From loic.minier at ubuntu.com Thu Sep 1 00:36:15 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Thu, 1 Sep 2016 02:36:15 +0200 Subject: New ubuntu-device-flash cannot use own gadget snap? In-Reply-To: References: <39865ede-2842-64f2-81a4-7c4b7397514f@parrot.com> <8d9a0ed6-b546-9a3c-c531-129c65d96d7f@canonical.com> Message-ID: Hi, I wanted to capture one of the issues that was encountered. ubuntu-device-flash snap can not work from an Ubuntu Core system; this is because it calls kpartx with the -s flag which relies on libdevmapper coordinating with its udev counterpart. Because dmsetup isn't installed in the core image, the udev signaling never comes back. To sum up: - Ubuntu classic + kpartx -avs: works - Ubuntu Core + kpartx -avs: hangs - Ubuntu Core + kpartx -av: works I wanted to mention this for people who try running ubuntu-device-flash from an Ubuntu Core system. (Various workarounds would be possible but because ubuntu-device-flash ought to be replaced by ubuntu-image soon it's probably not worth addressing right now.) Cheers, - Loïc Minier On Wed, Aug 31, 2016 at 8:27 PM, Loïc Minier wrote: > Yep; but the failure in kpartx didn't seem specific to local vs remote > snaps. > > For people on the list, I followed up on IRC and Yann was running into > weird issues with kpartx and ext4 in his environments, but from a clean > Ubuntu vm things worked and that's a viable solution for now. > > Cheers, > - Loïc > > On Wed, Aug 31, 2016 at 12:26 PM, Simon Fels > wrote: > >> On 31.08.2016 12:16, Loïc Minier wrote: >> > Works for me: >> > >> > $ sudo -E UBUNTU_DEVICE_FLASH_IGNORE_UNSTABLE_GADGET_DEFINITION=1 >> > /snap/bin/ubuntu-device-flash --verbose core 16 -o snappy.img --channel >> > edge --gadget pc --kernel pc-kernel --os ubuntu-core >> >> Yann is trying to use a gadget snap which isn't in the store but placed >> on the local disk. So I suspect he is running something like >> >> $ sudo -E UBUNTU_DEVICE_FLASH_IGNORE_UNSTABLE_GADGET_DEFINITION=1 >> /snap/bin/ubuntu-device-flash ... --gadget /home/user/my-gadget.snap ... >> >> regards, >> Simon >> >> > Determining gadget configuration >> > >> > 836.00 KB / 836.00 KB [==================================] 100.00% >> > 12.77 MB/s 0 >> > >> > Partitioning... >> > >> > Formatting... >> > >> > Mounting... >> > >> > Provisioning... >> > >> > 74.40 MB / 74.40 MB [===================================] 100.00% >> 24.02 >> > MB/s 3s >> > >> > 74.40 MB / 74.40 MB [===================================] 100.00% >> 59.46 >> > MB/s 1s >> > >> > 110.54 MB / 110.54 MB [=================================] 100.00% >> 23.19 >> > MB/s 4s >> > >> > 836.00 KB / 836.00 KB [==================================] 100.00% >> > 68.23 MB/s 0 >> > >> > Unmounting... >> > >> > New image complete >> > >> > Summary: >> > >> > Output: snappy.img >> > >> > Architecture: amd64 >> > >> > Channel: edge >> > >> > Version: 0 >> > >> > >> > On Wed, Aug 31, 2016 at 11:11 AM, Yann Sionneau >> > > wrote: >> > >> > >> > >> > Le 08/31/2016 à 10:56 AM, Simon Fels a écrit : >> > > On 31.08.2016 10:49, Yann Sionneau wrote: >> > >> Hello, >> > >> >> > >> It seems the new (snapp'ed) ubuntu-device-flash cannot use my >> own gadget >> > >> snap anymore. >> > >> >> > >> yann at imperium$ sudo -E /snap/bin/ubuntu-device-flash --verbose >> core 16 >> > >> -o snappy.img --channel edge --gadget >> $PWD/../../../xxx_2.0_all.snap >> > >> --kernel ../../../xxx_kernel/xxx-kernel_3.10.97_armhf.snap --os >> > >> ubuntu-core --developer-mode --enable-ssh >> > >> cannot use >> > >> "/home/yann/dev/snappy_xxx/tools/snappy/xxx_image/../../../ >> xxx_2.0_all.snap", >> > >> must be one of: ["canonical-i386" "canonical-pc" "pc" >> "canonical-pi2" >> > >> "pi2" "pi3" "canonical-dragon" "dragonboard" "beagleblack" >> "plano-amd64"] >> > >> >> > >> Is porting Snappy on non official devices not supported anymore? >> > > No, that is not the case. >> > > >> > >> How is it supposed to work now? I must confess that I am blocked >> in my >> > >> work because of this change, I cannot generate nor flash images >> anymore >> > >> and my project is thus stalled :/ >> > > We're currently in a phase where ubuntu-device-flash is still >> being used >> > > but the future will be a new tool called ubuntu-image which will >> allow >> > > you to create images in a much better way. >> > Yes I've heard of the new ubuntu-image tool. It's a good idea to >> make >> > this new tool! >> > It's just important I think that the old tools stay functional >> until the >> > new ones are ready. >> > > >> > > From what I got from Michael a lot things are currently hard coded >> > > inside ubuntu-device-flash. See [1] for the relevant code bits. >> > > >> > > You can override the sanity check for the gadget names with >> setting >> > > UBUNTU_DEVICE_FLASH_IGNORE_UNSTABLE_GADGET_DEFINITION=1 in the >> command >> > > line you're executing. Didn't tested this but maybe Michael can >> comment >> > > how this should work. >> > Ok, the environment variable works, thanks a lot! >> > Now I get this: >> > >> > error while executing external command mkfs.ext4 -F -L writable >> > /dev/mapper/loop3p2: mke2fs 1.42.13 (17-May-2015) >> > Invalid filesystem option set: >> > has_journal,extent,huge_file,flex_bg,metadata_csum,64bit,di >> r_nlink,extra_isize >> > >> > any idea? >> > >> > > >> > > regards, >> > > Simon >> > > >> > > [1]: >> > > >> > https://bazaar.launchpad.net/~mvo/goget-ubuntu-touch/minima >> l-first-boot-no-prepare-image/view/head:/ubuntu-device- >> flash/snappy.go#L116 >> > > nimal-first-boot-no-prepare-image/view/head:/ubuntu-device- >> flash/snappy.go#L116> >> > > >> > >> > >> > -- >> > Snapcraft mailing list >> > Snapcraft at lists.snapcraft.io >> > Modify settings or unsubscribe at: >> > https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > >> > >> > >> > >> > >> > -- >> > - Loïc >> > >> > >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> > > > > -- > - Loïc > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From yann.sionneau at parrot.com Thu Sep 1 09:30:02 2016 From: yann.sionneau at parrot.com (Yann Sionneau) Date: Thu, 1 Sep 2016 11:30:02 +0200 Subject: Snappy takes long time to boot when no network Message-ID: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> Hello, I've just updated my snappy image and now it takes a very long time to boot (140 seconds) because of [FAILED] Failed to start Wait for Network to be Configured. See 'systemctl status systemd-networkd-wait-online.service' for details. which is : https://www.freedesktop.org/software/systemd/man/systemd-networkd-wait-online.service.html Is there a way to configure Snappy to not wait for network to go up at bootup? Lots of embedded devices don't have a 24/24 network connection. Thanks! PS : I already tried systemctl disable systemctl status systemd-networkd-wait-online.service without success. Regards, -- Yann From mabnhdev at gmail.com Thu Sep 1 10:10:36 2016 From: mabnhdev at gmail.com (MikeB) Date: Thu, 1 Sep 2016 06:10:36 -0400 Subject: Stuck in Ubuntu Core Profile Setup Message-ID: I've been working on bringing up Ubuntu Core on some new hardware targets - network switch whiteboxes. Very recently, the first boot of Ubuntu Core has added a Profile Setup step. I'm stuck in this step and can't get out. My target is now a brick... My Xenial Desktop development system is up to date as of this morning. I built the kernel snap for the target using snapcraft 2.15.1. I then use ubuntu-device-flash (Version 7 Rev 7 from the edge channel) to build an image for the target using the following command. ubuntu-device-flash core 16 --channel=edge --kernel=$snap --gadget=pc --os=ubuntu-core -o $IMG I then create an ONIE NOS Installer from the resulting image and install it on my target switch. Here's where the fun begins... While the target is booting, it gets to the following and just sits there forever with no indication that it is waiting for manual intervention - requiring manual intervention during a boot is problematic for network switches in any case. =================================== ... [ 24.633353] cloud-init[2405]: Cloud-init v. 0.7.7 running 'modules:config' at Thu, 01 Sep 2016 09:43:05 +0000. Up 24.04 seconds. [ 24.652503] cloud-init[2405]: 2016-09-01 09:43:06,375 - util.py[WARNING]: Running module snappy (Sep 1 09:43:09 ec2: <14>Sep 1 09:43:09 ec2: ############################################################# <14>Sep 1 09:43:09 ec2: -----BEGIN SSH HOST KEY FINGERPRINTS----- <14>Sep 1 09:43:09 ec2: 1024 SHA256:TrfoEO2enrY3+HYlcxtKDV6MZjRSDvxwmdLaf07mEOI root at localhost.localdomain (DSA) <14>Sep 1 09:43:09 ec2: 256 SHA256:C0jrA04aKDV+z56ftvVNvd+NhUMf1zUncqpQPVSUnLg root at localhost.localdomain (ECDSA) <14>Sep 1 09:43:09 ec2: 256 SHA256:8t9b4izHmXc/RPKekQ6qP3yQM6icVOgrQgPLJ9d+XB8 root at localhost.localdomain (ED25519) <14>Sep 1 09:43:09 ec2: 2048 SHA256:a2eY2+sMacC5UUy3k3hYGbXXP4KSJVhqYLu6vAu7xIw root at localhost.localdomain (RSA) <14>Sep 1 09:43:09 ec2: -----END SSH HOST KEY FINGERPRINTS----- <14>Sep 1 09:43:09 ec2: ############################################################# -----BEGIN SSH HOST KEY KEYS----- ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBATz2hu5/KC3AMAbx5ngRj+pzcaqntJ2Vu+YG3fk21MCJy0RwyorZWpcfJmRYu9YsWExNNLqI3ibkG4pe06MeCo= roon ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAICcyf82/UPZsuSzegwFjpegDn4l0jTW8ydd1/gtlGdqh root at localhost.localdomain ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQC4XYVVMexJ4gISwT5/GpiAZ/p2aPKJ9mHNq7bDUhSnNqGd++j5n9LbX8AeZSWJrvfXNI4h9HpEoc2Fcisp3xxGO/J+FeZXbkA5TTRtzLhi3AbPDbRA7ptEU3/xMhwIn -----END SSH HOST KEY KEYS----- [ 27.730512] cloud-init[2461]: Cloud-init v. 0.7.7 running 'modules:final' at Thu, 01 Sep 2016 09:43:08 +0000. Up 27.21 seconds. [ 27.748699] cloud-init[2461]: ci-info: no authorized ssh keys fingerprints found for user ubuntu. [ 27.764506] cloud-init[2461]: Cloud-init v. 0.7.7 finished at Thu, 01 Sep 2016 09:43:09 +0000. Datasource DataSourceNoCloud [seed=/var/lib/cloud/seed/nocloud-nets =================================== At this point, I happened to hit on the console keyboard, and a new screen came up to "Configure the network and setup an administrator account on this all-snap Ubuntu Core system. After this setup process you will have secure web or command access to the system." I followed the instructions and took default values for the network setup since the switch gets everything from DHCP. All good so far as it sets up the network configuration. At this point, it asks me to "Enter an email address from your account in the store. ". I enter the e-mail address I use whenever I log into 'Ubuntu One'. At this point, I get one of two results. 1. I see a message saying 'Contacting the Store', then get the error "Creating user failed: error: bad user result: cannot create user for : no ssh keys found" -- OR -- 2. I immediately see the error "Creating user failed: error: bad user result: cannot create user "": Get https://login.ubuntu.com/api/v2/keys/: dial tcp: lookup login.ubuntu.com on [::1]:53: read udp [::1]:40286->[::1]:53: read: connection refused I'l usually see the first error on the first try and the second error on subsequent tries until I power-cycle the target switch. In either case, I can never get past this setup, so as I said above, my target switch is now a brick. Can someone tell me how to get past this screen so that the boot completes and I can get back to work? Thanks, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From mabnhdev at gmail.com Thu Sep 1 10:47:14 2016 From: mabnhdev at gmail.com (MikeB) Date: Thu, 1 Sep 2016 06:47:14 -0400 Subject: Stuck in Ubuntu Core Profile Setup Message-ID: On 01-Sep-2016, MikeB wrote: 1. I see a message saying 'Contacting the Store', then get the error > "Creating user failed: error: bad user result: cannot create user for > : no ssh keys found" -- OR -- 2. I immediately see the > error "Creating user failed: error: bad user > result: cannot create user "": Get > https://login.ubuntu.com/api/v2/keys/: dial tcp: lookup > login.ubuntu.com on [::1]:53: read udp [::1]:40286->[::1]:53: read: > connection refused I'l usually see the first error on the first try and > the second error on > subsequent tries until I power-cycle the target switch. In either case, I > can never get past this setup, so as I said above, my > target switch is now a brick. Can someone tell me how to get past this > screen so that the boot completes > and I can get back to work? > ​ ​I went to my Ubuntu One account and ​imported my Public SSH Key and I was able to successfully complete the 'Profile Setup'. I hadn't realized that my account even required a public key. So, I'm now pass the 'Profile Setup' and can use the switch again. However, I have some concerns about this new feature... 1. I'm concerned that the bootload just "froze" with no indication that it was looking for manual intervention. 2. I'm concerned that there was no obvious way to bypass this profile setup and get on with the boot. 3. I'm concerned that network equipment has to be registered to a particular user that has to have an Ubuntu One account with an imported public key. In my particular case, these switches are used by many developers. I don't want all the developers forced to create Ubuntu One accounts and I don't want to give all the developers my Ubuntu One credentials. 4. When the unit finally booted, I was unable to perform a 'sudo snap install hello-world'. I encountered what looked like network errors. I rebooted the switch and then was able to 'snap install'. It doesn't look like the first boot left the network configuration in a good state as the boot finished up. Regards, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Thu Sep 1 10:57:35 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Thu, 1 Sep 2016 12:57:35 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> Message-ID: Hi Yann, Following up from our IRC exchange :-) What would you expect to be working before network is up? Is it startup of the snaps' services, or the console, or the Snapweb web UI etc. [1]? Cheers, - Loïc [1] of course, still need to address Snapweb not starting properly with no network On Thu, Sep 1, 2016 at 11:30 AM, Yann Sionneau wrote: > Hello, > > I've just updated my snappy image and now it takes a very long time to > boot (140 seconds) because of > > [FAILED] Failed to start Wait for Network to be Configured. > See 'systemctl status systemd-networkd-wait-online.service' for details. > > which is : > https://www.freedesktop.org/software/systemd/man/systemd- > networkd-wait-online.service.html > > Is there a way to configure Snappy to not wait for network to go up at > bootup? > > Lots of embedded devices don't have a 24/24 network connection. > > Thanks! > > PS : I already tried systemctl disable systemctl status > systemd-networkd-wait-online.service without success. > > Regards, > > -- > > Yann > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From yann.sionneau at parrot.com Thu Sep 1 11:29:36 2016 From: yann.sionneau at parrot.com (Yann Sionneau) Date: Thu, 1 Sep 2016 13:29:36 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> Message-ID: <41df9529-73ee-e096-3966-92f35956fc25@parrot.com> Hi Loic, I expect the system to boot normally but everything needing internet would not work obviously. For instance indeed the snap services would be started, snapweb, the uart console. But indeed I would not expect the auto refresh of snaps to work. PS : and in fact I have 2 interfaces with internet access, and still the service "networkd-wait-online" stalls. Cheers, Yann Le 09/01/2016 à 12:57 PM, Loïc Minier a écrit : > Hi Yann, > > Following up from our IRC exchange :-) > > What would you expect to be working before network is up? Is it > startup of the snaps' services, or the console, or the Snapweb web UI > etc. [1]? > > Cheers, > - Loïc > > [1] of course, still need to address Snapweb not starting properly > with no network > > On Thu, Sep 1, 2016 at 11:30 AM, Yann Sionneau > > wrote: > > Hello, > > I've just updated my snappy image and now it takes a very long time to > boot (140 seconds) because of > > [FAILED] Failed to start Wait for Network to be Configured. > See 'systemctl status systemd-networkd-wait-online.service' for > details. > > which is : > https://www.freedesktop.org/software/systemd/man/systemd-networkd-wait-online.service.html > > > Is there a way to configure Snappy to not wait for network to go up at > bootup? > > Lots of embedded devices don't have a 24/24 network connection. > > Thanks! > > PS : I already tried systemctl disable systemctl status > systemd-networkd-wait-online.service without success. > > Regards, > > -- > > Yann > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > -- > - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Thu Sep 1 11:42:51 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 01 Sep 2016 13:42:51 +0200 Subject: Stuck in Ubuntu Core Profile Setup In-Reply-To: References: Message-ID: <1472730171.15344.13.camel@ubuntu.com> hi, On Do, 2016-09-01 at 06:10 -0400, MikeB wrote: > Very recently, the first boot of Ubuntu Core has added a Profile > Setup step.  I'm stuck in this step and can't get out.  My target is > now a brick... > i filed  https://bugs.launchpad.net/ubuntu/+source/subiquity/+bug/1619245 for this ... feel free to confirm ... ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From ogra at ubuntu.com Thu Sep 1 11:47:50 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 01 Sep 2016 13:47:50 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> Message-ID: <1472730470.15344.14.camel@ubuntu.com> hi, On Do, 2016-09-01 at 11:30 +0200, Yann Sionneau wrote: > > Lots of embedded devices don't have a 24/24 network connection. >  feel free to confirm: https://bugs.launchpad.net/ubuntu/+source/nplan/+bug/1619258 ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From jamie.bennett at canonical.com Thu Sep 1 11:47:58 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Thu, 1 Sep 2016 12:47:58 +0100 Subject: Stuck in Ubuntu Core Profile Setup In-Reply-To: <1472730171.15344.13.camel@ubuntu.com> References: <1472730171.15344.13.camel@ubuntu.com> Message-ID: <4F189796-F594-477A-BE5C-32880DFC68D3@canonical.com> > On 1 Sep 2016, at 12:42, Oliver Grawert wrote: > > hi, > On Do, 2016-09-01 at 06:10 -0400, MikeB wrote: > >> Very recently, the first boot of Ubuntu Core has added a Profile >> Setup step. I'm stuck in this step and can't get out. My target is >> now a brick... >> > i filed > https://bugs.launchpad.net/ubuntu/+source/subiquity/+bug/1619245 > > for this ... feel free to confirm … I believe we have two bugs here but the same solution may fix both. In #1619245 the user cannot be created because of a lack of network connectivity, in this case it is an incomplete profile in Ubuntu One (i.e. no public SSH key). Either way a fix would be to allow a local user to be created on first boot by console-conf but this local user should be limited to side-loaded snap and not have authority to install snaps from the store. Regards, Jamie. > ciao > oli-- From yann.sionneau at parrot.com Thu Sep 1 12:05:12 2016 From: yann.sionneau at parrot.com (Yann Sionneau) Date: Thu, 1 Sep 2016 14:05:12 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: <1472730470.15344.14.camel@ubuntu.com> References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> <1472730470.15344.14.camel@ubuntu.com> Message-ID: I feel like this is the bug I'm hitting. But, how do you explain that my boot is stalled even if I have 2 NICs with internet access? One is via wifi, the other is via usb-ethernet. Thanks! Le 09/01/2016 à 01:47 PM, Oliver Grawert a écrit : > hi, > On Do, 2016-09-01 at 11:30 +0200, Yann Sionneau wrote: >> Lots of embedded devices don't have a 24/24 network connection. >> > feel free to confirm: > > https://bugs.launchpad.net/ubuntu/+source/nplan/+bug/1619258 > > ciao > oli > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Thu Sep 1 12:30:28 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 01 Sep 2016 14:30:28 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> <1472730470.15344.14.camel@ubuntu.com> Message-ID: <1472733028.15344.29.camel@ubuntu.com> hi, On Do, 2016-09-01 at 14:05 +0200, Yann Sionneau wrote: > I feel like this is the bug I'm hitting. > But, how do you explain that my boot is stalled even if I have 2 NICs > with internet access? > One is via wifi, the other is via usb-ethernet. > Thanks! hmm, this sounds more like a different bug ... how about: https://launchpad.net/bugs/1618522 ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From yann.sionneau at parrot.com Thu Sep 1 12:44:19 2016 From: yann.sionneau at parrot.com (Yann Sionneau) Date: Thu, 1 Sep 2016 14:44:19 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: <1472733028.15344.29.camel@ubuntu.com> References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> <1472730470.15344.14.camel@ubuntu.com> <1472733028.15344.29.camel@ubuntu.com> Message-ID: I'm not sure I understand everything in the bug ticket as I am not a systemd / networkd / netplan expert at all. But : root at Paros:~# cat /etc/netplan/00- 00-initial-config.yaml 00-snapd-config.yaml root at Paros:~# cat /etc/netplan/00-initial-config.yaml network: version: 2 ethernets: all: match: name: "*" dhcp4: true root at Paros:~# cat /run/systemd/network/10-netplan-all.network [Match] Name=* [Network] DHCP=ipv4 root at Paros:~# networkctl IDX LINK TYPE OPERATIONAL SETUP 1 lo loopback carrier configured 2 sit0 sit routable configuring 3 eth1 ether routable configured 4 eth0 ether routable configured 5 rndis0 ether no-carrier configuring 5 links listed. Does this help? eth0 is a WiFi interface. eth1 is USB ethernet device (plugged at boot) Le 09/01/2016 à 02:30 PM, Oliver Grawert a écrit : > hi, > On Do, 2016-09-01 at 14:05 +0200, Yann Sionneau wrote: >> I feel like this is the bug I'm hitting. >> But, how do you explain that my boot is stalled even if I have 2 NICs >> with internet access? >> One is via wifi, the other is via usb-ethernet. >> Thanks! > hmm, this sounds more like a different bug ... > how about: > > https://launchpad.net/bugs/1618522 > > ciao > oli From mark at ubuntu.com Thu Sep 1 16:38:05 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 1 Sep 2016 12:38:05 -0400 Subject: Stuck in Ubuntu Core Profile Setup In-Reply-To: References: Message-ID: Hi Mike You are seeing the bootstrapping of a brand new all-snap OS, sorry for the rough patches but we have just added this first-boot experience. The rest is shaping up very nicely, that part is just brand new. We expect a usable image on Friday this week, and a release candidate later in the month. This new Ubuntu Core 16 image is shaping up to be something we can be very proud of: * everything is a snap and those snaps are super-fast and super-transactional * the bootstrap process is very factory-friendly for people making physical appliances * there is a very strong security foundation that raises the bar for all appliances and IoT This first-user experience addresses some long-standing issues that have bugged me about Ubuntu across cloud and devices for nearly a decade! * the 'ubuntu' default user is removed so appliance vendors have much more brand control * we eliminate default users from the bootstrap process * we work the same way from cloud to edge appliance Apologies again for the rough spots in the first code drop, but I bet you love the end result. Mark On 01/09/16 06:47, MikeB wrote: > On 01-Sep-2016, MikeB wrote: > > 1. I see a message saying 'Contacting the Store', then get the error > "Creating user failed: error: bad user result: cannot create user for > : no ssh keys found" -- OR -- 2. I immediately see > the error "Creating user failed: error: bad user > result: cannot create user "": Get > https://login.ubuntu.com/api/v2/keys/: dial tcp: lookup > login.ubuntu.com on [::1]:53: read udp > [::1]:40286->[::1]:53: read: > connection refused I'l usually see the first error on the first > try and the second error on > subsequent tries until I power-cycle the target switch. In either > case, I can never get past this setup, so as I said above, my > target switch is now a brick. Can someone tell me how to get past > this screen so that the boot completes > and I can get back to work? > ​ > > > ​I went to my Ubuntu One account and ​imported my Public SSH Key and I > was able to successfully complete the 'Profile Setup'. I hadn't > realized that my account even required a public key. So, I'm now pass > the 'Profile Setup' and can use the switch again. > > However, I have some concerns about this new feature... > > 1. I'm concerned that the bootload just "froze" with no indication > that it was looking for manual intervention. > 2. I'm concerned that there was no obvious way to bypass this profile > setup and get on with the boot. > 3. I'm concerned that network equipment has to be registered to a > particular user that has to have an Ubuntu One account with an > imported public key. In my particular case, these switches are used > by many developers. I don't want all the developers forced to create > Ubuntu One accounts and I don't want to give all the developers my > Ubuntu One credentials. > 4. When the unit finally booted, I was unable to perform a 'sudo snap > install hello-world'. I encountered what looked like network errors. > I rebooted the switch and then was able to 'snap install'. It doesn't > look like the first boot left the network configuration in a good > state as the boot finished up. > > Regards, Mike > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mabnhdev at gmail.com Thu Sep 1 16:40:41 2016 From: mabnhdev at gmail.com (MikeB) Date: Thu, 1 Sep 2016 12:40:41 -0400 Subject: Stuck in Ubuntu Core Profile Setup In-Reply-To: References: Message-ID: Understood about the rough spots - it's the cost of playing on the edge. This one was a bit more disruptive than usual. I'm looking forward to the official release. Cheers, Mike On Thu, Sep 1, 2016 at 12:38 PM, Mark Shuttleworth wrote: > Hi Mike > > You are seeing the bootstrapping of a brand new all-snap OS, sorry for the > rough patches but we have just added this first-boot experience. The rest > is shaping up very nicely, that part is just brand new. We expect a usable > image on Friday this week, and a release candidate later in the month. > > This new Ubuntu Core 16 image is shaping up to be something we can be very > proud of: > > * everything is a snap and those snaps are super-fast and > super-transactional > * the bootstrap process is very factory-friendly for people making > physical appliances > * there is a very strong security foundation that raises the bar for all > appliances and IoT > > This first-user experience addresses some long-standing issues that have > bugged me about Ubuntu across cloud and devices for nearly a decade! > > * the 'ubuntu' default user is removed so appliance vendors have much > more brand control > * we eliminate default users from the bootstrap process > * we work the same way from cloud to edge appliance > > Apologies again for the rough spots in the first code drop, but I bet you > love the end result. > > Mark > > > On 01/09/16 06:47, MikeB wrote: > > On 01-Sep-2016, MikeB wrote: > > 1. I see a message saying 'Contacting the Store', then get the error >> "Creating user failed: error: bad user result: cannot create user for >> : no ssh keys found" -- OR -- 2. I immediately see the >> error "Creating user failed: error: bad user >> result: cannot create user "": Get >> https://login.ubuntu.com/api/v2/keys/: dial tcp: lookup >> login.ubuntu.com on [::1]:53: read udp [::1]:40286->[::1]:53: read: >> connection refused I'l usually see the first error on the first try and >> the second error on >> subsequent tries until I power-cycle the target switch. In either case, I >> can never get past this setup, so as I said above, my >> target switch is now a brick. Can someone tell me how to get past this >> screen so that the boot completes >> and I can get back to work? >> ​ > > > ​I went to my Ubuntu One account and ​imported my Public SSH Key and I was > able to successfully complete the 'Profile Setup'. I hadn't realized that > my account even required a public key. So, I'm now pass the 'Profile > Setup' and can use the switch again. > > However, I have some concerns about this new feature... > > 1. I'm concerned that the bootload just "froze" with no indication that it > was looking for manual intervention. > 2. I'm concerned that there was no obvious way to bypass this profile > setup and get on with the boot. > 3. I'm concerned that network equipment has to be registered to a > particular user that has to have an Ubuntu One account with an imported > public key. In my particular case, these switches are used by many > developers. I don't want all the developers forced to create Ubuntu One > accounts and I don't want to give all the developers my Ubuntu One > credentials. > 4. When the unit finally booted, I was unable to perform a 'sudo snap > install hello-world'. I encountered what looked like network errors. I > rebooted the switch and then was able to 'snap install'. It doesn't look > like the first boot left the network configuration in a good state as the > boot finished up. > > Regards, Mike > > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From casey.marshall at canonical.com Thu Sep 1 17:05:32 2016 From: casey.marshall at canonical.com (Casey Marshall) Date: Thu, 1 Sep 2016 12:05:32 -0500 Subject: Setting release version in snapcraft build Message-ID: When automating the build process for snaps, I'd like to be able to provide the release version as an argument to snapcraft, which could then be used as a variable in the snapcraft.yaml. For example, say I'd like to release "foo" version "1.2.3". I'd then like to build the snap with version: 1.2.3, and use the git tag "v1.2.3" in its source. For example: name: foo version: ${release-version} ... parts: foo: source: git at ... source-tag: v${release-version} The alternatives at the moment seem to be: 1. Parse the version out of snapcraft.yaml, using that as the authoritative release version string. Workable for new projects, but a tough sell for existing projects with an established release process. 2. Generate the snapcraft.yaml from a template (jinja, for example), which works but is kind of awkward. So, how about a ${release-version} variable in snapcraft.yaml, which you could set on the snapcraft command-line with a --release-version flag? This would be darn useful for Jenkins CI scripts that build snaps. Does this sound like a reasonable feature to add to snapcraft? Or is there already a preferred method for managing the release version string? -Casey -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Thu Sep 1 17:15:41 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Thu, 1 Sep 2016 14:15:41 -0300 Subject: Setting release version in snapcraft build In-Reply-To: References: Message-ID: El 01/09/16 a las 14:05, Casey Marshall escribió: > When automating the build process for snaps, I'd like to be able to > provide the release version as an argument to snapcraft, which could > then be used as a variable in the snapcraft.yaml. > > For example, say I'd like to release "foo" version "1.2.3". I'd then > like to build the snap with version: 1.2.3, and use the git tag > "v1.2.3" in its source. For example: > > name: foo > version: ${release-version} > ... > parts: > foo: > source: git at ... > source-tag: v${release-version} In 2.16 we are introducing a variable to set this one, it comes from the `version` defined above. > > The alternatives at the moment seem to be: > > 1. Parse the version out of snapcraft.yaml, using that as the > authoritative release version string. Workable for new projects, but a > tough sell for existing projects with an established release process. > > 2. Generate the snapcraft.yaml from a template (jinja, for example), > which works but is kind of awkward. > > So, how about a ${release-version} variable in snapcraft.yaml, which > you could set on the snapcraft command-line with a --release-version > flag? This would be darn useful for Jenkins CI scripts that build snaps. What would the version be set to during CI? The version from a snappy point of view is just a friendly name, the architecture only cares about revisions. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Thu Sep 1 17:40:43 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Thu, 1 Sep 2016 14:40:43 -0300 Subject: Setting release version in snapcraft build In-Reply-To: References: Message-ID: Readding the list El 01/09/16 a las 14:27, Casey Marshall escribió: > On Thu, Sep 1, 2016 at 12:15 PM, Sergio Schvezov > > > wrote: > > El 01/09/16 a las 14:05, Casey Marshall escribió: >> When automating the build process for snaps, I'd like to be able >> to provide the release version as an argument to snapcraft, which >> could then be used as a variable in the snapcraft.yaml. >> >> For example, say I'd like to release "foo" version "1.2.3". I'd >> then like to build the snap with version: 1.2.3, and use the git >> tag "v1.2.3" in its source. For example: >> >> name: foo >> version: ${release-version} >> ... >> parts: >> foo: >> source: git at ... >> source-tag: v${release-version} > > In 2.16 we are introducing a variable to set this one, it comes > from the `version` defined above. > > > That certainly helps, thanks for this! > > >> >> The alternatives at the moment seem to be: >> >> 1. Parse the version out of snapcraft.yaml, using that as the >> authoritative release version string. Workable for new projects, >> but a tough sell for existing projects with an established >> release process. >> >> 2. Generate the snapcraft.yaml from a template (jinja, for >> example), which works but is kind of awkward. >> >> So, how about a ${release-version} variable in snapcraft.yaml, >> which you could set on the snapcraft command-line with a >> --release-version flag? This would be darn useful for Jenkins CI >> scripts that build snaps. > > What would the version be set to during CI? The version from a > snappy point of view is just a friendly name, the architecture > only cares about revisions. > > > Well, I'm thinking of the case where CI knows the release version up > front, and wants to set that on the snap it's building. > > For example, let's say I've added a snapcraft.yaml to my source tree. > I manage my releases by tagging them with a version tag, and then I > run a Jenkins job with the release version as a parameter. > > This release version is used to check out the source tree at that > release version tag, build, run tests, and then build the snap. > Jenkins knows the release version it's building, and should be able to > set the version in the snap. If I can't, I either have to try to the > snapcraft.yaml contents in sync with my release tags, or modify the > snapcraft.yaml contents at build time. > > So instead, I'd want my Jenkins job to be able to set the version on > the command line, something like `snapcraft --version > ${RELEASE_VERSION}`. This avoids fragile repetition of release > versions (among the tag & snap), or the need for > generating/substituting snapcraft.yaml contents in the job. I think it would be good to behave like dch does and have a `snapcraft set-version ` command instead. I let this sit for a bit and see what others think. In this scenario you'd do snapcraft set-version snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From leo.arias at canonical.com Thu Sep 1 17:46:22 2016 From: leo.arias at canonical.com (Leo Arias) Date: Thu, 1 Sep 2016 11:46:22 -0600 Subject: Setting release version in snapcraft build In-Reply-To: References: Message-ID: <57C8696E.70905@canonical.com> This is a pain point for me too. Most upstreams store the version number in only one place, because editing more than one for every tagged release is too boring. When we add the snapcraft.yaml, I suspect that the version field will be forgotten often, and it will not be in sync. However, there are a few issues with this. First, launchpad would need a new text field to fill the version. And then what would you write in that field? I see problems if you write a static string and if you write a command to be appended to the snapcraft run. Also, it's not always straight-forward to get the version from the code. Some projects store it in multiple statements, like major=2; minor=1; patch=3; Transforming that into "v2.1.3" results in a script that's not so nice. Now, I can see it working at least for the snaps I'm building using travis and docker. I can get the sha1 when building from master, or the name of tag when there is one, and use that as the version. I would love that. We could make that snapcraft --snap-version=$(git rev-parse HEAD) just overwrites the version field, useful immediately for people not using launchpad. pura vida -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From leo.arias at canonical.com Thu Sep 1 17:52:31 2016 From: leo.arias at canonical.com (Leo Arias) Date: Thu, 1 Sep 2016 11:52:31 -0600 Subject: Setting release version in snapcraft build In-Reply-To: References: Message-ID: <57C86ADF.1090900@canonical.com> On 2016-09-01 11:40, Sergio Schvezov wrote: > snapcraft set-version > snapcraft works for me. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From joseph.wakeling at webdrake.net Thu Sep 1 21:06:53 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Thu, 1 Sep 2016 23:06:53 +0200 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <7cff62a8-b4ec-34ab-b674-69b186d340d0@webdrake.net> <5620d7bf-458b-330f-92cb-83e48f0a2bad@ubuntu.com> <622ce8d3-75f9-f1f9-0797-21e7af47beb7@webdrake.net> <30004526-4ac8-996c-5d25-31373469252b@webdrake.net> Message-ID: On 31.08.2016 21:35, Joseph Rushton Wakeling wrote: > First things first, run `snapcraft stage`. Then edit the auto-generated > stage/etc/ldc2.conf and replace it with this: > https://gist.github.com/WebDrake/229645efeca14fa54b0b1c82bcbb6477 > > ... which as you can see includes a compiler flag: `-gcc=ldc2.gcc`. This should > ensure that ldc2.gcc is called when LDC wants to call GCC. OK, I think I have tied down why this is failing. Under the hood, LDC does a lookup of the full path to the executable whose name is passed via the -gcc flag, using llvm::sys::findProgramByName to do so. If it doesn't get a result, it first tries to look up CC, and if that fails, it reverts to looking up the path for gcc. This would suggest that, running within the snap, LDC is not able to find the path to ldc2.gcc. I've also tried passing it -gcc=%%ldcbinarypath%%/../command-gcc.wrapper and also -gcc=/snap/ldc2/x1/command-gcc.wrapper and in both cases it still fails. The weird thing is that the latter works when it's passed manually: ldc2.ldmd2 -gcc=/snap/ldc2/x1/command-gcc.wrapper hello.d ... but not when it's included in the etc/ldc2.conf file. This would suggest that something about the snap container is preventing LDC from resolving the path correctly. Can anyone more experienced with snappy containment advise what could be going on here? Thanks & best wishes, -- Joe From sergio.schvezov at canonical.com Thu Sep 1 21:18:46 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Thu, 1 Sep 2016 18:18:46 -0300 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <7cff62a8-b4ec-34ab-b674-69b186d340d0@webdrake.net> <5620d7bf-458b-330f-92cb-83e48f0a2bad@ubuntu.com> <622ce8d3-75f9-f1f9-0797-21e7af47beb7@webdrake.net> <30004526-4ac8-996c-5d25-31373469252b@webdrake.net> Message-ID: El 01/09/16 a las 18:06, Joseph Rushton Wakeling escribió: > On 31.08.2016 21:35, Joseph Rushton Wakeling wrote: >> First things first, run `snapcraft stage`. Then edit the auto-generated >> stage/etc/ldc2.conf and replace it with this: >> https://gist.github.com/WebDrake/229645efeca14fa54b0b1c82bcbb6477 >> >> ... which as you can see includes a compiler flag: `-gcc=ldc2.gcc`. >> This should >> ensure that ldc2.gcc is called when LDC wants to call GCC. > > OK, I think I have tied down why this is failing. > > Under the hood, LDC does a lookup of the full path to the executable > whose name is passed via the -gcc flag, using > llvm::sys::findProgramByName to do so. If it doesn't get a result, it > first tries to look up CC, and if that fails, it reverts to looking up > the path for gcc. > > This would suggest that, running within the snap, LDC is not able to > find the path to ldc2.gcc. > > I've also tried passing it > -gcc=%%ldcbinarypath%%/../command-gcc.wrapper and also > -gcc=/snap/ldc2/x1/command-gcc.wrapper and in both cases it still > fails. The weird thing is that the latter works when it's passed > manually: > > ldc2.ldmd2 -gcc=/snap/ldc2/x1/command-gcc.wrapper hello.d Please don't use the wrapper :-) > > ... but not when it's included in the etc/ldc2.conf file. > > This would suggest that something about the snap container is > preventing LDC from resolving the path correctly. Can anyone more > experienced with snappy containment advise what could be going on here? > Any reason why you don't just set it to gcc? If gcc is inside the snap, it should be in the PATH. To experiment simply run: snap run --shell ldc2ldmd2 If run from a directory not available while in confinement do a change dir to $SNAP (or $SNAP_DATA, $SNAP_USER_DATA, ...): cd $SNAP Verify the environment: which gcc # anything else From michael.hudson at canonical.com Thu Sep 1 21:19:49 2016 From: michael.hudson at canonical.com (Michael Hudson-Doyle) Date: Fri, 2 Sep 2016 09:19:49 +1200 Subject: Snappy takes long time to boot when no network In-Reply-To: References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> <1472730470.15344.14.camel@ubuntu.com> <1472733028.15344.29.camel@ubuntu.com> Message-ID: On 2 September 2016 at 00:44, Yann Sionneau wrote: > I'm not sure I understand everything in the bug ticket as I am not a > systemd / networkd / netplan expert at all. > > But : > > root at Paros:~# cat /etc/netplan/00- > 00-initial-config.yaml 00-snapd-config.yaml > root at Paros:~# cat /etc/netplan/00-initial-config.yaml > > network: > version: 2 > ethernets: > all: > match: > name: "*" > dhcp4: true > root at Paros:~# cat /run/systemd/network/10-netplan-all.network > [Match] > Name=* > > [Network] > DHCP=ipv4 > root at Paros:~# networkctl > IDX LINK TYPE OPERATIONAL SETUP > 1 lo loopback carrier configured > 2 sit0 sit routable configuring > 3 eth1 ether routable configured > 4 eth0 ether routable configured > 5 rndis0 ether no-carrier configuring > > 5 links listed. > > Does this help? > > eth0 is a WiFi interface. > > eth1 is USB ethernet device (plugged at boot) > > I don't know if it's the cause of all your issues, but having both files in /etc/netplan is a sign of console-conf / snapd version skew. It's fixed now and new installs won't have this problem, but if you don't want to re-install just deleted the 00-initial-config.yaml file and run sudo netplan apply. Cheers, mwh > Le 09/01/2016 à 02:30 PM, Oliver Grawert a écrit : > > hi, > > On Do, 2016-09-01 at 14:05 +0200, Yann Sionneau wrote: > >> I feel like this is the bug I'm hitting. > >> But, how do you explain that my boot is stalled even if I have 2 NICs > >> with internet access? > >> One is via wifi, the other is via usb-ethernet. > >> Thanks! > > hmm, this sounds more like a different bug ... > > how about: > > > > https://launchpad.net/bugs/1618522 > > > > ciao > > oli > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.wakeling at webdrake.net Thu Sep 1 21:35:00 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Thu, 1 Sep 2016 23:35:00 +0200 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <7cff62a8-b4ec-34ab-b674-69b186d340d0@webdrake.net> <5620d7bf-458b-330f-92cb-83e48f0a2bad@ubuntu.com> <622ce8d3-75f9-f1f9-0797-21e7af47beb7@webdrake.net> <30004526-4ac8-996c-5d25-31373469252b@webdrake.net> Message-ID: <775921a9-a627-5fcd-6cce-cd7be53b37fb@webdrake.net> On 01.09.2016 23:18, Sergio Schvezov wrote: > > > El 01/09/16 a las 18:06, Joseph Rushton Wakeling escribió: >> On 31.08.2016 21:35, Joseph Rushton Wakeling wrote: >>> First things first, run `snapcraft stage`. Then edit the auto-generated >>> stage/etc/ldc2.conf and replace it with this: >>> https://gist.github.com/WebDrake/229645efeca14fa54b0b1c82bcbb6477 >>> >>> ... which as you can see includes a compiler flag: `-gcc=ldc2.gcc`. This should >>> ensure that ldc2.gcc is called when LDC wants to call GCC. >> >> OK, I think I have tied down why this is failing. >> >> Under the hood, LDC does a lookup of the full path to the executable whose >> name is passed via the -gcc flag, using llvm::sys::findProgramByName to do >> so. If it doesn't get a result, it first tries to look up CC, and if that >> fails, it reverts to looking up the path for gcc. >> >> This would suggest that, running within the snap, LDC is not able to find the >> path to ldc2.gcc. >> >> I've also tried passing it -gcc=%%ldcbinarypath%%/../command-gcc.wrapper and >> also -gcc=/snap/ldc2/x1/command-gcc.wrapper and in both cases it still fails. Actually, turns out I was completely wrong in the comments above: putting -gcc=%%ldcbinarypath%%/../command-gcc.wrapper ... _does_ work; it was failing because I'd manually included an extra -gcc= flag in the snapcraft.yaml for the ldc2 and ldmd2 commands. With that fixed, all is good. But that brings us to ... > Please don't use the wrapper :-) Well, I didn't _want_ to; it was something I tried as an experiment to see what would happen, and it's the first setup that has worked. BTW what is problematic about using that wrapper in this way? Just to make sure I understand. > Any reason why you don't just set it to gcc? If gcc is inside the snap, it > should be in the PATH. To experiment simply run: Yea, the problem is not gcc being in the path, but the --sysroot of gcc being correct. With the default gcc being called, the LDC build process falls over something like this (LDC's verbose output of its gcc call): /snap/ldc2/x1/usr/bin/gcc hello.o -o hello -L/snap/ldc2/x1/bin/../lib -lphobos2-ldc -ldruntime-ldc -Wl,--gc-sections -lrt -ldl -lpthread -lm -m64 /snap/ldc2/x1/usr/bin/ld: cannot find /usr/lib/x86_64-linux-gnu/libpthread_nonshared.a collect2: error: ld returned 1 exit status Error: /snap/ldc2/x1/usr/bin/gcc failed with status: 1 So, gcc needs to be wrapped with something that adds the --sysroot=$SNAP option. I'll try now to create such a minimal wrapper. From joseph.wakeling at webdrake.net Thu Sep 1 21:52:24 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Thu, 1 Sep 2016 23:52:24 +0200 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> Message-ID: <0d2b3901-eb19-b286-c223-c7a47f6e3431@webdrake.net> On 27.08.2016 22:45, Joseph Rushton Wakeling wrote: > I thought I'd have a go at making a snap of LDC, the LLVM-based compiler for the > D programming language. I recognize that snapping a compiler might be jumping > ahead of the current intended use-case(s), but it's fun to see what could be > possible -- and besides, D's compilers are updated fairly often, so it would be > great to be able to package them easily in a truly cross-distro form. OK, a certain amount of wailing and gnashing of teeth later (and following some great help here and from the LDC devs -- big thanks to all!), I have a provisional first working setup (as in, at least, "Works For Me"). First, here's my snapcraft.yaml: https://gist.github.com/WebDrake/6ad7e5fe48abc999460cb67a31972afe So, put this in a working directory, and run `snapcraft stage`. Now edit stage/etc/ldc2.conf and replace it with this: https://gist.github.com/WebDrake/65cb3aaa355ea325af98d58f8ca52e3f ... and run `snapcraft prime`. Now create a file `prime/bin/gcc.wrapper` with the following contents: https://gist.github.com/WebDrake/0081628d70b61e271006c9b8fa3454ef (This is basically the command-gcc.wrapper that would be generated if I were exposing the gcc command outside the snap, but minus the path-related stuff that I presume is the objectionable part of using that wrapper.) Now run `snapcraft snap` to finish things up, install, and ... voilà, a working LDC snap. So, with something provisionally working, questions: * anything obvious that I could do better in the general setup? * is there any way to automate some of the stages of the above, e.g. the editing of stage/etc/ldc2.conf or the generation of prime/bin/gcc.wrapper ... ? For the second point, obviously I could create a makefile or other script to run things, but I wondered if there was any support in snapcraft itself for running simple post-stage or post-prime hooks ... ? Thanks again and best wishes, -- Joe From gustavo.niemeyer at canonical.com Thu Sep 1 22:15:25 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 1 Sep 2016 19:15:25 -0300 Subject: Let's kill "sideloading" Message-ID: Hello all, With assertions finally being put to great use, it's time to kill the term "sideloading". That term does a disservice to our conversations, because it is vague and also limits the thinking around what is possible. Whenever we use "sideloading", we mean one of two things: 1. The installation of a snap from the local filesystem 2. The installation of a snap that is not backed by assertions We can talk about these cases using this actual terminology. To talk about the second case tersely we can use "unasserted", which is apparently a real term [1]: "1. resting on a statement or claim unsupported by evidence or proof; alleged:" That's exactly what we mean by that. With assertions, we can have the first case without the second, though. A snap in the local filesystem doesn't necessarily have to be unasserted. So: Case 1: sideload => local snap Case 2: sideload => unasserted snap How does that sound? [1] http://www.dictionary.com/browse/unasserted gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From kirkland at canonical.com Thu Sep 1 22:28:07 2016 From: kirkland at canonical.com (Dustin Kirkland) Date: Thu, 1 Sep 2016 17:28:07 -0500 Subject: Let's kill "sideloading" In-Reply-To: References: Message-ID: On Thu, Sep 1, 2016 at 5:15 PM, Gustavo Niemeyer wrote: > Hello all, > > With assertions finally being put to great use, it's time to kill the term > "sideloading". That term does a disservice to our conversations, because it > is vague and also limits the thinking around what is possible. > > Whenever we use "sideloading", we mean one of two things: > > 1. The installation of a snap from the local filesystem > 2. The installation of a snap that is not backed by assertions > > We can talk about these cases using this actual terminology. To talk about > the second case tersely we can use "unasserted", which is apparently a real > term [1]: > > "1. resting on a statement or claim unsupported by evidence or proof; > alleged:" > > That's exactly what we mean by that. > > With assertions, we can have the first case without the second, though. A > snap in the local filesystem doesn't necessarily have to be unasserted. > > So: > > Case 1: sideload => local snap > Case 2: sideload => unasserted snap > > How does that sound? It's a reasonable suggestion, but you'll need to match the part of speech. "sideload" here is a verb, whereas "local snap" and "unasserted snap" are adjective + noun tuples. (And "sideloading" is a gerund, or a present participle verb form.) I think you're suggesting, s/sideload/install local/ and s/sideload/install unasserted/. So, colloquially: "Hey Jim, yeah, to do that you'll need to just sideload your snap" becomes "Hey Jim, yeah, to do that, you'll need to just install a local snap" or "Hey Jim, yeah, to do that, you'll need to install an unasserted snap." (While I find "unasserted" a little bumpy, but I'm sure it may feel a little more comfortable with practice.) From joseph.wakeling at webdrake.net Thu Sep 1 22:33:30 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Fri, 2 Sep 2016 00:33:30 +0200 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: <0d2b3901-eb19-b286-c223-c7a47f6e3431@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0d2b3901-eb19-b286-c223-c7a47f6e3431@webdrake.net> Message-ID: <140e596e-d55a-88db-0a46-36635e3a22e6@webdrake.net> On 01.09.2016 23:52, Joseph Rushton Wakeling wrote: > Now create a file `prime/bin/gcc.wrapper` with the following contents: > https://gist.github.com/WebDrake/0081628d70b61e271006c9b8fa3454ef > > (This is basically the command-gcc.wrapper that would be generated if I were > exposing the gcc command outside the snap, but minus the path-related stuff that > I presume is the objectionable part of using that wrapper.) Even easier solution: remove the `-gcc=` flag from the ldc2.conf, and just name the wrapper in prime/bin as `gcc`. I also note that libgcc-5-dev doesn't need to be in the `parts:` section of snapcraft.yaml. From seth.arnold at canonical.com Thu Sep 1 23:36:50 2016 From: seth.arnold at canonical.com (Seth Arnold) Date: Thu, 1 Sep 2016 16:36:50 -0700 Subject: Let's kill "sideloading" In-Reply-To: References: Message-ID: <20160901233650.GC11564@hunt> On Thu, Sep 01, 2016 at 07:15:25PM -0300, Gustavo Niemeyer wrote: > With assertions finally being put to great use, it's time to kill the term > "sideloading". That term does a disservice to our conversations, because it What are assertions? Thanks -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 473 bytes Desc: not available URL: From leo.arias at canonical.com Fri Sep 2 00:23:19 2016 From: leo.arias at canonical.com (Leo Arias) Date: Thu, 1 Sep 2016 18:23:19 -0600 Subject: Publishing to the snap store from Travis In-Reply-To: References: Message-ID: <57C8C677.1060506@canonical.com> Thanks Evan, this is great and I've already copied a couple of times. In a similar fashion, take a look at the script Kyle wrote to push the branch to launchpad: https://github.com/nextcloud/nextcloud-snap/blob/master/.travis.yml The benefit here is that launchpad will build and publish for many architectures. The downside is that you have to configure the project in launchpad: create a team, create a bot user, make a passwordless ssh key for it, create the project, push the branch once, set up the snap build. This will be a lot nicer with imports from github: https://bugs.launchpad.net/launchpad/+bug/1469459 pura vida. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From gustavo.niemeyer at canonical.com Fri Sep 2 00:36:23 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 1 Sep 2016 21:36:23 -0300 Subject: Let's kill "sideloading" In-Reply-To: <20160901233650.GC11564@hunt> References: <20160901233650.GC11564@hunt> Message-ID: Hi Seth, On Thu, Sep 1, 2016 at 8:36 PM, Seth Arnold wrote:t > On Thu, Sep 01, 2016 at 07:15:25PM -0300, Gustavo Niemeyer wrote: > > With assertions finally being put to great use, it's time to kill the > term > > "sideloading". That term does a disservice to our conversations, because > it > > What are assertions? An assertion is just a signed document that we can hand a snappy system. So the point here is that installing a local snap and having that signed information are orthogonal to each other. Historically one implied the other, and we had that term to refer to both of them at once. Nowadays, speaking of a "local installation" and the assertions/signed document clearly is much more clear. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From bret.barker at canonical.com Fri Sep 2 01:03:57 2016 From: bret.barker at canonical.com (Bret A. Barker) Date: Thu, 1 Sep 2016 21:03:57 -0400 Subject: Let's kill "sideloading" In-Reply-To: References: Message-ID: <20160902010356.GL34264@abitrandom.net> Consider also that we likely want to support remote unasserted installs for alternate stores [1], so they are fully orthogonal concepts. So we have all four local/remote asserted/unasserted combos. And I agree that "sideloading" is no longer a useful term. -bret [1] With whatever warnings or override flags/configs we feel appropriate for that use-case. On Thu, Sep 01, 2016 at 07:15:25PM -0300, Gustavo Niemeyer wrote: > Hello all, > > With assertions finally being put to great use, it's time to kill the term > "sideloading". That term does a disservice to our conversations, because it > is vague and also limits the thinking around what is possible. > > Whenever we use "sideloading", we mean one of two things: > > 1. The installation of a snap from the local filesystem > 2. The installation of a snap that is not backed by assertions > > We can talk about these cases using this actual terminology. To talk about > the second case tersely we can use "unasserted", which is apparently a real > term [1]: > > "1. resting on a statement or claim unsupported by evidence or proof; > alleged:" > > That's exactly what we mean by that. > > With assertions, we can have the first case without the second, though. A > snap in the local filesystem doesn't necessarily have to be unasserted. > > So: > > Case 1: sideload => local snap > Case 2: sideload => unasserted snap > > How does that sound? > > > [1] http://www.dictionary.com/browse/unasserted > > > gustavo @ http://niemeyer.net From mark at ubuntu.com Fri Sep 2 01:25:07 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 1 Sep 2016 21:25:07 -0400 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: <0d2b3901-eb19-b286-c223-c7a47f6e3431@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0d2b3901-eb19-b286-c223-c7a47f6e3431@webdrake.net> Message-ID: <464a4cbd-d46b-0a76-6bcd-4697b3f26c22@ubuntu.com> On 01/09/16 17:52, Joseph Rushton Wakeling wrote: > * is there any way to automate some of the stages of the above, > e.g. the editing of stage/etc/ldc2.conf or the generation of > prime/bin/gcc.wrapper ... ? I think we need a recommended way to: * carry patches for a part in the snap definition * provide wrapper scripts etc (could just be in a patch) * trigger hooks at each step in the snapcraft lifecycle I know Seb and others asked for a patch plan early on, which I resisted, but having made a few snaps I understand better why its useful as an impedance buffer before things can land upstream. Mark From kyle.fazzari at canonical.com Fri Sep 2 02:29:59 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Thu, 1 Sep 2016 19:29:59 -0700 Subject: Publishing to the snap store from Travis In-Reply-To: References: <57C8C677.1060506@canonical.com> Message-ID: On Sep 1, 2016 5:24 PM, "Leo Arias" wrote: > > Thanks Evan, this is great and I've already copied a couple of times. > > In a similar fashion, take a look at the script Kyle wrote to push the > branch to launchpad: > https://github.com/nextcloud/nextcloud-snap/blob/master/.travis.yml Note that if I were less lazy^Wefficient, I would have used the Launchpad python API for this. That would also allow for polling and noting build failures, etc. Kyle -------------- next part -------------- An HTML attachment was scrubbed... URL: From andrew.wilkins at canonical.com Fri Sep 2 03:42:12 2016 From: andrew.wilkins at canonical.com (Andrew Wilkins) Date: Fri, 02 Sep 2016 03:42:12 +0000 Subject: Snapping LDC (LLVM-based D compiler) In-Reply-To: <140e596e-d55a-88db-0a46-36635e3a22e6@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0d2b3901-eb19-b286-c223-c7a47f6e3431@webdrake.net> <140e596e-d55a-88db-0a46-36635e3a22e6@webdrake.net> Message-ID: On Fri, Sep 2, 2016 at 6:34 AM Joseph Rushton Wakeling < joseph.wakeling at webdrake.net> wrote: > On 01.09.2016 23:52, Joseph Rushton Wakeling wrote: > > Now create a file `prime/bin/gcc.wrapper` with the following contents: > > https://gist.github.com/WebDrake/0081628d70b61e271006c9b8fa3454ef > > > > (This is basically the command-gcc.wrapper that would be generated if I > were > > exposing the gcc command outside the snap, but minus the path-related > stuff that > > I presume is the objectionable part of using that wrapper.) > > Even easier solution: remove the `-gcc=` flag from the ldc2.conf, and just > name > the wrapper in prime/bin as `gcc`. > What I'm going to do with llgo is: - use "organize" in the gcc part, renaming usr/bin/gcc to usr/bin/gcc.real - add another part using the "dump" plugin, with a simple #!/bin/sh script called usr/bin/gcc that invokes gcc.real with --sysroot=$SNAP You can do something similar for ldc2.conf: omit the default one from your ldc2 part's installed files (add "-etc/ldc2.conf" to the "snap"), and provide another part that has the ldc2.conf you want. I've forked your gist here: https://gist.github.com/axw/371db15c5139c9dc64bf920f30118e19 If you create gcc-wrapper and ldc2-conf directories in the same directory as snapcraft.yaml, and add the gcc-wrapper and ldc2.conf files to those, respectively, it should build cleanly with a single "snapcraft". At least it did for me - and "snap run ldc2" was able to compile something from my home directory. Cheers, Andrew I also note that libgcc-5-dev doesn't need to be in the `parts:` section of > snapcraft.yaml. > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Fri Sep 2 07:07:49 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 02 Sep 2016 09:07:49 +0200 Subject: Let's kill "sideloading" In-Reply-To: References: Message-ID: <1472800069.5902.11.camel@ubuntu.com> hi, Am Donnerstag, den 01.09.2016, 19:15 -0300 schrieb Gustavo Niemeyer: > Hello all, > > With assertions finally being put to great use, it's time to kill the > term "sideloading". That term does a disservice to our conversations, > because it is vague and also limits the thinking around what is > possible. > > Whenever we use "sideloading", we mean one of two things: > does it actually matter what *we* use ? its the slang of the android kids, it is what they know and will use when talking about the topic of installing local packages, no matter what we say in our marketing docs ...  i dont mind changing it in docs or watching my personal language to not use it if we decide on a term, but i doubt the crowd will follow if you dont find a term that feels like a natural replacement for them... ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From victor.palau at canonical.com Fri Sep 2 07:14:30 2016 From: victor.palau at canonical.com (Victor Palau) Date: Fri, 2 Sep 2016 08:14:30 +0100 Subject: Let's kill "sideloading" In-Reply-To: <1472800069.5902.11.camel@ubuntu.com> References: <1472800069.5902.11.camel@ubuntu.com> Message-ID: Hi, In the case of installing snaps from the local file system, where will the assertions require to installation reside? Also in the local file system? I can see this use case when either the snap is not yet in the store (during development) or the device has no access to the/a store (maybe behind a firewall) Thanks Victor Thanks Victor On Sep 2, 2016 8:08 AM, "Oliver Grawert" wrote: > hi, > Am Donnerstag, den 01.09.2016, 19:15 -0300 schrieb Gustavo Niemeyer: > > Hello all, > > > > With assertions finally being put to great use, it's time to kill the > > term "sideloading". That term does a disservice to our conversations, > > because it is vague and also limits the thinking around what is > > possible. > > > > Whenever we use "sideloading", we mean one of two things: > > > does it actually matter what *we* use ? its the slang of the android > kids, it is what they know and will use when talking about the topic of > installing local packages, no matter what we say in our marketing docs > ... > > i dont mind changing it in docs or watching my personal language to not > use it if we decide on a term, but i doubt the crowd will follow if you > dont find a term that feels like a natural replacement for them... > > ciao > oli > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stuart.bishop at canonical.com Fri Sep 2 08:01:46 2016 From: stuart.bishop at canonical.com (Stuart Bishop) Date: Fri, 2 Sep 2016 15:01:46 +0700 Subject: Setting release version in snapcraft build In-Reply-To: <57C8696E.70905@canonical.com> References: <57C8696E.70905@canonical.com> Message-ID: On 2 September 2016 at 00:46, Leo Arias wrote: > This is a pain point for me too. Most upstreams store the version number > in only one place, because editing more than one for every tagged > release is too boring. > This is what https://bugs.launchpad.net/snapcraft/+bug/1594794 is about. The plugin can extract a version number in many cases (and plugins could be added that just do this - eg. pull the version from debian/changelog or git tags | grep release | naturalsort | tail -1 or run staging/usr/bin/foo --version or whatever). So a particular part might know its version, and we could declare which part's version is used for the snap. No need to specify anything on the command line, and it would work with Launchpad without extra work. When we add the snapcraft.yaml, I suspect that the version field will be > forgotten often, and it will not be in sync. > Especially when juggling multiple release branches, which differ only by the version number. > However, there are a few issues with this. First, launchpad would need a > new text field to fill the version. And then what would you write in > that field? I see problems if you write a static string and if you write > a command to be appended to the snapcraft run. > > Also, it's not always straight-forward to get the version from the code. > Some projects store it in multiple statements, like major=2; minor=1; > patch=3; Transforming that into "v2.1.3" results in a script that's not > so nice. > > Now, I can see it working at least for the snaps I'm building using > travis and docker. I can get the sha1 when building from master, or the > name of tag when there is one, and use that as the version. I would love > that. > > We could make that snapcraft --snap-version=$(git rev-parse HEAD) just > overwrites the version field, useful immediately for people not using > launchpad. > > pura vida > > > -- Stuart Bishop -------------- next part -------------- An HTML attachment was scrubbed... URL: From stuart.bishop at canonical.com Fri Sep 2 08:38:06 2016 From: stuart.bishop at canonical.com (Stuart Bishop) Date: Fri, 2 Sep 2016 15:38:06 +0700 Subject: Setting release version in snapcraft build In-Reply-To: References: <57C8696E.70905@canonical.com> Message-ID: On 2 September 2016 at 15:01, Stuart Bishop wrote: > > > On 2 September 2016 at 00:46, Leo Arias wrote: > >> This is a pain point for me too. Most upstreams store the version number >> in only one place, because editing more than one for every tagged >> release is too boring. >> > > This is what https://bugs.launchpad.net/snapcraft/+bug/1594794 is about. > > The plugin can extract a version number in many cases (and plugins could > be added that just do this - eg. pull the version from debian/changelog or > git tags | grep release | naturalsort | tail -1 or run staging/usr/bin/foo > --version or whatever). So a particular part might know its version, and we > could declare which part's version is used for the snap. No need to specify > anything on the command line, and it would work with Launchpad without > extra work. > (Or if that is too complicated, just a command that is run at the end of staging or at the start of priming to generate the version number) -- Stuart Bishop -------------- next part -------------- An HTML attachment was scrubbed... URL: From om26er at ubuntu.com Fri Sep 2 09:43:02 2016 From: om26er at ubuntu.com (Omer Akram) Date: Fri, 2 Sep 2016 14:43:02 +0500 Subject: Need LXD for Ubuntu Core 16 image Message-ID: Hi! I am working with Ubuntu Core 16 image[1] on a Raspberry Pi. I am wondering if I there is a LXD snap that I can install on that image ? I found this snap[2] but it seems its not available for Ubuntu Core 16. Is there a place where I can download the LXD snap from ? [1] https://people.canonical.com/~mvo/all-snaps/16/ [2] https://uappexplorer.com/app/lxd.stgraber Thanks! -------------- next part -------------- An HTML attachment was scrubbed... URL: From simon.fels at canonical.com Fri Sep 2 09:54:44 2016 From: simon.fels at canonical.com (Simon Fels) Date: Fri, 2 Sep 2016 11:54:44 +0200 Subject: Need LXD for Ubuntu Core 16 image In-Reply-To: References: Message-ID: On 02.09.2016 11:43, Omer Akram wrote: > Hi! > > I am working with Ubuntu Core 16 image[1] on a Raspberry Pi. I am > wondering if I there is a LXD snap that I can install on that image ? I > found this snap[2] but it seems its not available for Ubuntu Core 16. > > Is there a place where I can download the LXD snap from ? The snap is AFIAK currently only available in the edge channel. You should be able to install it with $ snap install --edge --devmode lxd.stgraber An interface should have landed in snapd too but I can't comment if that is already in use by the lxd snap. regards, Simon From om26er at ubuntu.com Fri Sep 2 09:59:59 2016 From: om26er at ubuntu.com (Omer Akram) Date: Fri, 2 Sep 2016 14:59:59 +0500 Subject: Need LXD for Ubuntu Core 16 image In-Reply-To: References: Message-ID: Thanks Simon for the reply, I still got: error: cannot install "lxd": snap not found And yes, you are right lxd-support interface have landed I can see that with `snap interfaces`. Cheers! On Fri, Sep 2, 2016 at 2:54 PM, Simon Fels wrote: > On 02.09.2016 11:43, Omer Akram wrote: > > Hi! > > > > I am working with Ubuntu Core 16 image[1] on a Raspberry Pi. I am > > wondering if I there is a LXD snap that I can install on that image ? I > > found this snap[2] but it seems its not available for Ubuntu Core 16. > > > > Is there a place where I can download the LXD snap from ? > > The snap is AFIAK currently only available in the edge channel. You > should be able to install it with > > $ snap install --edge --devmode lxd.stgraber > > An interface should have landed in snapd too but I can't comment if that > is already in use by the lxd snap. > > regards, > Simon > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From simon.fels at canonical.com Fri Sep 2 10:10:16 2016 From: simon.fels at canonical.com (Simon Fels) Date: Fri, 2 Sep 2016 12:10:16 +0200 Subject: Need LXD for Ubuntu Core 16 image In-Reply-To: References: Message-ID: <71418cd2-70ad-f22e-5861-b77a4504475c@canonical.com> On 02.09.2016 11:59, Omer Akram wrote: > Thanks Simon for the reply, I still got: > > error: cannot install "lxd": snap not found On which architecture do you try to install the snap? It looks like it is only available for amd64 and armhf. regards, Simon > And yes, you are right lxd-support interface have landed I can see that > with `snap interfaces`. > > Cheers! > > On Fri, Sep 2, 2016 at 2:54 PM, Simon Fels > wrote: > > On 02.09.2016 11:43, Omer Akram wrote: > > Hi! > > > > I am working with Ubuntu Core 16 image[1] on a Raspberry Pi. I am > > wondering if I there is a LXD snap that I can install on that image ? I > > found this snap[2] but it seems its not available for Ubuntu Core 16. > > > > Is there a place where I can download the LXD snap from ? > > The snap is AFIAK currently only available in the edge channel. You > should be able to install it with > > $ snap install --edge --devmode lxd.stgraber > > An interface should have landed in snapd too but I can't comment if that > is already in use by the lxd snap. > > regards, > Simon > > From om26er at ubuntu.com Fri Sep 2 10:18:07 2016 From: om26er at ubuntu.com (Omer Akram) Date: Fri, 2 Sep 2016 15:18:07 +0500 Subject: Need LXD for Ubuntu Core 16 image In-Reply-To: <71418cd2-70ad-f22e-5861-b77a4504475c@canonical.com> References: <71418cd2-70ad-f22e-5861-b77a4504475c@canonical.com> Message-ID: Its a Raspberry Pi 2 so armhf. On Fri, Sep 2, 2016 at 3:10 PM, Simon Fels wrote: > On 02.09.2016 11:59, Omer Akram wrote: > > Thanks Simon for the reply, I still got: > > > > error: cannot install "lxd": snap not found > On which architecture do you try to install the snap? It looks like it is > only available for amd64 and armhf. > > regards, > Simon > > > And yes, you are right lxd-support interface have landed I can see that > > with `snap interfaces`. > > > > Cheers! > > > > On Fri, Sep 2, 2016 at 2:54 PM, Simon Fels > > wrote: > > > > On 02.09.2016 11:43, Omer Akram wrote: > > > Hi! > > > > > > I am working with Ubuntu Core 16 image[1] on a Raspberry Pi. I am > > > wondering if I there is a LXD snap that I can install on that > image ? I > > > found this snap[2] but it seems its not available for Ubuntu Core > 16. > > > > > > Is there a place where I can download the LXD snap from ? > > > > The snap is AFIAK currently only available in the edge channel. You > > should be able to install it with > > > > $ snap install --edge --devmode lxd.stgraber > > > > An interface should have landed in snapd too but I can't comment if > that > > is already in use by the lxd snap. > > > > regards, > > Simon > > > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Fri Sep 2 11:29:17 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 2 Sep 2016 07:29:17 -0400 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: References: Message-ID: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> One of my colleagues wrote regarding his experimental postgres-9-6 RC1 snap, and I'm replying to the list with permission, because I think he raises a lot of good points and perspectives and questions that folks here would have comments on. On 31/08/16 13:08, Stuart Bishop wrote: > > PostgreSQL is well behaved and well polished, so snapping was not > problematic. Unlike what most people are recommending, I started > directly with strict confinement. My gut feeling is that it is better > to trip over issues one at a time when they come up rather than try to > deal with them in bulk once you have something working in devmode. > That said, I didn't trip over many. I think there should be a meme-generator for this. "I don't often make snaps but when I do I start strictly". > > The biggest problem I had was dropping privileges to a non-root user. > There is currently no way to do this (nor to create such a > non-privileged user when the snap is installed). PostgreSQL upstream > block the root user from running tools this, with deliberately no way > to disable it since they traditionally took flak from poorly managed > installs (and they still bitch about Debian using non-standard paths). > I needed to patch several places in the code, and there may be some > more lurking in there not triggered by the PostgreSQL test suite. OK, we definitely want to support this. I think the scaffolding we need is: * such user names should be managed globally (i.e. assigned in an assertion) * such user names should by design not conflict with real users I think this requirement has come up before and what we decided was to pre-reserve a name in Ubuntu which would be in /etc/passwd up front. Perhaps this was for LXD, I forget. That's not a scalable solution but it might work for Pg. > Usability is currently poor. PostgreSQL is cli heavy, so having all > the well known commands with munged names like > 'postgresql-9-6.pg-dump' instead of 'pg_dump' would stop adoption all > by itself. I understand that this is being addressed, and I will be > able to present the dozen or so commands with their preferred names. > But this will also cause different issues, when I snap postgresql-10. > postgresql-10 will have the same set of tools. So I'll have two sets > on my path, each only able to deal with their own confinement. I think > the search path needs a defined ordering (eg. alphabetically), and > tools need to be available in both their prefixed and unprefixed form. > I will need to have both postgresql-9-6 and postgresql-10 snaps > installed in the same container, as this is the only migration path I > can come up with (allowing postgresql-10.pg_upgrade access to the > postgresql-9-6 containment via the content interface). Yes, these are very useful practical items of feedback for the command work. Let's promise to take that up in a sprint, when it's close to the top of our priority list. As a straw man it seems that we want groups of snaps (your pg versions) to be able to overlap in command names, but have only one of them get that top-level space at a time. That's a little bit like update-alternatives but with a snap as the level of granularity. If pg-10 is your default then all the top-level commands are pg-10, others are pg-X.command. > > PostgreSQL is very extensible, and I haven't worked out the best way > to handle it. This includes adding arbitrary 3rd party Python, Perl > and tcl libraries for use from the built in stored procedure > languages. This includes building, linking and installing entirely new > stored procedure languages such as Ruby and all their dependencies and > extensions. This includes building, linking and installing C stored > procedures. This includes building other PostgreSQL extensions and > installing them where they can be installed using 'CREATE EXTENSION', > such as Citus for massive scale out and sharding or BDR for multi > master replication. The ecosystem is too vast to package everything, > and a lot of people are using bespoke and proprietary tools. Yes, the general question about plugins and extensions is super interesting. Snaps offer tight and deterministic binaries on demand, and that slightly conflicts with extensible ecosystems, but we must explore the boundaries of the known world to better understand where we're going. Simplistically, a snap author could make an ecosystem of things they bring into their snap. But what you want is a little less controlled than that. > > Launchpad was invaluable, as my attention span doesn't cover uploading > 100MB++ blobs over my awful 3rd world ADSL connection. Awful hacks > were used to get test suites running as part of the build process from > the static snapcraft.yaml, which in hindsite would have been done > better as a custom plugin. > Yeah, a custom plugin is a much easier way to solve build issues than I expected too. We could do a much better job of helping people write such plugins, for example by documenting the BasePlugin much better, perhaps as a skeleton that 'snapcraft init-plugin' gives you. > I needed to write my own log rotation daemon. There seems to be no way > to schedule regular operations. System logging locally via syslog and systemd journal should be straightforward interfaces (I think the latter is in-progress by Jamie). The system should handle rotation of *those* logs. I like the idea of having log rotation as a capability for logs in $SNAP_DATA and cousins. > locales was tricky to get right, and I don't know if I got it right. I > need *all* the locales available, as you can declare indexes and > columns in particular locales to get particular collation orders. I > ended up pulling in locales-all from universe as a stage-package, and > my wrappers set LOCPATH to the snapped locales location. Yes, I think numerous folks have run into this. I think an official solution for locales-in-snap and locales-from-system would be most useful. > Access to dotfiles in $HOME is necessary, like most of the CLI tools > I'm interested in snapping. I suspect the use case for a home > interface that blocks 'hidden' files is small, possibly non-existent, > and it might be best to just remove the restriction than add a new > interface. Noted. I'd like to preserve the current plan, which is to extend the home interface for parameterised access to known dotfiles, but this is another good data point to consider. > > Storage is going to be a problem. I can't use the snap in the > PostgreSQL juju charm, as I need to store WAL and data files on > partitions provided by Juju, outside of containment. Non-juju > production installs will also need to split WAL and datafiles onto > separate partitions (even with pure SSD you can want separate channels > for the increased bandwidth). And another potentially huge issue is > that the database gets destroyed when the snap is uninstalled. I get > the feeling that I need to be able to define arbitrary paths paths on > the host to be accessible from inside containment. Yes we will need to look again at storage for really serious data-persistence software like Postgres. > And on opening up paths on the host, unix sockets. By default, > PostgreSQL clients try to connect to the server using sockets stored > in a well known directory (/var/run/postgresql on Debian). Tools > compiled against the standard PostgreSQL packages will not find the > listening socket in /var/run/postgresql, but need to be told to look > in /var/snap/postgresql-9-6/common/run. If the server running inside > containment could write to /var/run/postgresql, then everyone could > play together happily. > This should be handled pretty cleanly by an interface, and I would suggest going ahead and making one for Postgres right away as that will allow other snaps to use Postgres if they want it. > Upgrades will be slow, as data (possibly terabytes) will need to be > copied from the postgresql-9-6 container into the postgresql-10 > container. Non-snap installs get to hardlink the datafiles into the > new location and migrate them quickly. This is another use case for > allowing access to areas of the host. The content interface should enable this - a snap could offer the data spaces up to different versions of postgres. I'm not sure if the content interface is yet ready for this particular use-case, but its definitely in progress. > Where is the best place to store data? I'm using $SNAP_COMMON because > I can't let the data be rolled back (it might be a good idea on a > standalone system, but rollback would desynchronize and break a > replicated system). > For now that's probably best, until we have the content interface settled. > How to customize the snap? PostgreSQL can be customized by sudo vi > /var/snap/postgresql-9-6/common/data/postgresql.conf, but for snap > configuration I went with /var/snap/postgresql-9-6/common/snap.ini > (currently it just defines the log retention time). For now I think an editable config file in a known writable location is best. Later we'll have a standardised mechanism which your snap can use to hook into bigger systems. > I have lots and lots and lots of man pages not on my MANPATH. Not > urgent, but nice to expose them. > Noted and I think in-progress. > For general adoption, the bar is really high. PostgreSQL running in an > machine container is exactly what people want (and what we have been > pushing on the Cloud side of things). Even if we can address all the > above issues, a PostgreSQL snap will still be more cumbersome than > running PostgreSQL in an lxd container (or a docker container for that > matter). Operators get to control and grow the environment they need, > rather than being stuck with the static one I provide and the > interfaces, plugs and canned scripts that customize it in preapproved > ways. I've been wondering if the solution is adding a new 'machine' > confinement, where the snap software is installed and updated in a > real lxd machine container on the host. Maybe the best of both worlds? Let's start with a more focused used-case: postgres as a storage engine for IoT. In that case the level of customization is much reduced, people are looking for some highly predictable behavior more than every possible degree of freedom in customization. Remember they can also build postgres INTO their own snap (and it actually might be worth exporting a shared pg part so people can do that trivially). Mark From yann.sionneau at parrot.com Fri Sep 2 12:46:58 2016 From: yann.sionneau at parrot.com (Yann Sionneau) Date: Fri, 2 Sep 2016 14:46:58 +0200 Subject: Snappy takes long time to boot when no network In-Reply-To: References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> <1472730470.15344.14.camel@ubuntu.com> <1472733028.15344.29.camel@ubuntu.com> Message-ID: <84e58e9b-2070-95e1-70a3-01b81548c8da@parrot.com> Your solution fixed my issue : root at Paros:~# rm /etc/netplan/00-initial-config.yaml root at Paros:~# netplan apply Cannot replug rndis0: [Errno 2] No such file or directory: '/sys/devices/platform/tegra-xudc/gadget/driver/unbind' and then reboot. I now boot in 24 seconds instead of 140 seconds :) Thanks! Le 09/01/2016 à 11:19 PM, Michael Hudson-Doyle a écrit : > > > On 2 September 2016 at 00:44, Yann Sionneau > wrote: > > I'm not sure I understand everything in the bug ticket as I am not a > systemd / networkd / netplan expert at all. > > But : > > root at Paros:~# cat /etc/netplan/00- > 00-initial-config.yaml 00-snapd-config.yaml > root at Paros:~# cat /etc/netplan/00-initial-config.yaml > > network: > version: 2 > ethernets: > all: > match: > name: "*" > dhcp4: true > root at Paros:~# cat /run/systemd/network/10-netplan-all.network > [Match] > Name=* > > [Network] > DHCP=ipv4 > root at Paros:~# networkctl > IDX LINK TYPE OPERATIONAL SETUP > 1 lo loopback carrier configured > 2 sit0 sit routable configuring > 3 eth1 ether routable configured > 4 eth0 ether routable configured > 5 rndis0 ether no-carrier configuring > > 5 links listed. > > Does this help? > > eth0 is a WiFi interface. > > eth1 is USB ethernet device (plugged at boot) > > > I don't know if it's the cause of all your issues, but having both > files in /etc/netplan is a sign of console-conf / snapd version skew. > It's fixed now and new installs won't have this problem, but if you > don't want to re-install just deleted the 00-initial-config.yaml file > and run sudo netplan apply. > > Cheers, > mwh > > > Le 09/01/2016 à 02:30 PM, Oliver Grawert a écrit : > > hi, > > On Do, 2016-09-01 at 14:05 +0200, Yann Sionneau wrote: > >> I feel like this is the bug I'm hitting. > >> But, how do you explain that my boot is stalled even if I have > 2 NICs > >> with internet access? > >> One is via wifi, the other is via usb-ethernet. > >> Thanks! > > hmm, this sounds more like a different bug ... > > how about: > > > > https://launchpad.net/bugs/1618522 > > > > > ciao > > oli > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From espy at canonical.com Fri Sep 2 13:35:39 2016 From: espy at canonical.com (Tony Espy) Date: Fri, 2 Sep 2016 09:35:39 -0400 Subject: Let's kill "sideloading" In-Reply-To: References: Message-ID: <0d1dd0a5-f76f-470c-fedd-56c5586b7d2c@canonical.com> On 09/01/2016 06:15 PM, Gustavo Niemeyer wrote: > Hello all, > > With assertions finally being put to great use, it's time to kill the > term "sideloading". That term does a disservice to our conversations, > because it is vague and also limits the thinking around what is possible. I have a question related to "sideloading" a snap. Yesterday while testing a fix for our network-manager snap, I refreshed my rpi2 ( running the 'experimental' image ) which resulted in a new ubuntu-core snap, which I discovered now enforces the assertion that a snap must be signed in order to install, even when side-loaded. I was told on #snappy that I could circumvent this check via the --force-dangerous parameter, which worked for me. I was also told that this parameter may just be shortened to "--dangerous", and that "--devmode" may cause this to automatically set. My question is what is the process for getting a snap signed? Is this something that's done automatically when a snap is published to the store? The snap I was testing was built by launchpad. Is it possible to sign a snap locally ( ie. like debsign )? Regards, /tony From ogra at ubuntu.com Fri Sep 2 13:43:24 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 02 Sep 2016 15:43:24 +0200 Subject: Let's kill "sideloading" In-Reply-To: <0d1dd0a5-f76f-470c-fedd-56c5586b7d2c@canonical.com> References: <0d1dd0a5-f76f-470c-fedd-56c5586b7d2c@canonical.com> Message-ID: <1472823804.5902.18.camel@ubuntu.com> hi, Am Freitag, den 02.09.2016, 09:35 -0400 schrieb Tony Espy: > My question is what is the process for getting a snap signed?  Is > this  > something that's done automatically when a snap is published to the > store? yes ... this is just about a store signature .. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From gustavo at niemeyer.net Fri Sep 2 13:58:36 2016 From: gustavo at niemeyer.net (Gustavo Niemeyer) Date: Fri, 2 Sep 2016 10:58:36 -0300 Subject: Let's kill "sideloading" In-Reply-To: References: <1472800069.5902.11.camel@ubuntu.com> Message-ID: Hey Victor, On Fri, Sep 2, 2016 at 4:14 AM, Victor Palau wrote: > Hi, > > In the case of installing snaps from the local file system, where will the > assertions require to installation reside? Also in the local file system? > The knowledge of which assertions are being respected is internal to snapd. It may be fed new assertions at any point by running "snap ack file.assert", and it also retrieves new assertions from the server end when necessary during operations (install, etc). > I can see this use case when either the snap is not yet in the store > (during development) or the device has no access to the/a store (maybe > behind a firewall) > Right. It effectively disentangles system knowledge from access to a particular server or network. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Fri Sep 2 14:10:59 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Fri, 2 Sep 2016 11:10:59 -0300 Subject: Let's kill "sideloading" In-Reply-To: <0d1dd0a5-f76f-470c-fedd-56c5586b7d2c@canonical.com> References: <0d1dd0a5-f76f-470c-fedd-56c5586b7d2c@canonical.com> Message-ID: On Fri, Sep 2, 2016 at 10:35 AM, Tony Espy wrote: > On 09/01/2016 06:15 PM, Gustavo Niemeyer wrote: > >> Hello all, >> >> With assertions finally being put to great use, it's time to kill the >> term "sideloading". That term does a disservice to our conversations, >> because it is vague and also limits the thinking around what is possible. >> > > I have a question related to "sideloading" a snap. > > Yesterday while testing a fix for our network-manager snap, I refreshed my > rpi2 ( running the 'experimental' image ) which resulted in a new > ubuntu-core snap, which I discovered now enforces the assertion that a snap > must be signed in order to install, even when side-loaded. I was told on > #snappy that I could circumvent this check via the --force-dangerous > parameter, which worked for me. I was also told that this parameter may > just be shortened to "--dangerous", and that "--devmode" may cause this to > automatically set. > Indeed, we'll do those changes in the next couple of days. > My question is what is the process for getting a snap signed? Is this > something that's done automatically when a snap is published to the store? > Yes, the goal is for the whole process to be mostly transparent. When you build a snap you'll get an assertion next to it saying that you built it. When you upload it, the assertion is shipped to the server, the snap gets additional server assertions backing that process. No effort on the developer end. The snap I was testing was built by launchpad. Is it possible to sign a > snap locally ( ie. like debsign )? > Yes, Launchpad is likely using snapcraft already, which means it'll do that by default once updated. We'll need to put a developer key there, though. Sergio and Colin Watson should know more details here. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From wesley.mason at canonical.com Fri Sep 2 14:23:15 2016 From: wesley.mason at canonical.com (Wes Mason) Date: Fri, 02 Sep 2016 14:23:15 +0000 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> Message-ID: On Fri, 2 Sep 2016, 12:30 pm Mark Shuttleworth, wrote: > > One of my colleagues wrote regarding his experimental postgres-9-6 RC1 > snap, and I'm replying to the list with permission, because I think he > raises a lot of good points and perspectives and questions that folks > here would have comments on. > > On 31/08/16 13:08, Stuart Bishop wrote: > > > > PostgreSQL is well behaved and well polished, so snapping was not > > problematic. Unlike what most people are recommending, I started > > directly with strict confinement. My gut feeling is that it is better > > to trip over issues one at a time when they come up rather than try to > > deal with them in bulk once you have something working in devmode. > > That said, I didn't trip over many. > > I think there should be a meme-generator for this. "I don't often make > snaps but when I do I start strictly". > https://cdn.meme.am/instances/500x/71468200.jpg > > > > > The biggest problem I had was dropping privileges to a non-root user. > > There is currently no way to do this (nor to create such a > > non-privileged user when the snap is installed). PostgreSQL upstream > > block the root user from running tools this, with deliberately no way > > to disable it since they traditionally took flak from poorly managed > > installs (and they still bitch about Debian using non-standard paths). > > I needed to patch several places in the code, and there may be some > > more lurking in there not triggered by the PostgreSQL test suite. > > OK, we definitely want to support this. > > I think the scaffolding we need is: > > * such user names should be managed globally (i.e. assigned in an > assertion) > * such user names should by design not conflict with real users > > I think this requirement has come up before and what we decided was to > pre-reserve a name in Ubuntu which would be in /etc/passwd up front. > Perhaps this was for LXD, I forget. That's not a scalable solution but > it might work for Pg. > > > Usability is currently poor. PostgreSQL is cli heavy, so having all > > the well known commands with munged names like > > 'postgresql-9-6.pg-dump' instead of 'pg_dump' would stop adoption all > > by itself. I understand that this is being addressed, and I will be > > able to present the dozen or so commands with their preferred names. > > But this will also cause different issues, when I snap postgresql-10. > > postgresql-10 will have the same set of tools. So I'll have two sets > > on my path, each only able to deal with their own confinement. I think > > the search path needs a defined ordering (eg. alphabetically), and > > tools need to be available in both their prefixed and unprefixed form. > > I will need to have both postgresql-9-6 and postgresql-10 snaps > > installed in the same container, as this is the only migration path I > > can come up with (allowing postgresql-10.pg_upgrade access to the > > postgresql-9-6 containment via the content interface). > > Yes, these are very useful practical items of feedback for the command > work. Let's promise to take that up in a sprint, when it's close to the > top of our priority list. As a straw man it seems that we want groups of > snaps (your pg versions) to be able to overlap in command names, but > have only one of them get that top-level space at a time. That's a > little bit like update-alternatives but with a snap as the level of > granularity. If pg-10 is your default then all the top-level commands > are pg-10, others are pg-X.command. > > > > > > PostgreSQL is very extensible, and I haven't worked out the best way > > to handle it. This includes adding arbitrary 3rd party Python, Perl > > and tcl libraries for use from the built in stored procedure > > languages. This includes building, linking and installing entirely new > > stored procedure languages such as Ruby and all their dependencies and > > extensions. This includes building, linking and installing C stored > > procedures. This includes building other PostgreSQL extensions and > > installing them where they can be installed using 'CREATE EXTENSION', > > such as Citus for massive scale out and sharding or BDR for multi > > master replication. The ecosystem is too vast to package everything, > > and a lot of people are using bespoke and proprietary tools. > > Yes, the general question about plugins and extensions is super > interesting. Snaps offer tight and deterministic binaries on demand, and > that slightly conflicts with extensible ecosystems, but we must explore > the boundaries of the known world to better understand where we're going. > > Simplistically, a snap author could make an ecosystem of things they > bring into their snap. But what you want is a little less controlled > than that. > > > > > Launchpad was invaluable, as my attention span doesn't cover uploading > > 100MB++ blobs over my awful 3rd world ADSL connection. Awful hacks > > were used to get test suites running as part of the build process from > > the static snapcraft.yaml, which in hindsite would have been done > > better as a custom plugin. > > > > Yeah, a custom plugin is a much easier way to solve build issues than I > expected too. We could do a much better job of helping people write such > plugins, for example by documenting the BasePlugin much better, perhaps > as a skeleton that 'snapcraft init-plugin' gives you. > > > I needed to write my own log rotation daemon. There seems to be no way > > to schedule regular operations. > > System logging locally via syslog and systemd journal should be > straightforward interfaces (I think the latter is in-progress by Jamie). > The system should handle rotation of *those* logs. I like the idea of > having log rotation as a capability for logs in $SNAP_DATA and cousins. > > > locales was tricky to get right, and I don't know if I got it right. I > > need *all* the locales available, as you can declare indexes and > > columns in particular locales to get particular collation orders. I > > ended up pulling in locales-all from universe as a stage-package, and > > my wrappers set LOCPATH to the snapped locales location. > > Yes, I think numerous folks have run into this. I think an official > solution for locales-in-snap and locales-from-system would be most useful. > > > Access to dotfiles in $HOME is necessary, like most of the CLI tools > > I'm interested in snapping. I suspect the use case for a home > > interface that blocks 'hidden' files is small, possibly non-existent, > > and it might be best to just remove the restriction than add a new > > interface. > > Noted. I'd like to preserve the current plan, which is to extend the > home interface for parameterised access to known dotfiles, but this is > another good data point to consider. > > > > > Storage is going to be a problem. I can't use the snap in the > > PostgreSQL juju charm, as I need to store WAL and data files on > > partitions provided by Juju, outside of containment. Non-juju > > production installs will also need to split WAL and datafiles onto > > separate partitions (even with pure SSD you can want separate channels > > for the increased bandwidth). And another potentially huge issue is > > that the database gets destroyed when the snap is uninstalled. I get > > the feeling that I need to be able to define arbitrary paths paths on > > the host to be accessible from inside containment. > > Yes we will need to look again at storage for really serious > data-persistence software like Postgres. > > > And on opening up paths on the host, unix sockets. By default, > > PostgreSQL clients try to connect to the server using sockets stored > > in a well known directory (/var/run/postgresql on Debian). Tools > > compiled against the standard PostgreSQL packages will not find the > > listening socket in /var/run/postgresql, but need to be told to look > > in /var/snap/postgresql-9-6/common/run. If the server running inside > > containment could write to /var/run/postgresql, then everyone could > > play together happily. > > > > This should be handled pretty cleanly by an interface, and I would > suggest going ahead and making one for Postgres right away as that will > allow other snaps to use Postgres if they want it. > > > Upgrades will be slow, as data (possibly terabytes) will need to be > > copied from the postgresql-9-6 container into the postgresql-10 > > container. Non-snap installs get to hardlink the datafiles into the > > new location and migrate them quickly. This is another use case for > > allowing access to areas of the host. > > The content interface should enable this - a snap could offer the data > spaces up to different versions of postgres. I'm not sure if the content > interface is yet ready for this particular use-case, but its definitely > in progress. > > > Where is the best place to store data? I'm using $SNAP_COMMON because > > I can't let the data be rolled back (it might be a good idea on a > > standalone system, but rollback would desynchronize and break a > > replicated system). > > > > For now that's probably best, until we have the content interface settled. > > > How to customize the snap? PostgreSQL can be customized by sudo vi > > /var/snap/postgresql-9-6/common/data/postgresql.conf, but for snap > > configuration I went with /var/snap/postgresql-9-6/common/snap.ini > > (currently it just defines the log retention time). > > For now I think an editable config file in a known writable location is > best. Later we'll have a standardised mechanism which your snap can use > to hook into bigger systems. > > > I have lots and lots and lots of man pages not on my MANPATH. Not > > urgent, but nice to expose them. > > > > Noted and I think in-progress. > > > For general adoption, the bar is really high. PostgreSQL running in an > > machine container is exactly what people want (and what we have been > > pushing on the Cloud side of things). Even if we can address all the > > above issues, a PostgreSQL snap will still be more cumbersome than > > running PostgreSQL in an lxd container (or a docker container for that > > matter). Operators get to control and grow the environment they need, > > rather than being stuck with the static one I provide and the > > interfaces, plugs and canned scripts that customize it in preapproved > > ways. I've been wondering if the solution is adding a new 'machine' > > confinement, where the snap software is installed and updated in a > > real lxd machine container on the host. Maybe the best of both worlds? > > Let's start with a more focused used-case: postgres as a storage engine > for IoT. In that case the level of customization is much reduced, people > are looking for some highly predictable behavior more than every > possible degree of freedom in customization. Remember they can also > build postgres INTO their own snap (and it actually might be worth > exporting a shared pg part so people can do that trivially). > > Mark > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo at niemeyer.net Fri Sep 2 14:37:22 2016 From: gustavo at niemeyer.net (Gustavo Niemeyer) Date: Fri, 2 Sep 2016 11:37:22 -0300 Subject: Let's kill "sideloading" In-Reply-To: <1472800069.5902.11.camel@ubuntu.com> References: <1472800069.5902.11.camel@ubuntu.com> Message-ID: Hi Oliver, On Fri, Sep 2, 2016 at 4:07 AM, Oliver Grawert wrote: > > does it actually matter what *we* use ? its the slang of the android > kids, it is what they know and will use when talking about the topic of installing local packages, The terminology we use for our conversations absolutely matters. The problem is precisely that people using "sideloaded" today already have a clear picture in mind of what it means, and when you have richer semantics that don't match what they mean, using the term does more damage than it helps. Speaking of terminology, it also feels a bit awkward to be talking about Android developers as "kids". I know you mean it well, but it's derogatory. > no matter what we say in our marketing docs > This is not about marketing. It's about what we talk on our daily conversations. > i dont mind changing it in docs or watching my personal language to not > use it if we decide on a term, but i doubt the crowd will follow if you > dont find a term that feels like a natural replacement for them... > In my experience terminology used daily by the larger development group always ends up being used everywhere. Both the good terminology and the bad one. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Fri Sep 2 14:59:10 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 02 Sep 2016 16:59:10 +0200 Subject: Let's kill "sideloading" In-Reply-To: References: <1472800069.5902.11.camel@ubuntu.com> Message-ID: <1472828350.5902.24.camel@ubuntu.com> hi, Am Freitag, den 02.09.2016, 11:37 -0300 schrieb Gustavo Niemeyer: > > Speaking of terminology, it also feels a bit awkward to be talking > about Android developers as "kids". I know you mean it well, but it's > derogatory. well, if i had meant developers i had said that ... with "kids" i was exactly referring to the persons that blindly grab some binary android image that someone they do not know provides on the xda forums ;)  (which is the bigger majority i guess) admittedly "kids" was a bit harsh, i apologize to anyone who feels offended by this ...  ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From sergio.schvezov at canonical.com Fri Sep 2 15:55:52 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Fri, 2 Sep 2016 12:55:52 -0300 Subject: Let's kill "sideloading" In-Reply-To: References: <0d1dd0a5-f76f-470c-fedd-56c5586b7d2c@canonical.com> Message-ID: <79a2a820-6089-00aa-a24b-529d2161e66b@canonical.com> El 02/09/16 a las 11:10, Gustavo Niemeyer escribió: > > > On Fri, Sep 2, 2016 at 10:35 AM, Tony Espy > wrote: > > On 09/01/2016 06:15 PM, Gustavo Niemeyer wrote: > > Hello all, > > With assertions finally being put to great use, it's time to > kill the > term "sideloading". That term does a disservice to our > conversations, > because it is vague and also limits the thinking around what > is possible. > > > I have a question related to "sideloading" a snap. > > Yesterday while testing a fix for our network-manager snap, I > refreshed my rpi2 ( running the 'experimental' image ) which > resulted in a new ubuntu-core snap, which I discovered now > enforces the assertion that a snap must be signed in order to > install, even when side-loaded. I was told on #snappy that I > could circumvent this check via the --force-dangerous parameter, > which worked for me. I was also told that this parameter may just > be shortened to "--dangerous", and that "--devmode" may cause this > to automatically set. > > > Indeed, we'll do those changes in the next couple of days. > > My question is what is the process for getting a snap signed? Is > this something that's done automatically when a snap is published > to the store? > > > Yes, the goal is for the whole process to be mostly transparent. When > you build a snap you'll get an assertion next to it saying that you > built it. When you upload it, the assertion is shipped to the server, > the snap gets additional server assertions backing that process. No > effort on the developer end. > > The snap I was testing was built by launchpad. Is it possible to > sign a snap locally ( ie. like debsign )? > > > Yes, Launchpad is likely using snapcraft already, which means it'll do > that by default once updated. We'll need to put a developer key there, > though. So I guess what Tony might get value on knowing is which assertion needs to be available to avoid --devmode/--dangerous. > > Sergio and Colin Watson should know more details here. Maybe subscribe to https://github.com/snapcore/snapcraft/pull/726 and https://bugs.launchpad.net/snapcraft/+bug/1612730 > > > gustavo @ http://niemeyer.net > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stgraber at ubuntu.com Fri Sep 2 16:23:50 2016 From: stgraber at ubuntu.com (=?iso-8859-1?Q?St=E9phane?= Graber) Date: Fri, 2 Sep 2016 12:23:50 -0400 Subject: Need LXD for Ubuntu Core 16 image In-Reply-To: References: <71418cd2-70ad-f22e-5861-b77a4504475c@canonical.com> Message-ID: <20160902162350.GD15622@castiana> Hi, The LXD snap for series 16 will be published under the canonical account, not under mine. As was mentioned earlier, we needed the lxd-support interface to land, which it now has but we're blocked on a snap-confine bug which prevents LXD from running right now. Once snap-confine is sorted out, we should be able to build and publish a new LXD snap for series 16. Note that there may still be some blockers for an all snap system. Testing so far is only being done on classic. One possible issue that may need resolving, if it wasn't already, is the lack of a "lxd" group on the system. LXD needs this for access control and there is currently no interface or mechanism we can use to create it. On classic we just assume it already exists on the system, but for other systems we'll either need a mechanism for snapd to create the group for us or we'll need snapd itself to create that group on all systems. Stéphane On Fri, Sep 02, 2016 at 03:18:07PM +0500, Omer Akram wrote: > Its a Raspberry Pi 2 so armhf. > > On Fri, Sep 2, 2016 at 3:10 PM, Simon Fels wrote: > > > On 02.09.2016 11:59, Omer Akram wrote: > > > Thanks Simon for the reply, I still got: > > > > > > error: cannot install "lxd": snap not found > > On which architecture do you try to install the snap? It looks like it is > > only available for amd64 and armhf. > > > > regards, > > Simon > > > > > And yes, you are right lxd-support interface have landed I can see that > > > with `snap interfaces`. > > > > > > Cheers! > > > > > > On Fri, Sep 2, 2016 at 2:54 PM, Simon Fels > > > wrote: > > > > > > On 02.09.2016 11:43, Omer Akram wrote: > > > > Hi! > > > > > > > > I am working with Ubuntu Core 16 image[1] on a Raspberry Pi. I am > > > > wondering if I there is a LXD snap that I can install on that > > image ? I > > > > found this snap[2] but it seems its not available for Ubuntu Core > > 16. > > > > > > > > Is there a place where I can download the LXD snap from ? > > > > > > The snap is AFIAK currently only available in the edge channel. You > > > should be able to install it with > > > > > > $ snap install --edge --devmode lxd.stgraber > > > > > > An interface should have landed in snapd too but I can't comment if > > that > > > is already in use by the lxd snap. > > > > > > regards, > > > Simon -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: not available URL: From sergio.schvezov at canonical.com Fri Sep 2 16:29:21 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Fri, 2 Sep 2016 13:29:21 -0300 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> Message-ID: <7ffb5326-868a-a9e7-0ee3-2b9409f73a76@canonical.com> El 02/09/16 a las 08:29, Mark Shuttleworth escribió: >> >Usability is currently poor. PostgreSQL is cli heavy, so having all >> >the well known commands with munged names like >> >'postgresql-9-6.pg-dump' instead of 'pg_dump' would stop adoption all >> >by itself. I understand that this is being addressed, and I will be >> >able to present the dozen or so commands with their preferred names. >> >But this will also cause different issues, when I snap postgresql-10. >> >postgresql-10 will have the same set of tools. So I'll have two sets >> >on my path, each only able to deal with their own confinement. I think >> >the search path needs a defined ordering (eg. alphabetically), and >> >tools need to be available in both their prefixed and unprefixed form. >> >I will need to have both postgresql-9-6 and postgresql-10 snaps >> >installed in the same container, as this is the only migration path I >> >can come up with (allowing postgresql-10.pg_upgrade access to the >> >postgresql-9-6 containment via the content interface). > Yes, these are very useful practical items of feedback for the command > work. Let's promise to take that up in a sprint, when it's close to the > top of our priority list. As a straw man it seems that we want groups of > snaps (your pg versions) to be able to overlap in command names, but > have only one of them get that top-level space at a time. That's a > little bit like update-alternatives but with a snap as the level of > granularity. If pg-10 is your default then all the top-level commands > are pg-10, others are pg-X.command. > > A solution for this that could work today is to have some sort of env setup (like done with python virtual envs), source the correct env and with aliases or PATH setups make sure the right top level names are seen. Here's a friday parlor trick source $(pg.env) And `pg.env` would just echo the path to the file withing the snap to source. Not totally user friendly and probably could be made more straightforward but it does get the job done. Maybe one day we can get a `snap env ` command (thinking out loud/writing), who knows :-) -------------- next part -------------- An HTML attachment was scrubbed... URL: From mhall119 at ubuntu.com Fri Sep 2 16:59:46 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Fri, 2 Sep 2016 12:59:46 -0400 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: <7ffb5326-868a-a9e7-0ee3-2b9409f73a76@canonical.com> References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> <7ffb5326-868a-a9e7-0ee3-2b9409f73a76@canonical.com> Message-ID: On 09/02/2016 12:29 PM, Sergio Schvezov wrote: > El 02/09/16 a las 08:29, Mark Shuttleworth escribió: >>> > Usability is currently poor. PostgreSQL is cli heavy, so having all >>> > the well known commands with munged names like >>> > 'postgresql-9-6.pg-dump' instead of 'pg_dump' would stop adoption all >>> > by itself. I understand that this is being addressed, and I will be >>> > able to present the dozen or so commands with their preferred names. >>> > But this will also cause different issues, when I snap postgresql-10. >>> > postgresql-10 will have the same set of tools. So I'll have two sets >>> > on my path, each only able to deal with their own confinement. I think >>> > the search path needs a defined ordering (eg. alphabetically), and >>> > tools need to be available in both their prefixed and unprefixed form. >>> > I will need to have both postgresql-9-6 and postgresql-10 snaps >>> > installed in the same container, as this is the only migration path I >>> > can come up with (allowing postgresql-10.pg_upgrade access to the >>> > postgresql-9-6 containment via the content interface). >> Yes, these are very useful practical items of feedback for the command >> work. Let's promise to take that up in a sprint, when it's close to the >> top of our priority list. As a straw man it seems that we want groups of >> snaps (your pg versions) to be able to overlap in command names, but >> have only one of them get that top-level space at a time. That's a >> little bit like update-alternatives but with a snap as the level of >> granularity. If pg-10 is your default then all the top-level commands >> are pg-10, others are pg-X.command. >> >> > > A solution for this that could work today is to have some sort of env > setup (like done with python virtual envs), source the correct env and > with aliases or PATH setups make sure the right top level names are seen. > > Here's a friday parlor trick > > source $(pg.env) > > And `pg.env` would just echo the path to the file withing the snap to > source. Not totally user friendly and probably could be made more > straightforward but it does get the job done. Maybe one day we can get a > `snap env ` command (thinking out loud/writing), who knows :-) > > That would only work if the overlapping command name are from the same upstream project (postgresql in this example) but what if they are from different upstreams? For instance, some other database (let's say mariadb for sake of example) might also have a createdb command. Then you might page postgresql-9-6.createdb, postgresql-10.createdb and mariadb.createdb. Michael Hall mhall119 at ubuntu.com From evan.dandrea at canonical.com Fri Sep 2 17:06:19 2016 From: evan.dandrea at canonical.com (Evan Dandrea) Date: Fri, 02 Sep 2016 17:06:19 +0000 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> Message-ID: On Fri, 2 Sep 2016 at 04:30 Mark Shuttleworth wrote: > Remember they can also build postgres INTO their own snap (and it actually > might be worth exporting a shared pg part so people can do that trivially). > And this is exactly what happened with Mongo. Gustavo wrote a Mongo snap. The Rocket.Chat guys picked it up and used it as a part. Then we pushed it to the parts server (`snapcraft search mongodb`) so others can benefit. +1 to a postgres part. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mhall119 at ubuntu.com Fri Sep 2 17:15:43 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Fri, 2 Sep 2016 13:15:43 -0400 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> Message-ID: <27768989-e4d3-840d-f11d-e8616e110286@ubuntu.com> On 09/02/2016 01:06 PM, Evan Dandrea wrote: > On Fri, 2 Sep 2016 at 04:30 Mark Shuttleworth > wrote: > > Remember they can also build postgres INTO their own snap (and it > actually might be worth exporting a shared pg part so people can do > that trivially). > > > And this is exactly what happened with Mongo. Gustavo wrote a Mongo > snap. The Rocket.Chat guys picked it up and used it as a part. Then we > pushed it to the parts server (`snapcraft search mongodb`) so others can > benefit. > > +1 to a postgres part. > > Yet another option would be to use the new content interface to build postgresql snap that can share it's binaries and libraries into your snap, so you can call them to setup and manage a local database in your own application's directories. Michael Hall mhall119 at ubuntu.com From sergio.schvezov at canonical.com Fri Sep 2 17:24:34 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Fri, 2 Sep 2016 14:24:34 -0300 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> <7ffb5326-868a-a9e7-0ee3-2b9409f73a76@canonical.com> Message-ID: <5b702d89-377a-7a3b-d61c-56e9b2616334@canonical.com> El 02/09/16 a las 13:59, Michael Hall escribió: > On 09/02/2016 12:29 PM, Sergio Schvezov wrote: >> El 02/09/16 a las 08:29, Mark Shuttleworth escribió: >>>>> Usability is currently poor. PostgreSQL is cli heavy, so having all >>>>> the well known commands with munged names like >>>>> 'postgresql-9-6.pg-dump' instead of 'pg_dump' would stop adoption all >>>>> by itself. I understand that this is being addressed, and I will be >>>>> able to present the dozen or so commands with their preferred names. >>>>> But this will also cause different issues, when I snap postgresql-10. >>>>> postgresql-10 will have the same set of tools. So I'll have two sets >>>>> on my path, each only able to deal with their own confinement. I think >>>>> the search path needs a defined ordering (eg. alphabetically), and >>>>> tools need to be available in both their prefixed and unprefixed form. >>>>> I will need to have both postgresql-9-6 and postgresql-10 snaps >>>>> installed in the same container, as this is the only migration path I >>>>> can come up with (allowing postgresql-10.pg_upgrade access to the >>>>> postgresql-9-6 containment via the content interface). >>> Yes, these are very useful practical items of feedback for the command >>> work. Let's promise to take that up in a sprint, when it's close to the >>> top of our priority list. As a straw man it seems that we want groups of >>> snaps (your pg versions) to be able to overlap in command names, but >>> have only one of them get that top-level space at a time. That's a >>> little bit like update-alternatives but with a snap as the level of >>> granularity. If pg-10 is your default then all the top-level commands >>> are pg-10, others are pg-X.command. >>> >>> >> A solution for this that could work today is to have some sort of env >> setup (like done with python virtual envs), source the correct env and >> with aliases or PATH setups make sure the right top level names are seen. >> >> Here's a friday parlor trick >> >> source $(pg.env) >> >> And `pg.env` would just echo the path to the file withing the snap to >> source. Not totally user friendly and probably could be made more >> straightforward but it does get the job done. Maybe one day we can get a >> `snap env ` command (thinking out loud/writing), who knows :-) >> >> > That would only work if the overlapping command name are from the same > upstream project (postgresql in this example) but what if they are from > different upstreams? For instance, some other database (let's say > mariadb for sake of example) might also have a createdb command. Then > you might page postgresql-9-6.createdb, postgresql-10.createdb and > mariadb.createdb. I don't understand how this relates to my response. Can you clarify? I am not even covering a case of the same upstream project but the same snap which could be made up of many upstream projects. From sergio.schvezov at canonical.com Fri Sep 2 17:26:59 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Fri, 2 Sep 2016 14:26:59 -0300 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: <27768989-e4d3-840d-f11d-e8616e110286@ubuntu.com> References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> <27768989-e4d3-840d-f11d-e8616e110286@ubuntu.com> Message-ID: El 02/09/16 a las 14:15, Michael Hall escribió: > On 09/02/2016 01:06 PM, Evan Dandrea wrote: >> On Fri, 2 Sep 2016 at 04:30 Mark Shuttleworth > > wrote: >> >> Remember they can also build postgres INTO their own snap (and it >> actually might be worth exporting a shared pg part so people can do >> that trivially). >> >> >> And this is exactly what happened with Mongo. Gustavo wrote a Mongo >> snap. The Rocket.Chat guys picked it up and used it as a part. Then we >> pushed it to the parts server (`snapcraft search mongodb`) so others can >> benefit. >> >> +1 to a postgres part. >> >> > Yet another option would be to use the new content interface to build > postgresql snap that can share it's binaries and libraries into your > snap, so you can call them to setup and manage a local database in your > own application's directories. Yes, this is indeed an option for snaps from the same publisher. But you are now discussing inter snap communication and while the content interface seems good in the beginning, some things might be better served by using an interface specific to databases (as this is the topic at hand). From sergio.schvezov at canonical.com Fri Sep 2 17:35:23 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Fri, 2 Sep 2016 14:35:23 -0300 Subject: snapcraft's python plugins, road ahead and changes. Message-ID: Hi there, after seeing many problems with the current python plugins I decided to overhaul them. It has basically been re-written, fixing many bugs and annoyances. Among those nice things these changes provide are: smaller resulting snaps, easier iteration, wheel support and a clear separation of packages brought in through pip and those provided by a `stage-packages` entry. I invite any interested party to have a look at it here: https://github.com/snapcore/snapcraft/pull/771 It is in my interest to make this part of 2.17 (which will start its SRU process next week). With all this glory comes a problem, given the separation of THE python runtime and its stage-packages from whatever else you install iif you are a heavy fileset user or use command paths to entry points created from an install you will run into problems (a simple edit solves this). That said I am keen to hear from people using the python plugins, and to get a response from them if this is a better take on it. I am also interested to know if aside from fileset (these keywords in parts: `filesets`, `organize`, `stage` and `snap`) and `command` adaptations to whatever new path is used if there are any other blockers you might see with this branch. If the previous incantations of the python plugins did not work for you and this one doesn't either, then please also speak up (in the PR would be better) and maybe we can craft a fix. Cheers Sergio From ericoporto2008 at gmail.com Fri Sep 2 17:58:16 2016 From: ericoporto2008 at gmail.com (=?UTF-8?B?w4lyaWNvIFA=?=) Date: Fri, 2 Sep 2016 14:58:16 -0300 Subject: snapcraft's python plugins, road ahead and changes. In-Reply-To: References: Message-ID: Hey Sergio, I have tried to use the python plugin before but had some trouble relating to encodings. Posted here a question http://askubuntu.com/questions/783758/self-built-snap-run-fail-on-locale-error What do I need to do to use your version of the python plugin? Em 2 de set de 2016 2:36 PM, "Sergio Schvezov" < sergio.schvezov at canonical.com> escreveu: > Hi there, after seeing many problems with the current python plugins I > decided to overhaul them. It has basically been re-written, fixing many > bugs and annoyances. Among those nice things these changes provide are: > smaller resulting snaps, easier iteration, wheel support and a clear > separation of packages brought in through pip and those provided by a > `stage-packages` entry. > > I invite any interested party to have a look at it here: > > https://github.com/snapcore/snapcraft/pull/771 > > It is in my interest to make this part of 2.17 (which will start its SRU > process next week). > > With all this glory comes a problem, given the separation of THE python > runtime and its stage-packages from whatever else you install iif you are a > heavy fileset user or use command paths to entry points created from an > install you will run into problems (a simple edit solves this). > > That said I am keen to hear from people using the python plugins, and to > get a response from them if this is a better take on it. > > I am also interested to know if aside from fileset (these keywords in > parts: `filesets`, `organize`, `stage` and `snap`) and `command` > adaptations to whatever new path is used if there are any other blockers > you might see with this branch. > > If the previous incantations of the python plugins did not work for you > and this one doesn't either, then please also speak up (in the PR would be > better) and maybe we can craft a fix. > > Cheers > > Sergio > > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Fri Sep 2 18:39:24 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Fri, 2 Sep 2016 15:39:24 -0300 Subject: snapcraft's python plugins, road ahead and changes. In-Reply-To: References: Message-ID: El 02/09/16 a las 14:58, Érico P escribió: > > Hey Sergio, > > I have tried to use the python plugin before but had some trouble > relating to encodings. Posted here a question > > http://askubuntu.com/questions/783758/self-built-snap-run-fail-on-locale-error > I addressed that on askubuntu. > What do I need to do to use your version of the python plugin? > Here are a set of instructions given you have snapcraft installed on your system: git clone https://github.com/sergiusens/snapcraft.git cd snapcraft git checkout python-plugin-improvements alias snapcraft=$(pwd)/bin/snapcraft cd snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From leo.arias at canonical.com Fri Sep 2 20:01:03 2016 From: leo.arias at canonical.com (Leo Arias) Date: Fri, 2 Sep 2016 14:01:03 -0600 Subject: Let's kill "sideloading" In-Reply-To: <1472828350.5902.24.camel@ubuntu.com> References: <1472800069.5902.11.camel@ubuntu.com> <1472828350.5902.24.camel@ubuntu.com> Message-ID: <57C9DA7F.2070109@canonical.com> On 2016-09-02 08:59, Oliver Grawert wrote: > admittedly "kids" was a bit harsh, i apologize to anyone who feels > offended by this ... maybe you just meant that you are old ;) -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From mwinter at opensourcerouting.org Fri Sep 2 22:35:16 2016 From: mwinter at opensourcerouting.org (Martin Winter) Date: Fri, 02 Sep 2016 15:35:16 -0700 Subject: copy -> dump plugin migration Message-ID: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> So I’ve updated the quagga snap from using the copy plugin to the dump plugin. I’ve got it working, but it looks like a horrible hack using the dump plugin to copy files. I use it to copy default configs to the correct location. This is what I had before using the copy plugin: […] quagga-defaults: plugin: copy source: defaults files: "*.conf.default": /etc/quagga/ […] and this is what I use now with the dump plugin: […] quagga-defaults: plugin: dump source: defaults organize: zebra.conf.default: etc/quagga/zebra.conf.default bgpd.conf.default: etc/quagga/bgpd.conf.default isisd.conf.default: etc/quagga/isisd.conf.default ospf6d.conf.default: etc/quagga/ospf6d.conf.default ospfd.conf.default: etc/quagga/ospfd.conf.default pimd.conf.default: etc/quagga/pimd.conf.default ripd.conf.default: etc/quagga/ripd.conf.default ripngd.conf.default: etc/quagga/ripngd.conf.default […] Observations: I want to copy all files from my “defaults” directory (in the snapcraft build dir) to $SNAP/etc/quagga. The dump plugin doesn’t seem to support a destination, except with the “organize” function. And the “organize” function doesn’t support wildcards. This looks to me like bad code, but I can’t find a way to make this cleaner (better doc for the dump plugin would be highly appreciated) Maybe someone can comment on a better way? (It DOES WORK as shown, just trying to get feedback on a better/cleaner way, preferably with wildcard matching) - Martin Winter From gustavo.niemeyer at canonical.com Fri Sep 2 23:19:13 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Fri, 2 Sep 2016 20:19:13 -0300 Subject: copy -> dump plugin migration In-Reply-To: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> References: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> Message-ID: Hey Martin, Just put your files under defaults/etc/quagga, and use: defaults: plugin: dump source: defaults On Sep 2, 2016 7:36 PM, "Martin Winter" wrote: > So I’ve updated the quagga snap from using the copy plugin to the dump > plugin. > I’ve got it working, but it looks like a horrible hack using the dump > plugin > to copy files. > I use it to copy default configs to the correct location. > > This is what I had before using the copy plugin: > > […] > quagga-defaults: > plugin: copy > source: defaults > files: > "*.conf.default": /etc/quagga/ > […] > > and this is what I use now with the dump plugin: > > […] > quagga-defaults: > plugin: dump > source: defaults > organize: > zebra.conf.default: etc/quagga/zebra.conf.default > bgpd.conf.default: etc/quagga/bgpd.conf.default > isisd.conf.default: etc/quagga/isisd.conf.default > ospf6d.conf.default: etc/quagga/ospf6d.conf.default > ospfd.conf.default: etc/quagga/ospfd.conf.default > pimd.conf.default: etc/quagga/pimd.conf.default > ripd.conf.default: etc/quagga/ripd.conf.default > ripngd.conf.default: etc/quagga/ripngd.conf.default > […] > > Observations: > I want to copy all files from my “defaults” directory (in the snapcraft > build dir) > to $SNAP/etc/quagga. > The dump plugin doesn’t seem to support a destination, except with the > “organize” > function. And the “organize” function doesn’t support wildcards. > > This looks to me like bad code, but I can’t find a way to make this > cleaner (better > doc for the dump plugin would be highly appreciated) > > Maybe someone can comment on a better way? (It DOES WORK as shown, just > trying to > get feedback on a better/cleaner way, preferably with wildcard matching) > > - Martin Winter > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From leo.arias at canonical.com Sat Sep 3 00:06:11 2016 From: leo.arias at canonical.com (Leo Arias) Date: Fri, 2 Sep 2016 18:06:11 -0600 Subject: copy -> dump plugin migration In-Reply-To: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> References: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> Message-ID: <57CA13F3.2060707@canonical.com> Hello Martin, We might have been a little too hasty when we decided to print that deprecation message, I'm sorry about that. Take a look at https://bugs.launchpad.net/snapcraft/+bug/1616459 We are currently closing the final details to release 2.16, so this bug is scheduled to be resolved next week. On 2016-09-02 16:35, Martin Winter wrote: > This looks to me like bad code, but I can’t find a way to make this > cleaner (better > doc for the dump plugin would be highly appreciated) One thing I've requested is a small guide explaining how to migrate from copy to dump. Please leave any comments or suggestions you might have in the bug linked above. pura vida. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From mwinter at opensourcerouting.org Sat Sep 3 00:17:03 2016 From: mwinter at opensourcerouting.org (Martin Winter) Date: Fri, 02 Sep 2016 17:17:03 -0700 Subject: copy -> dump plugin migration In-Reply-To: References: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> Message-ID: Gustavo, On 2 Sep 2016, at 16:19, Gustavo Niemeyer wrote: > Hey Martin, > > Just put your files under defaults/etc/quagga, and use: > > defaults: > plugin: dump > source: defaults That’s one way to do it - but just creating the extra subdirectories seems to be non-optimal as well. It looks like the other email from Leo points to the exact issue I had when I tried to use the wildcards. So I may just wait for this to be resolved and keep the complete list of the files in the meantime. - Martin > > On Sep 2, 2016 7:36 PM, "Martin Winter" > > wrote: > >> So I’ve updated the quagga snap from using the copy plugin to the >> dump >> plugin. >> I’ve got it working, but it looks like a horrible hack using the >> dump >> plugin >> to copy files. >> I use it to copy default configs to the correct location. >> >> This is what I had before using the copy plugin: >> >> […] >> quagga-defaults: >> plugin: copy >> source: defaults >> files: >> "*.conf.default": /etc/quagga/ >> […] >> >> and this is what I use now with the dump plugin: >> >> […] >> quagga-defaults: >> plugin: dump >> source: defaults >> organize: >> zebra.conf.default: etc/quagga/zebra.conf.default >> bgpd.conf.default: etc/quagga/bgpd.conf.default >> isisd.conf.default: etc/quagga/isisd.conf.default >> ospf6d.conf.default: etc/quagga/ospf6d.conf.default >> ospfd.conf.default: etc/quagga/ospfd.conf.default >> pimd.conf.default: etc/quagga/pimd.conf.default >> ripd.conf.default: etc/quagga/ripd.conf.default >> ripngd.conf.default: etc/quagga/ripngd.conf.default >> […] >> >> Observations: >> I want to copy all files from my “defaults” directory (in the >> snapcraft >> build dir) >> to $SNAP/etc/quagga. >> The dump plugin doesn’t seem to support a destination, except with >> the >> “organize” >> function. And the “organize” function doesn’t support >> wildcards. >> >> This looks to me like bad code, but I can’t find a way to make this >> cleaner (better >> doc for the dump plugin would be highly appreciated) >> >> Maybe someone can comment on a better way? (It DOES WORK as shown, >> just >> trying to >> get feedback on a better/cleaner way, preferably with wildcard >> matching) >> >> - Martin Winter >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> From mwinter at opensourcerouting.org Sat Sep 3 00:18:41 2016 From: mwinter at opensourcerouting.org (Martin Winter) Date: Fri, 02 Sep 2016 17:18:41 -0700 Subject: copy -> dump plugin migration In-Reply-To: <57CA13F3.2060707@canonical.com> References: <2186060E-CF67-416D-B4C7-711C5F99DB2E@opensourcerouting.org> <57CA13F3.2060707@canonical.com> Message-ID: Leo, On 2 Sep 2016, at 17:06, Leo Arias wrote: > Hello Martin, > > We might have been a little too hasty when we decided to print that > deprecation message, I'm sorry about that. > > Take a look at https://bugs.launchpad.net/snapcraft/+bug/1616459 Yes, exactly the issue I ran into when I’ve tried the wildcard with the dump plugin. > We are currently closing the final details to release 2.16, so this bug > is scheduled to be resolved next week. Cool. As I have it currently working (just not so clean), I’ll just wait for this to be resolved and will update it again afterwards. - Martin > On 2016-09-02 16:35, Martin Winter wrote: >> This looks to me like bad code, but I can’t find a way to make this >> cleaner (better >> doc for the dump plugin would be highly appreciated) > > One thing I've requested is a small guide explaining how to migrate from > copy to dump. Please leave any comments or suggestions you might have in > the bug linked above. > > pura vida. > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 842 bytes Desc: OpenPGP digital signature URL: From mabnhdev at gmail.com Sat Sep 3 14:07:48 2016 From: mabnhdev at gmail.com (MikeB) Date: Sat, 3 Sep 2016 10:07:48 -0400 Subject: snapcraft's python plugins, road ahead and changes. Message-ID: Don't forget that at one point several folks thought it would be a good idea to move the location of custom plugins out of the parts directory - that's an awkward place for them. Can't tell you how many times I've accidentally deleted the parts directory only to remember - too late - I had some changes in the custom plugins. Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From casey.marshall at canonical.com Sat Sep 3 17:45:54 2016 From: casey.marshall at canonical.com (Casey Marshall) Date: Sat, 3 Sep 2016 12:45:54 -0500 Subject: Interface for accessing /dev/net/tun Message-ID: I'm trying to write a snap for OpenVPN with some wrapper scripts that automate PKI setup, IP masq, etc. My goal is to be able to setup an OpenVPN server with a single command, and then issue new clients with a single command. I'm using strict mode, because that's how I roll. I've added the firewall-control and network-control plugs, and that allowed some things to work (like setting up IP masq, etc.). However openvpn cannot start up, it's getting denied from opening /dev/net/tap, https://github.com/cmars/easy-openvpn-pkg/blob/master/TODO.md#tuntap-access-for-openvpn-in-strict-mode Would it be reasonable to contribute a patch to snapd adding the tun and/or tap devices to the network-control interface? Or would these constitute a new interface? -Casey -------------- next part -------------- An HTML attachment was scrubbed... URL: From ericoporto2008 at gmail.com Sat Sep 3 20:32:25 2016 From: ericoporto2008 at gmail.com (=?UTF-8?B?w4lyaWNvIFA=?=) Date: Sat, 3 Sep 2016 17:32:25 -0300 Subject: Nvidia related bug with snap In-Reply-To: References: Message-ID: Hello, I had crash trying to run the Krita snap https://bugs.kde.org/show_bug.cgi?id=368198 I do have a Nvidia GPU, but using prime, I had selected the Intel GPU. Are graphical snaps expected to work with Ubuntu if the Nvidia GPU is disabled? I don't know if here, IRC, AskUbuntu or Launchpad is the proper place to ask this question. Regards, Érico -------------- next part -------------- An HTML attachment was scrubbed... URL: From zygmunt.krynicki at canonical.com Sat Sep 3 21:27:10 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Sat, 3 Sep 2016 23:27:10 +0200 Subject: Nvidia related bug with snap In-Reply-To: References: Message-ID: > Wiadomość napisana przez Érico P w dniu 03.09.2016, o godz. 22:32: > > Hello, > > I had crash trying to run the Krita snap > > https://bugs.kde.org/show_bug.cgi?id=368198 > > I do have a Nvidia GPU, but using prime, I had selected the Intel GPU. > > Are graphical snaps expected to work with Ubuntu if the Nvidia GPU is disabled? > > I don't know if here, IRC, AskUbuntu or Launchpad is the proper place to ask this question. > Hey. Can you share some more details please? - Which distribution are you using - Which version of the driver are you using (if you don’t know, which GPU exactly do you have) - How did the crash look like - Which revision of ubuntu-core and krita did you have (hint: snap list) Best regards ZK From ericoporto2008 at gmail.com Sat Sep 3 21:41:12 2016 From: ericoporto2008 at gmail.com (=?UTF-8?B?w4lyaWNvIFA=?=) Date: Sat, 3 Sep 2016 18:41:12 -0300 Subject: Nvidia related bug with snap In-Reply-To: References: Message-ID: Hi Zygmunt, I will assume I can paste the steps to crash here: user at host:~$ sudo snap install --beta krita 98.17 MB / 98.17 MB [======================================] 100.00 % 1.70 MB/s krita (beta) 3.0.1-beta-snap13 from 'krita' installed * user at host:~$ krita * * krita.lib.pigment: Legacy integer arithmetics implementation* * XmbTextListToTextProperty result code -2* * QOpenGLFunctions created with non-current context* * Segmentation fault (core dumped)* Additional info: user at host:~$ lsb_release -a LSB Version: core-9.20160110ubuntu0.2-amd64:core-9.20160110ubuntu0.2-noarch:printing-9.20160110ubuntu0.2-amd64:printing-9.20160110ubuntu0.2-noarch:security-9.20160110ubuntu0.2-amd64:security-9.20160110ubuntu0.2-noarch Distributor ID: Ubuntu Description: Ubuntu 16.04.1 LTS Release: 16.04 Codename: xenial user at host:~$ snap --version snap 2.13+ppa207-1 snapd 2.13+ppa207-1 series 16 ubuntu 16.04 user at host:~$ dpkg -l | grep nvidia ii nvidia-364 364.19-0ubuntu0~gpu16.04.6 amd64 NVIDIA binary driver - version 364.19 ii nvidia-opencl-icd-364 364.19-0ubuntu0~gpu16.04.6 amd64 NVIDIA OpenCL ICD ii nvidia-prime 0.8.2 amd64 Tools to enable NVIDIA's Prime ii nvidia-settings 370.23-0ubuntu0~gpu16.04.1 amd64 Tool for configuring the NVIDIA graphics user at host:~$ snap list Name Version Rev Developer Notes krita 3.0.1-beta-snap13 5 krita - ubuntu-core 16.04.1 352 canonical - Also my GPU is Nvidia GT555M, but I am actually using the Intel GPU. My PC uses Intel® Core™ i7-2670QM CPU @ 2.20GHz, 6GBRam and it has the ubuntu installed is 64bits. I don't know how to show this information from terminal command though. Érico V. Porto On Sat, Sep 3, 2016 at 6:27 PM, Zygmunt Krynicki < zygmunt.krynicki at canonical.com> wrote: > > > Wiadomość napisana przez Érico P w dniu > 03.09.2016, o godz. 22:32: > > > > Hello, > > > > I had crash trying to run the Krita snap > > > > https://bugs.kde.org/show_bug.cgi?id=368198 > > > > I do have a Nvidia GPU, but using prime, I had selected the Intel GPU. > > > > Are graphical snaps expected to work with Ubuntu if the Nvidia GPU is > disabled? > > > > I don't know if here, IRC, AskUbuntu or Launchpad is the proper place to > ask this question. > > > > Hey. > > Can you share some more details please? > > - Which distribution are you using > - Which version of the driver are you using (if you don’t know, which GPU > exactly do you have) > - How did the crash look like > - Which revision of ubuntu-core and krita did you have (hint: snap list) > > Best regards > ZK > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.wakeling at webdrake.net Sat Sep 3 23:42:52 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Sun, 4 Sep 2016 01:42:52 +0200 Subject: DEPRECATED: 'license' defined in snapcraft.yaml Message-ID: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> Just got the above error message as I was finishing up my LDC snap. What's the recommended way to include license information? I ask because the 'snapcraft syntax' page on snapcraft.io lists the `licence` attribute without any mention of deprecation or an alternative for it: http://snapcraft.io/docs/build-snaps/syntax Apologies if the question has been asked and answered before, but I didn't find anything obvious in the mailing list archives, so I thought I would ask the question very obviously in order to correct that ;-) Thanks & best wishes, -- Joe From joseph.wakeling at webdrake.net Sun Sep 4 11:55:49 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Sun, 4 Sep 2016 13:55:49 +0200 Subject: Working LDC snap [was: Re: Snapping LDC (LLVM-based D compiler)] In-Reply-To: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> Message-ID: <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> On 27/08/16 22:45, Joseph Rushton Wakeling wrote: > I thought I'd have a go at making a snap of LDC, the LLVM-based compiler for the > D programming language. A "final first draft" of the working snap is available here: https://github.com/WebDrake/ldc2.snap/pull/1 Obviously everyone here has already been very generous with their advice and feedback, but I thought it would be good to offer an opportunity for final comments before finishing this up. Thoughts as ever gratefully received :-) Thanks again to everyone & best wishes, -- Joe From mark at ubuntu.com Sun Sep 4 12:00:41 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 4 Sep 2016 08:00:41 -0400 Subject: snapcraft's python plugins, road ahead and changes. In-Reply-To: References: Message-ID: On 03/09/16 10:07, MikeB wrote: > Don't forget that at one point several folks thought it would be a > good idea to move the location of custom plugins out of the parts > directory - that's an awkward place for them. Can't tell you how many > times I've accidentally deleted the parts directory only to remember - > too late - I had some changes in the custom plugins. I'll second this. IIRC we wanted to use the parts directory to reduce the footprint of snapcraft in an upstream tree, but I think that it would be better to treat parts/ as entirely removable. That would suggest putting the plugin some where like .snapcraft/ Mark -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Sun Sep 4 12:22:02 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 4 Sep 2016 08:22:02 -0400 Subject: Working LDC snap [was: Re: Snapping LDC (LLVM-based D compiler)] In-Reply-To: <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> Message-ID: <3bd1534d-ade1-033a-3dec-00f2c56e7e17@ubuntu.com> On 04/09/16 07:55, Joseph Rushton Wakeling wrote: > On 27/08/16 22:45, Joseph Rushton Wakeling wrote: >> I thought I'd have a go at making a snap of LDC, the LLVM-based >> compiler for the >> D programming language. > > A "final first draft" of the working snap is available here: > https://github.com/WebDrake/ldc2.snap/pull/1 > > Obviously everyone here has already been very generous with their > advice and feedback, but I thought it would be good to offer an > opportunity for final comments before finishing this up. Thoughts as > ever gratefully received :-) > > Thanks again to everyone & best wishes, YW :) I'll try yours now. Would also appreciate some feedback on my holiday project :) https://github.com/markshuttle/rethinkdb-snap Mark From michael.hudson at canonical.com Sun Sep 4 22:09:50 2016 From: michael.hudson at canonical.com (Michael Hudson-Doyle) Date: Mon, 5 Sep 2016 10:09:50 +1200 Subject: Snappy takes long time to boot when no network In-Reply-To: <84e58e9b-2070-95e1-70a3-01b81548c8da@parrot.com> References: <8042a3a2-721d-8ee8-5fb0-0dae3cb9bc7e@parrot.com> <1472730470.15344.14.camel@ubuntu.com> <1472733028.15344.29.camel@ubuntu.com> <84e58e9b-2070-95e1-70a3-01b81548c8da@parrot.com> Message-ID: On 3 September 2016 at 00:46, Yann Sionneau wrote: > Your solution fixed my issue : > > root at Paros:~# rm /etc/netplan/00-initial-config.yaml > root at Paros:~# netplan apply > Cannot replug rndis0: [Errno 2] No such file or directory: > '/sys/devices/platform/tegra-xudc/gadget/driver/unbind' > > and then reboot. I now boot in 24 seconds instead of 140 seconds :) > > Thanks! > Glad to hear it, and sorry for the trouble. Cheers, mwh > Le 09/01/2016 à 11:19 PM, Michael Hudson-Doyle a écrit : > > > > On 2 September 2016 at 00:44, Yann Sionneau > wrote: > >> I'm not sure I understand everything in the bug ticket as I am not a >> systemd / networkd / netplan expert at all. >> >> But : >> >> root at Paros:~# cat /etc/netplan/00- >> 00-initial-config.yaml 00-snapd-config.yaml >> root at Paros:~# cat /etc/netplan/00-initial-config.yaml >> >> network: >> version: 2 >> ethernets: >> all: >> match: >> name: "*" >> dhcp4: true >> root at Paros:~# cat /run/systemd/network/10-netplan-all.network >> [Match] >> Name=* >> >> [Network] >> DHCP=ipv4 >> root at Paros:~# networkctl >> IDX LINK TYPE OPERATIONAL SETUP >> 1 lo loopback carrier configured >> 2 sit0 sit routable configuring >> 3 eth1 ether routable configured >> 4 eth0 ether routable configured >> 5 rndis0 ether no-carrier configuring >> >> 5 links listed. >> >> Does this help? >> >> eth0 is a WiFi interface. >> >> eth1 is USB ethernet device (plugged at boot) >> >> > I don't know if it's the cause of all your issues, but having both files > in /etc/netplan is a sign of console-conf / snapd version skew. It's fixed > now and new installs won't have this problem, but if you don't want to > re-install just deleted the 00-initial-config.yaml file and run sudo > netplan apply. > > Cheers, > mwh > > >> Le 09/01/2016 à 02:30 PM, Oliver Grawert a écrit : >> > hi, >> > On Do, 2016-09-01 at 14:05 +0200, Yann Sionneau wrote: >> >> I feel like this is the bug I'm hitting. >> >> But, how do you explain that my boot is stalled even if I have 2 NICs >> >> with internet access? >> >> One is via wifi, the other is via usb-ethernet. >> >> Thanks! >> > hmm, this sounds more like a different bug ... >> > how about: >> > >> > https://launchpad.net/bugs/1618522 >> > >> > ciao >> > oli >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.wakeling at webdrake.net Sun Sep 4 22:25:42 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Mon, 5 Sep 2016 00:25:42 +0200 Subject: Working LDC snap [was: Re: Snapping LDC (LLVM-based D compiler)] In-Reply-To: <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> Message-ID: <17152bf2-210c-f0d3-6c69-5cb0e9bf0300@webdrake.net> On 04/09/16 13:55, Joseph Rushton Wakeling wrote: > On 27/08/16 22:45, Joseph Rushton Wakeling wrote: >> I thought I'd have a go at making a snap of LDC, the LLVM-based compiler for the >> D programming language. > > A "final first draft" of the working snap is available here: > https://github.com/WebDrake/ldc2.snap/pull/1 One problem fixed (the inclusion of `build-essential` as a build dependency ensures the snap will build in a `cleanbuild` environment), but one remains: it looks like the `ldc-config` part (which manually replaces the incorrectly auto-generated ldc2.conf with a correct alternative, using the `dump` plugin) can sometimes fail. There doesn't seem to be any reason for it, just with some builds I end up with a wrong (auto-generated) ldc2.conf in the snap package, and some I end up with the correct one copied from the ldc-config directory. Any ideas what's up? I'm guessing bug either with the dump plugin or with the handling of the `snap` section of the `ldc` part ... ? From christian.ehrhardt at canonical.com Mon Sep 5 06:14:22 2016 From: christian.ehrhardt at canonical.com (Christian Ehrhardt) Date: Mon, 5 Sep 2016 08:14:22 +0200 Subject: PostgreSQL 9.6 snap and feedback In-Reply-To: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> References: <2c9974b9-b116-57ba-ac79-cb3933d917b1@ubuntu.com> Message-ID: On Fri, Sep 2, 2016 at 1:29 PM, Mark Shuttleworth wrote: > > > > I have lots and lots and lots of man pages not on my MANPATH. Not > > urgent, but nice to expose them. > > > > Noted and I think in-progress. > FYI - https://bugs.launchpad.net/snappy/+bug/1575593 And just as in your case it is still in kind of a "Not urgent, but nice" state. If we realize that this is really a more common need, it might be worth to chime in and bump priority for anybody affected. -- Christian Ehrhardt Software Engineer, Ubuntu Server Canonical Ltd -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie.bennett at canonical.com Mon Sep 5 10:51:33 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Mon, 5 Sep 2016 11:51:33 +0100 Subject: Interface for accessing /dev/net/tun In-Reply-To: References: Message-ID: <20160905105133.GA4443@phoenix> On 03/09/16 at 12:45pm, Casey Marshall wrote: > I'm trying to write a snap for OpenVPN with some wrapper scripts that automate > PKI setup, IP masq, etc. My goal is to be able to setup an OpenVPN server with > a single command, and then issue new clients with a single command. I'm using > strict mode, because that's how I roll. > > I've added the firewall-control and network-control plugs, and that allowed > some things to work (like setting up IP masq, etc.). However openvpn cannot > start up, it's getting denied from opening /dev/net/tap, https://github.com/ > cmars/easy-openvpn-pkg/blob/master/TODO.md# > tuntap-access-for-openvpn-in-strict-mode > > Would it be reasonable to contribute a patch to snapd adding the tun and/or tap > devices to the network-control interface? Or would these constitute a new > interface? As this interface is just a network device, albeit for user space programs, it makes sense for this to be part of the network-control interface IMHO. > -Casey Regards, Jamie. > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From jamie.bennett at canonical.com Mon Sep 5 11:03:49 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Mon, 5 Sep 2016 12:03:49 +0100 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> Message-ID: <20160905110349.GB4443@phoenix> On 04/09/16 at 01:42am, Joseph Rushton Wakeling wrote: > Just got the above error message as I was finishing up my LDC snap. > > What's the recommended way to include license information? I ask because > the 'snapcraft syntax' page on snapcraft.io lists the `licence` attribute > without any mention of deprecation or an alternative for it: > http://snapcraft.io/docs/build-snaps/syntax > > Apologies if the question has been asked and answered before, but I didn't > find anything obvious in the mailing list archives, so I thought I would ask > the question very obviously in order to correct that ;-) License files can be copied across if they are present in the setup/ directory. Reproduced from docs/meta.md " ## Fixed assets Some metadata is provided in the form of conventions, such as license files, icons and desktop files among others. For these fixed files to make it into your final snap they need to be in a `setup` directory at the same level of your `snapcraft.yaml`. " Of course the website needs updating to reflect the new syntax (and the fact that icon e.t.c also uses the setup/ dir now). > Thanks & best wishes, > > -- Joe Regards, Jamie. > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From david.calle at canonical.com Mon Sep 5 11:08:53 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Mon, 5 Sep 2016 13:08:53 +0200 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <20160905110349.GB4443@phoenix> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> Message-ID: <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> On 05/09/2016 13:03, Jamie Bennett wrote: > On 04/09/16 at 01:42am, Joseph Rushton Wakeling wrote: >> Just got the above error message as I was finishing up my LDC snap. >> >> What's the recommended way to include license information? I ask because >> the 'snapcraft syntax' page on snapcraft.io lists the `licence` attribute >> without any mention of deprecation or an alternative for it: >> http://snapcraft.io/docs/build-snaps/syntax >> >> Apologies if the question has been asked and answered before, but I didn't >> find anything obvious in the mailing list archives, so I thought I would ask >> the question very obviously in order to correct that ;-) > License files can be copied across if they are present in the setup/ > directory. Reproduced from docs/meta.md > > " > ## Fixed assets > > Some metadata is provided in the form of conventions, such as license files, > icons and desktop files among others. For these fixed files to make it into > your final snap they need to be in a `setup` directory at the same level of > your `snapcraft.yaml`. > " > > Of course the website needs updating to reflect the new syntax (and the fact > that icon e.t.c also uses the setup/ dir now). You can find these details in the docs here: http://snapcraft.io/docs/build-snaps/metadata#fixed-assets Cheers, David > >> Thanks & best wishes, >> >> -- Joe > Regards, > Jamie. > >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From stuart.bishop at canonical.com Mon Sep 5 12:06:37 2016 From: stuart.bishop at canonical.com (Stuart Bishop) Date: Mon, 5 Sep 2016 19:06:37 +0700 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> Message-ID: On 5 September 2016 at 18:08, David Callé wrote: > > > License files can be copied across if they are present in the setup/ > > directory. Reproduced from docs/meta.md > > > > " > > ## Fixed assets > > > > Some metadata is provided in the form of conventions, such as license > files, > > icons and desktop files among others. For these fixed files to make it > into > > your final snap they need to be in a `setup` directory at the same level > of > > your `snapcraft.yaml`. > > " > > > > Of course the website needs updating to reflect the new syntax (and the > fact > > that icon e.t.c also uses the setup/ dir now). > > You can find these details in the docs here: > http://snapcraft.io/docs/build-snaps/metadata#fixed-assets This does not mention the magic name(s) needed for license (nee licence, copyright, legal, copying etc.) files to be discovered and made available from the snap metadata by installers, the snap store or other consumers. Which is what we are really asking. -- Stuart Bishop -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie.bennett at canonical.com Mon Sep 5 12:31:53 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Mon, 5 Sep 2016 13:31:53 +0100 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> Message-ID: <20160905123153.GE4443@phoenix> On 05/09/16 at 07:06pm, Stuart Bishop wrote: > On 5 September 2016 at 18:08, David Callé wrote: > > > > License files can be copied across if they are present in the setup/ > > directory. Reproduced from docs/meta.md > > > > " > > ## Fixed assets > > > > Some metadata is provided in the form of conventions, such as license > files, > > icons and desktop files among others. For these fixed files to make it > into > > your final snap they need to be in a `setup` directory at the same level > of > > your `snapcraft.yaml`. > > " > > > > Of course the website needs updating to reflect the new syntax (and the > fact > > that icon e.t.c also uses the setup/ dir now). > > You can find these details in the docs here: > http://snapcraft.io/docs/build-snaps/metadata#fixed-assets > > > > This does not mention the magic name(s) needed for license (nee licence, > copyright, legal, copying etc.) files to be discovered and made available from > the snap metadata by installers, the snap store or other consumers. Which is > what we are really asking. Just put the license in the setup/ directory and name it license.txt. > -- > Stuart Bishop Regards, Jamie. > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From simon.fels at canonical.com Mon Sep 5 12:34:12 2016 From: simon.fels at canonical.com (Simon Fels) Date: Mon, 5 Sep 2016 13:34:12 +0100 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <20160905123153.GE4443@phoenix> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> <20160905123153.GE4443@phoenix> Message-ID: On 05.09.2016 13:31, Jamie Bennett wrote: > On 05/09/16 at 07:06pm, Stuart Bishop wrote: >> On 5 September 2016 at 18:08, David Callé wrote: >> >> >> > License files can be copied across if they are present in the setup/ >> > directory. Reproduced from docs/meta.md >> > >> > " >> > ## Fixed assets >> > >> > Some metadata is provided in the form of conventions, such as license >> files, >> > icons and desktop files among others. For these fixed files to make it >> into >> > your final snap they need to be in a `setup` directory at the same level >> of >> > your `snapcraft.yaml`. >> > " >> > >> > Of course the website needs updating to reflect the new syntax (and the >> fact >> > that icon e.t.c also uses the setup/ dir now). >> >> You can find these details in the docs here: >> http://snapcraft.io/docs/build-snaps/metadata#fixed-assets >> >> >> >> This does not mention the magic name(s) needed for license (nee licence, >> copyright, legal, copying etc.) files to be discovered and made available from >> the snap metadata by installers, the snap store or other consumers. Which is >> what we are really asking. > > Just put the license in the setup/ directory and name it license.txt. What does this actually trigger? Is it detected by the store? Or is it just shipped with the snap and then being available as part of $SNAP? regards, Simon From oriol.rius at gmail.com Mon Sep 5 14:04:44 2016 From: oriol.rius at gmail.com (Oriol Rius) Date: Mon, 05 Sep 2016 14:04:44 +0000 Subject: 4G on Dell IoT Gateway Message-ID: Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a problem trying to set up a 4G connection using nmcli and mmcli without success. Simple capabilities doesn't work and Dell support it's slow and insufficient. My first problem starts trying to validate the PIN. But in the end what I need is a simple tutorial or manual that help me how to set up the internet connection. I found some information reading mailing lists and googling, but everything is imprecise and incomplete. About the modem I use: # mmcli -m 0 /org/freedesktop/ModemManager1/Modem/0 (device id 'b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') ------------------------- Hardware | manufacturer: 'Telit' | model: 'HE910-D' | revision: '12.00.086' | supported: 'gsm-umts' | current: 'gsm-umts' | equipment id: '356136070745999' ------------------------- System | device: '/sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.1' | drivers: 'cdc_acm, cdc_ether' | plugin: 'Dell' | primary port: 'ttyACM3' | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 (at)' ------------------------- Numbers | own : 'unknown' ------------------------- Status | lock: 'unknown' | unlock retries: 'unknown' | state: 'failed' | failed reason: 'sim-missing' | power state: 'on' | access tech: 'unknown' | signal quality: '0' (cached) ------------------------- Modes | supported: 'allowed: 2g, 3g; preferred: none' | current: 'allowed: any; preferred: none' ------------------------- Bands | supported: 'unknown' | current: 'unknown' ------------------------- IP | supported: 'none' ------------------------- SIM | path: 'none' ------------------------- Bearers | paths: 'none' Can you recommend any resource that help me on that? Thank you very much. Regards. Oriol -------------- next part -------------- An HTML attachment was scrubbed... URL: From alfonso.sanchez-beato at canonical.com Mon Sep 5 14:30:24 2016 From: alfonso.sanchez-beato at canonical.com (Alfonso Sanchez-Beato) Date: Mon, 5 Sep 2016 16:30:24 +0200 Subject: 4G on Dell IoT Gateway In-Reply-To: References: Message-ID: Hi Oriol, On Mon, Sep 5, 2016 at 4:04 PM, Oriol Rius wrote: > Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a problem > trying to set up a 4G connection using nmcli and mmcli without success. > Simple capabilities doesn't work and Dell support it's slow and > insufficient. > > My first problem starts trying to validate the PIN. But in the end what I > need is a simple tutorial or manual that help me how to set up the internet > connection. I found some information reading mailing lists and googling, > but everything is imprecise and incomplete. About the modem I use: > First, note that the HE910 is a HSPA modem and does not support 4G. You can try to activate cellular data with these steps: 1. Deactivate SIM PIN with another device 2. Execute these commands: nmcli c add type gsm ifname ttyACM3 con-name gsmconn apn nmcli r wwan on As you see you need to use nmcli instead of mmcli because we are controlling the modem from NetworkManager so it can take control of the connection and configure routes. HTH, Alfonso > > # mmcli -m 0 > > /org/freedesktop/ModemManager1/Modem/0 (device id ' > b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') > ------------------------- > Hardware | manufacturer: 'Telit' > | model: 'HE910-D' > | revision: '12.00.086' > | supported: 'gsm-umts' > | current: 'gsm-umts' > | equipment id: '356136070745999' > ------------------------- > System | device: '/sys/devices/pci0000:00/0000: > 00:14.0/usb1/1-4/1-4.1' > | drivers: 'cdc_acm, cdc_ether' > | plugin: 'Dell' > | primary port: 'ttyACM3' > | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 (at)' > ------------------------- > Numbers | own : 'unknown' > ------------------------- > Status | lock: 'unknown' > | unlock retries: 'unknown' > | state: 'failed' > | failed reason: 'sim-missing' > | power state: 'on' > | access tech: 'unknown' > | signal quality: '0' (cached) > ------------------------- > Modes | supported: 'allowed: 2g, 3g; preferred: none' > | current: 'allowed: any; preferred: none' > ------------------------- > Bands | supported: 'unknown' > | current: 'unknown' > ------------------------- > IP | supported: 'none' > ------------------------- > SIM | path: 'none' > > ------------------------- > Bearers | paths: 'none' > > > Can you recommend any resource that help me on that? > > Thank you very much. > > Regards. > Oriol > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie.bennett at canonical.com Mon Sep 5 14:38:48 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Mon, 5 Sep 2016 15:38:48 +0100 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> <20160905123153.GE4443@phoenix> Message-ID: <20160905143848.GF4443@phoenix> On 05/09/16 at 01:34pm, Simon Fels wrote: > On 05.09.2016 13:31, Jamie Bennett wrote: > > On 05/09/16 at 07:06pm, Stuart Bishop wrote: > >> On 5 September 2016 at 18:08, David Callé wrote: > >> > >> > >> > License files can be copied across if they are present in the setup/ > >> > directory. Reproduced from docs/meta.md > >> > > >> > " > >> > ## Fixed assets > >> > > >> > Some metadata is provided in the form of conventions, such as license > >> files, > >> > icons and desktop files among others. For these fixed files to make it > >> into > >> > your final snap they need to be in a `setup` directory at the same level > >> of > >> > your `snapcraft.yaml`. > >> > " > >> > > >> > Of course the website needs updating to reflect the new syntax (and the > >> fact > >> > that icon e.t.c also uses the setup/ dir now). > >> > >> You can find these details in the docs here: > >> http://snapcraft.io/docs/build-snaps/metadata#fixed-assets > >> > >> > >> > >> This does not mention the magic name(s) needed for license (nee licence, > >> copyright, legal, copying etc.) files to be discovered and made available from > >> the snap metadata by installers, the snap store or other consumers. Which is > >> what we are really asking. > > > > Just put the license in the setup/ directory and name it license.txt. > > What does this actually trigger? Is it detected by the store? Or is it > just shipped with the snap and then being available as part of $SNAP? I don't know if it is used in the store, someone else with more knowledge will have to comment. > regards, > Simon Regards, Jamie. > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From ogra at ubuntu.com Mon Sep 5 15:20:44 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Mon, 05 Sep 2016 17:20:44 +0200 Subject: Let's kill "sideloading" In-Reply-To: <57C9DA7F.2070109@canonical.com> References: <1472800069.5902.11.camel@ubuntu.com> <1472828350.5902.24.camel@ubuntu.com> <57C9DA7F.2070109@canonical.com> Message-ID: <1473088844.5902.51.camel@ubuntu.com> hi, Am Freitag, den 02.09.2016, 14:01 -0600 schrieb Leo Arias: > On 2016-09-02 08:59, Oliver Grawert wrote: > > > > admittedly "kids" was a bit harsh, i apologize to anyone who feels > > offended by this ...  > maybe you just meant that you are old ;) > yeah, and i welcome everyone who felt offended to call me "stinky old fart" from now on :) ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From gustavo.niemeyer at canonical.com Mon Sep 5 15:54:21 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Mon, 5 Sep 2016 12:54:21 -0300 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <20160905143848.GF4443@phoenix> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> <20160905123153.GE4443@phoenix> <20160905143848.GF4443@phoenix> Message-ID: Hello all, There's no integrated license support in snaps yet. We've removed the original feature on the way to 16.04, and haven't had time to reintroduce it properly yet. It'll come soon. On Mon, Sep 5, 2016 at 11:38 AM, Jamie Bennett wrote: > On 05/09/16 at 01:34pm, Simon Fels wrote: > > On 05.09.2016 13:31, Jamie Bennett wrote: > > > On 05/09/16 at 07:06pm, Stuart Bishop wrote: > > >> On 5 September 2016 at 18:08, David Callé > wrote: > > >> > > >> > > >> > License files can be copied across if they are present in the > setup/ > > >> > directory. Reproduced from docs/meta.md > > >> > > > >> > " > > >> > ## Fixed assets > > >> > > > >> > Some metadata is provided in the form of conventions, such as > license > > >> files, > > >> > icons and desktop files among others. For these fixed files to > make it > > >> into > > >> > your final snap they need to be in a `setup` directory at the > same level > > >> of > > >> > your `snapcraft.yaml`. > > >> > " > > >> > > > >> > Of course the website needs updating to reflect the new syntax > (and the > > >> fact > > >> > that icon e.t.c also uses the setup/ dir now). > > >> > > >> You can find these details in the docs here: > > >> http://snapcraft.io/docs/build-snaps/metadata#fixed-assets > > >> > > >> > > >> > > >> This does not mention the magic name(s) needed for license (nee > licence, > > >> copyright, legal, copying etc.) files to be discovered and made > available from > > >> the snap metadata by installers, the snap store or other consumers. > Which is > > >> what we are really asking. > > > > > > Just put the license in the setup/ directory and name it license.txt. > > > > What does this actually trigger? Is it detected by the store? Or is it > > just shipped with the snap and then being available as part of $SNAP? > > I don't know if it is used in the store, someone else with more knowledge > will > have to comment. > > > regards, > > Simon > > Regards, > Jamie. > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Mon Sep 5 16:06:12 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Mon, 5 Sep 2016 13:06:12 -0300 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <20160905143848.GF4443@phoenix> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> <58f915a0-bc86-3402-a9c1-aeb6d08ae021@canonical.com> <20160905123153.GE4443@phoenix> <20160905143848.GF4443@phoenix> Message-ID: <6690b345-4ea8-f8e3-39c4-9ce85dbf5419@canonical.com> El 05/09/16 a las 11:38, Jamie Bennett escribió: > On 05/09/16 at 01:34pm, Simon Fels wrote: >> On 05.09.2016 13:31, Jamie Bennett wrote: >>> On 05/09/16 at 07:06pm, Stuart Bishop wrote: >>>> On 5 September 2016 at 18:08, David Callé wrote: >>>> >>>> >>>> > License files can be copied across if they are present in the setup/ >>>> > directory. Reproduced from docs/meta.md >>>> > >>>> > " >>>> > ## Fixed assets >>>> > >>>> > Some metadata is provided in the form of conventions, such as license >>>> files, >>>> > icons and desktop files among others. For these fixed files to make it >>>> into >>>> > your final snap they need to be in a `setup` directory at the same level >>>> of >>>> > your `snapcraft.yaml`. >>>> > " >>>> > >>>> > Of course the website needs updating to reflect the new syntax (and the >>>> fact >>>> > that icon e.t.c also uses the setup/ dir now). >>>> >>>> You can find these details in the docs here: >>>> http://snapcraft.io/docs/build-snaps/metadata#fixed-assets >>>> >>>> >>>> >>>> This does not mention the magic name(s) needed for license (nee licence, >>>> copyright, legal, copying etc.) files to be discovered and made available from >>>> the snap metadata by installers, the snap store or other consumers. Which is >>>> what we are really asking. >>> Just put the license in the setup/ directory and name it license.txt. >> What does this actually trigger? Is it detected by the store? Or is it >> just shipped with the snap and then being available as part of $SNAP? > I don't know if it is used in the store, someone else with more knowledge will > have to comment. Here's the history https://lists.ubuntu.com/archives/snappy-devel/2016-March/001596.html Here is where it should be documented: http://snapcraft.io/docs/build-snaps/metadata under fixed assets. That said, there are multiple parties wanting it to be by declaration instead of a fixed location (and I am in that camp, while going by path-convention is a good idea for the snap itself, it brought in multiple problems to the builders). Going by declaration would make it easy to just provide whatever asset from whatever sources are used (which is what the current deprecated entry points do). From simon.fels at canonical.com Mon Sep 5 16:21:19 2016 From: simon.fels at canonical.com (Simon Fels) Date: Mon, 5 Sep 2016 17:21:19 +0100 Subject: Snappy on OpenSwitch In-Reply-To: <55cdfee6-d6d2-22ea-5d1a-d5bfd9449dcf@canonical.com> References: <55cdfee6-d6d2-22ea-5d1a-d5bfd9449dcf@canonical.com> Message-ID: <592a97db-72f6-3638-8aab-49bbfe9c4b4b@canonical.com> Hey Mike, >> I'm very interested in adding Snappy support to OpenSwitch >> (www.openswitch.net ). >> >> OpenSwitch is a Network Operating System for network switches build upon >> OpenEmbedded/Yocto/Poky. > > Pretty nice idea! > >> I'm using a branch of OpenSwitch in which they've upgraded to Yocto 2.1 >> - https://github.com/open-switch/ops-build >> branch >> "feature/yocto_2.1". >> >> I've added meta-snappy to the bblayers.conf file used to build >> OpenSwitch and I was able to successfully 'bitbake snappy-demo-image' so >> that I at least know your meta-snappy layer is compatible with the >> OpenSwitch layers. >> >> My knowledge of OE/bitbake is limited and I'm hoping someone could >> provide some pointers or guidance to integrating Snappy onto an existing >> OE/Yocto distribution such as OpenSwitch. > > You already did the first step and verified the build of the relevant > components (snapd and snap-confine) works well in OpenSwitch. > > The next step is to get the different config options for the kernel into > your kernel defconfig. We have various sample kernel trees at [1] which > you can use as reference. In meta-snappy I extended the default > linux-yocto defconfig with the bits listed in [2]. I suspect OpenSwitch > may use a different kernel tree so you have to apply the different > options for it as well. You don't need any AppArmor related patches if > you want just unconfined snaps working for a first step. > > As next thing you have to add snapd and snap-confine to the content of > the OpenSwitch image. Adding snap-confine and snapd in [3] is ok to get > it things working. Later you want to add a packagegroup especially for > snappy which I am planing to add to meta-snappy which then will > automatically pull in all necessary dependencies for you. > > Once that is done you should be able to build a regular OpenSwitch > image, flash it on a device or boot it in an emulator and should have > snapd running and being able to install snaps and run their applications. > > If you need further help or have questions, just ask here or ping me > (nick morphis) on IRC in #snappy on FreeNode. You already had a chance to try my suggestions? regards, Simon From joseph.wakeling at webdrake.net Mon Sep 5 22:15:43 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Tue, 6 Sep 2016 00:15:43 +0200 Subject: (Not entirely) working LDC snap In-Reply-To: <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> Message-ID: On 04/09/16 13:55, Joseph Rushton Wakeling wrote: > On 27/08/16 22:45, Joseph Rushton Wakeling wrote: >> I thought I'd have a go at making a snap of LDC, the LLVM-based compiler for the >> D programming language. > > A "final first draft" of the working snap is available here: > https://github.com/WebDrake/ldc2.snap/pull/1 So, you'd think I'd have tested this before, given that I'm trying to snap a compiler, but ... :-P Turns out the snap-packaged LDC will fail if it's trying to compile a project that needs other libraries to be linked in. This isn't surprising, of course -- system libraries won't be on the library paths of the snap -- but it does put in place a BIG constraint on what can be built and what not. Any thoughts on how this might be addressed? Thanks & best wishes, -- Joe From joseph.wakeling at webdrake.net Mon Sep 5 22:19:08 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Tue, 6 Sep 2016 00:19:08 +0200 Subject: DEPRECATED: 'license' defined in snapcraft.yaml In-Reply-To: <20160905110349.GB4443@phoenix> References: <2c00c655-8768-40d8-7c85-fc599bfd18f8@webdrake.net> <20160905110349.GB4443@phoenix> Message-ID: <0e4cb2f6-0f2c-d091-8d0e-cdefd0d5a859@webdrake.net> On 05/09/16 13:03, Jamie Bennett wrote: > License files can be copied across if they are present in the setup/ > directory. Reproduced from docs/meta.md > > " > ## Fixed assets > > Some metadata is provided in the form of conventions, such as license files, > icons and desktop files among others. For these fixed files to make it into > your final snap they need to be in a `setup` directory at the same level of > your `snapcraft.yaml`. > " Thanks! I've fixed my snap accordingly. From mark at ubuntu.com Mon Sep 5 22:53:43 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Mon, 5 Sep 2016 18:53:43 -0400 Subject: (Not entirely) working LDC snap In-Reply-To: References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> Message-ID: On 05/09/16 18:15, Joseph Rushton Wakeling wrote: > On 04/09/16 13:55, Joseph Rushton Wakeling wrote: >> On 27/08/16 22:45, Joseph Rushton Wakeling wrote: >>> I thought I'd have a go at making a snap of LDC, the LLVM-based >>> compiler for the >>> D programming language. >> >> A "final first draft" of the working snap is available here: >> https://github.com/WebDrake/ldc2.snap/pull/1 > > So, you'd think I'd have tested this before, given that I'm trying to > snap a compiler, but ... :-P > > Turns out the snap-packaged LDC will fail if it's trying to compile a > project that needs other libraries to be linked in. This isn't > surprising, of course -- system libraries won't be on the library > paths of the snap -- but it does put in place a BIG constraint on what > can be built and what not. > > Any thoughts on how this might be addressed? Yeah, this is exactly what I ran into snapping asciinema - it's a tool best used WITH the other files on your system. In my case, I wanted to make an asciinema movie of a Juju deployment, but as soon as I hit "asciinema record" I was essentially inside the asciinema sandbox and unable to see binaries installed in the classic filesystem. I started a thread on one or other snapcraft list about 'snaps that can properly see the system'. A chef or puppet snap, for example is going to want to manipulate /etc/foo.conf. Your LDC snap wants to use libraries in "real" /lib/ not "sandbox" /lib/. There is a balance - the upside of having a "sandbox" /lib/ is that it is *the same everywhere* even if your snap is running on openwrt. But for this class of developer tool, snaps will be much more useful if they can really integrate into the CLASSIC environment. We'll get to this. My straw man proposal is an interface which essentially makes "/classic/lib/" and "/classic/bin/" available to your snap, so as long as your snap's $PATH is twisted appropriately it will find things there. That's a proper bit of yoga on where snaps came from and how we got them onto classic in the first place, but I think it has merit as a starting point. On the other hand, snapping redis, rethinkdb and rqlite has gone really smoothly :) Mark From c.lobrano at gmail.com Tue Sep 6 07:59:19 2016 From: c.lobrano at gmail.com (Carlo Lobrano) Date: Tue, 6 Sep 2016 09:59:19 +0200 Subject: Fwd: Re: 4G on Dell IoT Gateway In-Reply-To: References: Message-ID: Hi Oriol, according to this output Status | lock: 'unknown' | unlock retries: 'unknown' | state: 'failed' | failed reason: 'sim-missing' | power state: 'on' | access tech: 'unknown' | signal quality: '0' (cached) it looks like the SIM is not recognized at all, like it was not there. If the problem was only the PIN lock, the status should be something like lock: 'sim-pin' ... state: 'locked' in this last state, you could send the pin using mmcli -i --pin= or the others sim related mmcli commands. Sorry for the silly question, but are you sure that the SIM is properly inserted? Best regards, Carlo Hi Oriol, On Mon, Sep 5, 2016 at 4:04 PM, Oriol Rius wrote: > Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a problem > trying to set up a 4G connection using nmcli and mmcli without success. > Simple capabilities doesn't work and Dell support it's slow and > insufficient. > > My first problem starts trying to validate the PIN. But in the end what I > need is a simple tutorial or manual that help me how to set up the internet > connection. I found some information reading mailing lists and googling, > but everything is imprecise and incomplete. About the modem I use: > First, note that the HE910 is a HSPA modem and does not support 4G. You can try to activate cellular data with these steps: 1. Deactivate SIM PIN with another device 2. Execute these commands: nmcli c add type gsm ifname ttyACM3 con-name gsmconn apn nmcli r wwan on As you see you need to use nmcli instead of mmcli because we are controlling the modem from NetworkManager so it can take control of the connection and configure routes. HTH, Alfonso > > # mmcli -m 0 > > /org/freedesktop/ModemManager1/Modem/0 (device id > 'b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') > ------------------------- > Hardware | manufacturer: 'Telit' > | model: 'HE910-D' > | revision: '12.00.086' > | supported: 'gsm-umts' > | current: 'gsm-umts' > | equipment id: '356136070745999' > ------------------------- > System | device: '/sys/devices/pci0000:00/0000: > 00:14.0/usb1/1-4/1-4.1' > | drivers: 'cdc_acm, cdc_ether' > | plugin: 'Dell' > | primary port: 'ttyACM3' > | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 (at)' > ------------------------- > Numbers | own : 'unknown' > ------------------------- > Status | lock: 'unknown' > | unlock retries: 'unknown' > | state: 'failed' > | failed reason: 'sim-missing' > | power state: 'on' > | access tech: 'unknown' > | signal quality: '0' (cached) > ------------------------- > Modes | supported: 'allowed: 2g, 3g; preferred: none' > | current: 'allowed: any; preferred: none' > ------------------------- > Bands | supported: 'unknown' > | current: 'unknown' > ------------------------- > IP | supported: 'none' > ------------------------- > SIM | path: 'none' > > ------------------------- > Bearers | paths: 'none' > > > Can you recommend any resource that help me on that? > > Thank you very much. > > Regards. > Oriol > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > > -- Snapcraft mailing list Snapcraft at lists.snapcraft.io Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm an/listinfo/snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From seb128 at ubuntu.com Tue Sep 6 09:42:03 2016 From: seb128 at ubuntu.com (Sebastien Bacher) Date: Tue, 6 Sep 2016 11:42:03 +0200 Subject: No such file or directory when try to execute gsettings In-Reply-To: References: Message-ID: <07556db2-d35e-3aa6-e703-0a98ece000aa@ubuntu.com> Le 30/08/2016 à 10:15, Eloy García (PC Actual) a écrit : > Thanks for your answer, but I forgot to say I tried gsettings and > unity7 interfaces with no success. Nevertheless, I only put those > interfaces in snapcraft.yml, but I didn't do anyting else > (configurations, tweaks...). Adding more information, this is the > script executed when wallpaperdownloader is launched. Maybe it is here > where I should add some mappings to the "native" environment? Hey again, Did you figure it out? The gsettings command talk over dbus to the service which is in the user session, so the key should be written in your user db, the confined code in the snap doesn't get back the value since dconf does that by mapping a file from the user directory so it might mean your frontend might not get the new value of the key. Can you check from the outside the value of the key? If you use the shared desktop launcher it includes an hack to symlink the snap dconf db to the real one which should make reading work as well Cheers, Sebastien Bacher From oriol.rius at gmail.com Tue Sep 6 09:48:26 2016 From: oriol.rius at gmail.com (Oriol Rius) Date: Tue, 06 Sep 2016 09:48:26 +0000 Subject: Fwd: Re: 4G on Dell IoT Gateway In-Reply-To: References: Message-ID: Hi Carlo, thanks for your feedback playing about that with Alfonso. We found problems detecting SIM with that modem. The strange thing is that SIM cards that I tested are working on other devices and connected with Dell both of them fail. The error is always 'sim-missing'. I don't know how to solve that and I don't have feedback from Dell. Regards. Oriol ------ Original Message ------ From: "Carlo Lobrano" To: snapcraft at lists.snapcraft.io; oriol.rius at gmail.com; alfonso.sanchez-beato at canonical.com Sent: 06-Sep-16 09:59:19 Subject: Fwd: Re: 4G on Dell IoT Gateway >Hi Oriol, > >according to this output > >Status | lock: 'unknown' > | unlock retries: 'unknown' > | state: 'failed' > | failed reason: 'sim-missing' > | power state: 'on' > | access tech: 'unknown' > | signal quality: '0' (cached) > >it looks like the SIM is not recognized at all, like it was not there. >If the problem was only the PIN lock, the status should be something >like > >lock: 'sim-pin' >... >state: 'locked' > >in this last state, you could send the pin using > >mmcli -i --pin= > >or the others sim related mmcli commands. > >Sorry for the silly question, but are you sure that the SIM is properly >inserted? > >Best regards, >Carlo > > >Hi Oriol, > >On Mon, Sep 5, 2016 at 4:04 PM, Oriol Rius >wrote: >>Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a >>problem trying to set up a 4G connection using nmcli and mmcli without >>success. Simple capabilities doesn't work and Dell support it's slow >>and insufficient. >> >>My first problem starts trying to validate the PIN. But in the end >>what I need is a simple tutorial or manual that help me how to set up >>the internet connection. I found some information reading mailing >>lists and googling, but everything is imprecise and incomplete. About >>the modem I use: > >First, note that the HE910 is a HSPA modem and does not support 4G. > >You can try to activate cellular data with these steps: > >1. Deactivate SIM PIN with another device >2. Execute these commands: > >nmcli c add type gsm ifname ttyACM3 con-name gsmconn apn > >nmcli r wwan on > >As you see you need to use nmcli instead of mmcli because we are >controlling the modem from NetworkManager so it can take control of the >connection and configure routes. > >HTH, >Alfonso > >> >># mmcli -m 0 >> >>/org/freedesktop/ModemManager1/Modem/0 (device id >>'b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') >> ------------------------- >> Hardware | manufacturer: 'Telit' >> | model: 'HE910-D' >> | revision: '12.00.086' >> | supported: 'gsm-umts' >> | current: 'gsm-umts' >> | equipment id: '356136070745999' >> ------------------------- >> System | device: >>'/sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.1' >> | drivers: 'cdc_acm, cdc_ether' >> | plugin: 'Dell' >> | primary port: 'ttyACM3' >> | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 (at)' >> ------------------------- >> Numbers | own : 'unknown' >> ------------------------- >> Status | lock: 'unknown' >> | unlock retries: 'unknown' >> | state: 'failed' >> | failed reason: 'sim-missing' >> | power state: 'on' >> | access tech: 'unknown' >> | signal quality: '0' (cached) >> ------------------------- >> Modes | supported: 'allowed: 2g, 3g; preferred: none' >> | current: 'allowed: any; preferred: none' >> ------------------------- >> Bands | supported: 'unknown' >> | current: 'unknown' >> ------------------------- >> IP | supported: 'none' >> ------------------------- >> SIM | path: 'none' >> >> ------------------------- >> Bearers | paths: 'none' >> >> >>Can you recommend any resource that help me on that? >> >>Thank you very much. >> >>Regards. >>Oriol >> >>-- >>Snapcraft mailing list >>Snapcraft at lists.snapcraft.io >>Modify settings or unsubscribe at: >>https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > > >-- >Snapcraft mailing list >Snapcraft at lists.snapcraft.io >Modify settings or unsubscribe at: >https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Tue Sep 6 11:34:58 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Tue, 6 Sep 2016 12:34:58 +0100 Subject: Access to a Serial Port using Java RXTX library Message-ID: Hi, I have a java application running in a .snap package. The java application is going to access the serial port and uses the RXTX Java library to do this. The RXTX jar is included in the .snap. When it runs I get the following error : check_group_uucp(): error testing lock file creation Error details:Permission deniedcheck_lock_status: No permission to create lock file. please see: How can I use Lock Files with rxtx? in INSTALL Anyone have any experience of this ? Thanks in advance. Jenny -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Tue Sep 6 11:48:41 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Tue, 6 Sep 2016 13:48:41 +0200 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Hi Jenny, The INSTALL file suggests that RXTX implements custom locking which requires OS integration; because you want your app to be the only one accessing the serial port(s), you should build RXTX with --disable-lockfiles to disable this feature of the library. Cheers, - Loïc On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy wrote: > Hi, > > I have a java application running in a .snap package. > The java application is going to access the serial port and uses the RXTX > Java library to do this. The RXTX jar is included in the .snap. > > When it runs I get the following error : > > check_group_uucp(): error testing lock file creation Error > details:Permission deniedcheck_lock_status: No permission to create lock > file. > please see: How can I use Lock Files with rxtx? in INSTALL > > > Anyone have any experience of this ? > > Thanks in advance. > Jenny > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | +353 (0) 61 > 512 511 w | http://www.episensor.com > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From c.lobrano at gmail.com Tue Sep 6 12:14:00 2016 From: c.lobrano at gmail.com (Carlo Lobrano) Date: Tue, 6 Sep 2016 14:14:00 +0200 Subject: Fwd: Re: 4G on Dell IoT Gateway In-Reply-To: References: Message-ID: Hi Oriol, then I guess the problem is just that, unfortunately I am no help for this Regards On 6 September 2016 at 11:48, Oriol Rius wrote: > Hi Carlo, thanks for your feedback playing about that with Alfonso. We > found problems detecting SIM with that modem. The strange thing is that SIM > cards that I tested are working on other devices and connected with Dell > both of them fail. The error is always 'sim-missing'. > > I don't know how to solve that and I don't have feedback from Dell. > > Regards. > > Oriol > > ------ Original Message ------ > From: "Carlo Lobrano" > To: snapcraft at lists.snapcraft.io; oriol.rius at gmail.com; > alfonso.sanchez-beato at canonical.com > Sent: 06-Sep-16 09:59:19 > Subject: Fwd: Re: 4G on Dell IoT Gateway > > Hi Oriol, > > according to this output > > Status | lock: 'unknown' > | unlock retries: 'unknown' > | state: 'failed' > | failed reason: 'sim-missing' > | power state: 'on' > | access tech: 'unknown' > | signal quality: '0' (cached) > > it looks like the SIM is not recognized at all, like it was not there. If > the problem was only the PIN lock, the status should be something like > > lock: 'sim-pin' > ... > state: 'locked' > > in this last state, you could send the pin using > > mmcli -i --pin= > > or the others sim related mmcli commands. > > Sorry for the silly question, but are you sure that the SIM is properly > inserted? > > Best regards, > Carlo > > > Hi Oriol, > > On Mon, Sep 5, 2016 at 4:04 PM, Oriol Rius wrote: > >> Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a problem >> trying to set up a 4G connection using nmcli and mmcli without success. >> Simple capabilities doesn't work and Dell support it's slow and >> insufficient. >> >> My first problem starts trying to validate the PIN. But in the end what I >> need is a simple tutorial or manual that help me how to set up the internet >> connection. I found some information reading mailing lists and googling, >> but everything is imprecise and incomplete. About the modem I use: >> > > First, note that the HE910 is a HSPA modem and does not support 4G. > > You can try to activate cellular data with these steps: > > 1. Deactivate SIM PIN with another device > 2. Execute these commands: > > nmcli c add type gsm ifname ttyACM3 con-name gsmconn apn > > nmcli r wwan on > > As you see you need to use nmcli instead of mmcli because we are > controlling the modem from NetworkManager so it can take control of the > connection and configure routes. > > HTH, > Alfonso > > >> >> # mmcli -m 0 >> >> /org/freedesktop/ModemManager1/Modem/0 (device id >> 'b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') >> ------------------------- >> Hardware | manufacturer: 'Telit' >> | model: 'HE910-D' >> | revision: '12.00.086' >> | supported: 'gsm-umts' >> | current: 'gsm-umts' >> | equipment id: '356136070745999' >> ------------------------- >> System | device: '/sys/devices/pci0000:00/0000: >> 00:14.0/usb1/1-4/1-4.1' >> | drivers: 'cdc_acm, cdc_ether' >> | plugin: 'Dell' >> | primary port: 'ttyACM3' >> | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 (at)' >> ------------------------- >> Numbers | own : 'unknown' >> ------------------------- >> Status | lock: 'unknown' >> | unlock retries: 'unknown' >> | state: 'failed' >> | failed reason: 'sim-missing' >> | power state: 'on' >> | access tech: 'unknown' >> | signal quality: '0' (cached) >> ------------------------- >> Modes | supported: 'allowed: 2g, 3g; preferred: none' >> | current: 'allowed: any; preferred: none' >> ------------------------- >> Bands | supported: 'unknown' >> | current: 'unknown' >> ------------------------- >> IP | supported: 'none' >> ------------------------- >> SIM | path: 'none' >> >> ------------------------- >> Bearers | paths: 'none' >> >> >> Can you recommend any resource that help me on that? >> >> Thank you very much. >> >> Regards. >> Oriol >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Tue Sep 6 16:07:29 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 06 Sep 2016 11:07:29 -0500 Subject: Interface for accessing /dev/net/tun In-Reply-To: <20160905105133.GA4443@phoenix> References: <20160905105133.GA4443@phoenix> Message-ID: <1473178049.31168.7.camel@canonical.com> On Mon, 2016-09-05 at 11:51 +0100, Jamie Bennett wrote: > On 03/09/16 at 12:45pm, Casey Marshall wrote: > > > > I'm trying to write a snap for OpenVPN with some wrapper scripts that > > automate > > PKI setup, IP masq, etc. My goal is to be able to setup an OpenVPN server > > with > > a single command, and then issue new clients with a single command. I'm > > using > > strict mode, because that's how I roll. > > > > I've added the firewall-control and network-control plugs, and that allowed > > some things to work (like setting up IP masq, etc.). However openvpn cannot > > start up, it's getting denied from opening /dev/net/tap, https://github.com/ > > cmars/easy-openvpn-pkg/blob/master/TODO.md# > > tuntap-access-for-openvpn-in-strict-mode > > > > Would it be reasonable to contribute a patch to snapd adding the tun and/or > > tap > > devices to the network-control interface? Or would these constitute a new > > interface? > As this interface is just a network device, albeit for user space programs, it > makes sense for this to be part of the network-control interface IMHO. > I tend to agree. I suggest filing a bug and submitting a PR against network- control and we can review/discuss there. Thanks! :) -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From leo.arias at canonical.com Tue Sep 6 16:13:09 2016 From: leo.arias at canonical.com (Leo Arias) Date: Tue, 6 Sep 2016 10:13:09 -0600 Subject: command should be run as root Message-ID: <57CEEB15.5000903@canonical.com> Hello, I'm confused by commands that say they should run as root. I'm playing with runc: https://github.com/elopio/runc/tree/snapcraft It's really easy to snap, but when I run it I get: ~/snap/runc/x1/container$ runc run testcontainer runc should be run as root So if I try to run: ~/snap/runc/x1/container$ sudo runc run testcontainer open config.json: permission denied First question, does it make sense for a root user to be denied to access the files in /home/somebody/snap? And second question is what am I supposed to do now? I created the directory /root/snap/runc/x1/ and moved the container directory there. Then I'm able to read the file, but that's not a nice experience. pura vida. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From mabnhdev at gmail.com Tue Sep 6 20:27:50 2016 From: mabnhdev at gmail.com (MikeB) Date: Tue, 6 Sep 2016 16:27:50 -0400 Subject: More Ubuntu-Core boot up problems... Message-ID: I'm building a Ubuntu Core image using latest snapcraft to build a 4.4-based kernel snap. I then use ubuntu-device-flash core 16 --channel=edge --kernel=$snap --gadget=pc --os=ubuntu-core -o $IMG to create the image. I then have a script that converts the image file to an ONIE NOS installer - but that's irrelevant. Up until this weekend, things were going pretty well. I could boot ubuntu core on my target network switch, set up the Ubuntu profile, then login in using ubuntu/ubuntu. I rebuilt this weekend and now I can still boot up and set up the Ubuntu profile, but I can't login. The ubuntu/ubuntu doesn't work and I tried every combination of my Ubuntu profile user that I could think of, but can't login. So, my switch is now pretty much a brick. Here's the last things on the console I see... [ OK ] Started Generate sshd host keys. Starting OpenBSD Secure Shell server... [FAILED] Failed to start Run snappy firstboot setup. See 'systemctl status snapd.firstboot.service' for details. [ OK ] Started Snappy daemon. [ OK ] Started Network Service. Starting Update resolvconf for networkd DNS... [ OK ] Started OpenBSD Secure Shell server. [ OK ] Reached target Multi-User System. [ OK ] Reached target Graphical Interface. Starting Update UTMP about System Runlevel Changes... Starting Notify bootloader that boot was successful... [ OK ] Started Update UTMP about System Runlevel Changes. [FAILED] Failed to start Notify bootloader that boot was successful. See 'systemctl status snapd.boot-ok.service' for details. [ OK ] Started Update resolvconf for networkd DNS. The first 'Failed' has been happening all along. But the second 'Failed' is new to me. Usually, at this point I see some cloud-init/network-init messages but that's not happening. I can't login, so I can't tell you what 'systemctl status' is for those errors. Any help would be appreciated. Thanks, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From andreas at canonical.com Tue Sep 6 21:25:35 2016 From: andreas at canonical.com (Andreas Hasenack) Date: Tue, 6 Sep 2016 18:25:35 -0300 Subject: Getting http_proxy var into a snap service Message-ID: Hi, On Xenial, I just had to debug an issue where a snap service wasn't getting the proxy variables set. i traditionally set these variables in /etc/environment, and that usually worked. Some googling led me to https://bugs.launchpad.net/ubuntu/+source/lxd/+bug/1557161 where a similar problem was encountered, and the solution/workaround was to suck in /etc/environment via EnvironmentFile. So that's what I did: [Service] ... EnvironmentFile=-/etc/environment Environment="SNAP=/snap/.... " Then the service got the proxy variables, and worked. That file is provided by snappy, however, and not something that the snap developer controls with that level of detail I believe. Having the sysadm/user add an override file for just that service is also possible, but it's a bit more complicated than it should be I think. Finally, snappy would probably want to be very careful about the shell variables it imports into a snap. What would be the best way to make sure a snap gets the proxy variables it might need to work? Maybe I completely misunderstood how to set a proxy system-wide in Xenial? Thanks! -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.wakeling at webdrake.net Tue Sep 6 22:21:06 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Wed, 7 Sep 2016 00:21:06 +0200 Subject: (Not entirely) working LDC snap In-Reply-To: References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> Message-ID: <3a91ee5a-1279-a33e-b833-63a719c1cf00@webdrake.net> On 06/09/16 00:53, Mark Shuttleworth wrote: > There is a balance - the upside of having a "sandbox" /lib/ is that it > is *the same everywhere* even if your snap is running on openwrt. > > But for this class of developer tool, snaps will be much more useful if > they can really integrate into the CLASSIC environment. > > We'll get to this. My straw man proposal is an interface which > essentially makes "/classic/lib/" and "/classic/bin/" available to your > snap, so as long as your snap's $PATH is twisted appropriately it will > find things there. That's a proper bit of yoga on where snaps came from > and how we got them onto classic in the first place, but I think it has > merit as a starting point. Yea, makes sense. It'll be fun to play with when it's ready :-) There was an idea that occurred to me, which I'm not sure necessarily fits with the snappy vision, but here goes. The metaphor for interfaces is plugs and sockets -- which suggests one to one connections. But where things like development libraries are concerned, one could imagine this more like posts on a bulletin board: the individual services post their availability via a particular board (in this case, the "development library" board), and snaps interested in consuming a particular class of service can find out about and access all the services of a particular kind by asking to read the associated board. IOW instead of requesting access to one particular interface, snaps could request access to collections of services, and other snaps could add individual services to a given collection. This could be an interesting way to make development libraries available without relying on an underlying "classic" system. > On the other hand, snapping redis, rethinkdb and rqlite has gone really > smoothly :) Nice! Congratulations :-) From david.chen at canonical.com Wed Sep 7 02:15:12 2016 From: david.chen at canonical.com (David Chen) Date: Wed, 7 Sep 2016 10:15:12 +0800 Subject: Fwd: Re: 4G on Dell IoT Gateway In-Reply-To: References: Message-ID: Hi Oriol, Have you tried removing the SIM card and re-inserting it? Regards, -David On 09/06/2016 05:48 PM, Oriol Rius wrote: > Hi Carlo, thanks for your feedback playing about that with Alfonso. We > found problems detecting SIM with that modem. The strange thing is > that SIM cards that I tested are working on other devices and > connected with Dell both of them fail. The error is always 'sim-missing'. > > I don't know how to solve that and I don't have feedback from Dell. > > Regards. > > Oriol > > ------ Original Message ------ > From: "Carlo Lobrano" > > To: snapcraft at lists.snapcraft.io > ; oriol.rius at gmail.com > ; alfonso.sanchez-beato at canonical.com > > Sent: 06-Sep-16 09:59:19 > Subject: Fwd: Re: 4G on Dell IoT Gateway > >> Hi Oriol, >> >> according to this output >> >> Status | lock: 'unknown' >> | unlock retries: 'unknown' >> | state: 'failed' >> | failed reason: 'sim-missing' >> | power state: 'on' >> | access tech: 'unknown' >> | signal quality: '0' (cached) >> >> it looks like the SIM is not recognized at all, like it was not >> there. If the problem was only the PIN lock, the status should be >> something like >> >> lock: 'sim-pin' >> ... >> state: 'locked' >> >> in this last state, you could send the pin using >> >> mmcli -i --pin= >> >> or the others sim related mmcli commands. >> >> Sorry for the silly question, but are you sure that the SIM is >> properly inserted? >> >> Best regards, >> Carlo >> >> >> Hi Oriol, >> >> On Mon, Sep 5, 2016 at 4:04 PM, Oriol Rius > > wrote: >> >> Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a >> problem trying to set up a 4G connection using nmcli and mmcli >> without success. Simple capabilities doesn't work and Dell >> support it's slow and insufficient. >> >> My first problem starts trying to validate the PIN. But in the >> end what I need is a simple tutorial or manual that help me how >> to set up the internet connection. I found some information >> reading mailing lists and googling, but everything is imprecise >> and incomplete. About the modem I use: >> >> >> First, note that the HE910 is a HSPA modem and does not support 4G. >> >> You can try to activate cellular data with these steps: >> >> 1. Deactivate SIM PIN with another device >> 2. Execute these commands: >> >> nmcli c add type gsm ifname ttyACM3 con-name gsmconn apn >> >> nmcli r wwan on >> >> As you see you need to use nmcli instead of mmcli because we are >> controlling the modem from NetworkManager so it can take control of >> the connection and configure routes. >> >> HTH, >> Alfonso >> >> >> >> # mmcli -m 0 >> >> /org/freedesktop/ModemManager1/Modem/0 (device id >> 'b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') >> ------------------------- >> Hardware | manufacturer: 'Telit' >> | model: 'HE910-D' >> | revision: '12.00.086' >> | supported: 'gsm-umts' >> | current: 'gsm-umts' >> | equipment id: '356136070745999' >> ------------------------- >> System | device: >> '/sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.1' >> | drivers: 'cdc_acm, cdc_ether' >> | plugin: 'Dell' >> | primary port: 'ttyACM3' >> | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 (at)' >> ------------------------- >> Numbers | own : 'unknown' >> ------------------------- >> Status | lock: 'unknown' >> | unlock retries: 'unknown' >> | state: 'failed' >> | failed reason: 'sim-missing' >> | power state: 'on' >> | access tech: 'unknown' >> | signal quality: '0' (cached) >> ------------------------- >> Modes | supported: 'allowed: 2g, 3g; preferred: none' >> | current: 'allowed: any; preferred: none' >> ------------------------- >> Bands | supported: 'unknown' >> | current: 'unknown' >> ------------------------- >> IP | supported: 'none' >> ------------------------- >> SIM | path: 'none' >> >> ------------------------- >> Bearers | paths: 'none' >> >> >> Can you recommend any resource that help me on that? >> >> Thank you very much. >> >> Regards. >> Oriol >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From oriol.rius at gmail.com Wed Sep 7 05:08:20 2016 From: oriol.rius at gmail.com (Oriol Rius) Date: Wed, 07 Sep 2016 05:08:20 +0000 Subject: Fwd: Re: 4G on Dell IoT Gateway In-Reply-To: References: Message-ID: In the end the solution was so simple like that David, thanks! Alfonso, helped me yesterday afternoon through chat and after some tries with two SIMs we've got that running. Thank you for your support. Oriol ------ Original Message ------ From: "David Chen" To: "Oriol Rius" ; "Carlo Lobrano" ; snapcraft at lists.snapcraft.io; alfonso.sanchez-beato at canonical.com Sent: 07-Sep-16 04:15:12 Subject: Re: Fwd: Re: 4G on Dell IoT Gateway >Hi Oriol, > >Have you tried removing the SIM card and re-inserting it? > >Regards, > >-David > >On 09/06/2016 05:48 PM, Oriol Rius wrote: >>Hi Carlo, thanks for your feedback playing about that with Alfonso. We >>found problems detecting SIM with that modem. The strange thing is >>that SIM cards that I tested are working on other devices and >>connected with Dell both of them fail. The error is always >>'sim-missing'. >> >>I don't know how to solve that and I don't have feedback from Dell. >> >>Regards. >> >>Oriol >> >>------ Original Message ------ >>From: "Carlo Lobrano" >>To: snapcraft at lists.snapcraft.io; oriol.rius at gmail.com; >>alfonso.sanchez-beato at canonical.com >>Sent: 06-Sep-16 09:59:19 >>Subject: Fwd: Re: 4G on Dell IoT Gateway >> >>>Hi Oriol, >>> >>>according to this output >>> >>>Status | lock: 'unknown' >>> | unlock retries: 'unknown' >>> | state: 'failed' >>> | failed reason: 'sim-missing' >>> | power state: 'on' >>> | access tech: 'unknown' >>> | signal quality: '0' (cached) >>> >>>it looks like the SIM is not recognized at all, like it was not >>>there. If the problem was only the PIN lock, the status should be >>>something like >>> >>>lock: 'sim-pin' >>>... >>>state: 'locked' >>> >>>in this last state, you could send the pin using >>> >>>mmcli -i --pin= >>> >>>or the others sim related mmcli commands. >>> >>>Sorry for the silly question, but are you sure that the SIM is >>>properly inserted? >>> >>>Best regards, >>>Carlo >>> >>> >>>Hi Oriol, >>> >>>On Mon, Sep 5, 2016 at 4:04 PM, Oriol Rius >>>wrote: >>>>Hi, again about Dell IoT Gateway 5000 with Snappy 15.04 I have a >>>>problem trying to set up a 4G connection using nmcli and mmcli >>>>without success. Simple capabilities doesn't work and Dell support >>>>it's slow and insufficient. >>>> >>>>My first problem starts trying to validate the PIN. But in the end >>>>what I need is a simple tutorial or manual that help me how to set >>>>up the internet connection. I found some information reading mailing >>>>lists and googling, but everything is imprecise and incomplete. >>>>About the modem I use: >>> >>>First, note that the HE910 is a HSPA modem and does not support 4G. >>> >>>You can try to activate cellular data with these steps: >>> >>>1. Deactivate SIM PIN with another device >>>2. Execute these commands: >>> >>>nmcli c add type gsm ifname ttyACM3 con-name gsmconn apn >>> >>>nmcli r wwan on >>> >>>As you see you need to use nmcli instead of mmcli because we are >>>controlling the modem from NetworkManager so it can take control of >>>the connection and configure routes. >>> >>>HTH, >>>Alfonso >>> >>>> >>>># mmcli -m 0 >>>> >>>>/org/freedesktop/ModemManager1/Modem/0 (device id >>>>'b73aa014c0d2097ce1e62fd9aa2a82c6f8dc6307') >>>> ------------------------- >>>> Hardware | manufacturer: 'Telit' >>>> | model: 'HE910-D' >>>> | revision: '12.00.086' >>>> | supported: 'gsm-umts' >>>> | current: 'gsm-umts' >>>> | equipment id: '356136070745999' >>>> ------------------------- >>>> System | device: >>>>'/sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.1' >>>> | drivers: 'cdc_acm, cdc_ether' >>>> | plugin: 'Dell' >>>> | primary port: 'ttyACM3' >>>> | ports: 'ttyACM3 (at), wwan0 (net), ttyACM0 >>>>(at)' >>>> ------------------------- >>>> Numbers | own : 'unknown' >>>> ------------------------- >>>> Status | lock: 'unknown' >>>> | unlock retries: 'unknown' >>>> | state: 'failed' >>>> | failed reason: 'sim-missing' >>>> | power state: 'on' >>>> | access tech: 'unknown' >>>> | signal quality: '0' (cached) >>>> ------------------------- >>>> Modes | supported: 'allowed: 2g, 3g; preferred: none' >>>> | current: 'allowed: any; preferred: none' >>>> ------------------------- >>>> Bands | supported: 'unknown' >>>> | current: 'unknown' >>>> ------------------------- >>>> IP | supported: 'none' >>>> ------------------------- >>>> SIM | path: 'none' >>>> >>>> ------------------------- >>>> Bearers | paths: 'none' >>>> >>>> >>>>Can you recommend any resource that help me on that? >>>> >>>>Thank you very much. >>>> >>>>Regards. >>>>Oriol >>>> >>>>-- >>>>Snapcraft mailing list >>>>Snapcraft at lists.snapcraft.io >>>>Modify settings or unsubscribe at: >>>>https://lists.ubuntu.com/mailman/listinfo/snapcraft >>>> >>> >>> >>>-- >>>Snapcraft mailing list >>>Snapcraft at lists.snapcraft.io >>>Modify settings or unsubscribe at: >>>https://lists.ubuntu.com/mailman/listinfo/snapcraft >>> >>> >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie.bennett at canonical.com Wed Sep 7 09:02:43 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Wed, 7 Sep 2016 10:02:43 +0100 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> Hi Mike, > On 6 Sep 2016, at 21:27, MikeB wrote: > > I'm building a Ubuntu Core image using latest snapcraft to build a 4.4-based kernel snap. > > I then use ubuntu-device-flash core 16 --channel=edge --kernel=$snap --gadget=pc --os=ubuntu-core -o $IMG to create the image. > > I then have a script that converts the image file to an ONIE NOS installer - but that's irrelevant. > > Up until this weekend, things were going pretty well. I could boot ubuntu core on my target network switch, set up the Ubuntu profile, then login in using ubuntu/ubuntu. > > I rebuilt this weekend and now I can still boot up and set up the Ubuntu profile, but I can't login. The ubuntu/ubuntu doesn't work and I tried every combination of my Ubuntu profile user that I could think of, but can't login. Recently we changed the first boot experience, removing the default user (this is insecure if people forget to remove/change the password) and instead we require that the device is set up manually. This process uses console-conf and will walk you through setting up a first user (must have a store account) and any network changes that may be needed. For this you will need a screen and keyboard attached. We are looking at alternative options to configure a device for people who do not have a screen and keyboard attached but that is something for the near future. Regards, Jamie. > So, my switch is now pretty much a brick. > > Here's the last things on the console I see... > > [ OK ] Started Generate sshd host keys. > Starting OpenBSD Secure Shell server... > [FAILED] Failed to start Run snappy firstboot setup. > See 'systemctl status snapd.firstboot.service' for details. > [ OK ] Started Snappy daemon. > [ OK ] Started Network Service. > Starting Update resolvconf for networkd DNS... > [ OK ] Started OpenBSD Secure Shell server. > [ OK ] Reached target Multi-User System. > [ OK ] Reached target Graphical Interface. > Starting Update UTMP about System Runlevel Changes... > Starting Notify bootloader that boot was successful... > [ OK ] Started Update UTMP about System Runlevel Changes. > [FAILED] Failed to start Notify bootloader that boot was successful. > See 'systemctl status snapd.boot-ok.service' for details. > [ OK ] Started Update resolvconf for networkd DNS. > > > The first 'Failed' has been happening all along. But the second 'Failed' is new to me. Usually, at this point I see some cloud-init/network-init messages but that's not happening. > > I can't login, so I can't tell you what 'systemctl status' is for those errors. > > Any help would be appreciated. > > Thanks, Mike > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From victor.palau at canonical.com Wed Sep 7 09:11:13 2016 From: victor.palau at canonical.com (Victor Palau) Date: Wed, 7 Sep 2016 10:11:13 +0100 Subject: More Ubuntu-Core boot up problems... In-Reply-To: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> Message-ID: Hi Mike, We are seeing a similar issue, my understanding is to create a working image right now for the edge channel you require a model assertion. However, the changes to upload your key to the store, so you can sign your own model assertion and the right tools to then build an image from (ubuntu-image) have not quite landed yet. That might be why firstboot.service is failing, which I believe will preclude you from accessing console-conf to be able to set up a user to ssh into the box. Hope that helps Thanks Victor On Wed, Sep 7, 2016 at 10:02 AM, Jamie Bennett wrote: > Hi Mike, > > On 6 Sep 2016, at 21:27, MikeB wrote: > > I'm building a Ubuntu Core image using latest snapcraft to build a > 4.4-based kernel snap. > > I then use ubuntu-device-flash core 16 --channel=edge --kernel=$snap > --gadget=pc --os=ubuntu-core -o $IMG to create the image. > > I then have a script that converts the image file to an ONIE NOS installer > - but that's irrelevant. > > Up until this weekend, things were going pretty well. I could boot ubuntu > core on my target network switch, set up the Ubuntu profile, then login in > using ubuntu/ubuntu. > > I rebuilt this weekend and now I can still boot up and set up the Ubuntu > profile, but I can't login. The ubuntu/ubuntu doesn't work and I tried > every combination of my Ubuntu profile user that I could think of, but > can't login. > > > Recently we changed the first boot experience, removing the default user > (this is insecure if people forget to remove/change the password) and > instead we require that the device is set up manually. This process uses > console-conf and will walk you through setting up a first user (must have a > store account) and any network changes that may be needed. For this you > will need a screen and keyboard attached. We are looking at alternative > options to configure a device for people who do not have a screen and > keyboard attached but that is something for the near future. > > Regards, > Jamie. > > So, my switch is now pretty much a brick. > > Here's the last things on the console I see... > > [ OK ] Started Generate sshd host keys. > Starting OpenBSD Secure Shell server... > [FAILED] Failed to start Run snappy firstboot setup. > See 'systemctl status snapd.firstboot.service' for details. > [ OK ] Started Snappy daemon. > [ OK ] Started Network Service. > Starting Update resolvconf for networkd DNS... > [ OK ] Started OpenBSD Secure Shell server. > [ OK ] Reached target Multi-User System. > [ OK ] Reached target Graphical Interface. > Starting Update UTMP about System Runlevel Changes... > Starting Notify bootloader that boot was successful... > [ OK ] Started Update UTMP about System Runlevel Changes. > [FAILED] Failed to start Notify bootloader that boot was successful. > See 'systemctl status snapd.boot-ok.service' for details. > [ OK ] Started Update resolvconf for networkd DNS. > > > The first 'Failed' has been happening all along. But the second 'Failed' > is new to me. Usually, at this point I see some cloud-init/network-init > messages but that's not happening. > > I can't login, so I can't tell you what 'systemctl status' is for those > errors. > > Any help would be appreciated. > > Thanks, Mike > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Wed Sep 7 09:52:47 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 07 Sep 2016 11:52:47 +0200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> Message-ID: <1473241967.24404.5.camel@ubuntu.com> hi, On Mi, 2016-09-07 at 10:11 +0100, Victor Palau wrote: > Hi Mike, > > We are seeing a similar issue, my understanding is to create a > working image right now for the edge channel you require a model > assertion. However, the changes to upload your key to the store, so > you can sign your own model assertion and the right tools to then > build an image from (ubuntu-image) have not quite landed yet. > > That might be why firstboot.service is failing, which I believe will > preclude you from accessing console-conf to be able to set up a user > to ssh into the box. > the two are unrelated ... console-conf starts in any case so you can always set up an ssh user (as long as you have network access on the device and a store account with ssh key in launchpad least)  depending on the console (console-conf isnt actually great with serial) you will not see the "please press enter" message that starts the config tool though. if you press enter anyway it will start the config UI after a while. the failing firstboot service is indeed most likely related to a missing model assertion, but that should not prevent you from having a booting img or from being able to use console-conf to set up an admin user. it will (currently) only affect any snap related bits of the system (install, upgrade, roll back, searching for snaps etc). if firstboot would actually block the boot without model assertion, porting to a new device would be nearly impossible :) ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From mabnhdev at gmail.com Wed Sep 7 10:03:02 2016 From: mabnhdev at gmail.com (MikeB) Date: Wed, 7 Sep 2016 06:03:02 -0400 Subject: More Ubuntu-Core boot up problems... In-Reply-To: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> Message-ID: Yes, I have gone through the manual setup of the first user successfully. However, I've tried every combination of user name and password I could think of and none of them allow me to login. Given that I've set up the first user using the email address I use for Ubuntu One, what should I use for login credentials for that user - what is the username I should use and what is the password I should use? Regards, Mike On Wed, Sep 7, 2016 at 5:02 AM, Jamie Bennett wrote: > Hi Mike, > > On 6 Sep 2016, at 21:27, MikeB wrote: > > I'm building a Ubuntu Core image using latest snapcraft to build a > 4.4-based kernel snap. > > I then use ubuntu-device-flash core 16 --channel=edge --kernel=$snap > --gadget=pc --os=ubuntu-core -o $IMG to create the image. > > I then have a script that converts the image file to an ONIE NOS installer > - but that's irrelevant. > > Up until this weekend, things were going pretty well. I could boot ubuntu > core on my target network switch, set up the Ubuntu profile, then login in > using ubuntu/ubuntu. > > I rebuilt this weekend and now I can still boot up and set up the Ubuntu > profile, but I can't login. The ubuntu/ubuntu doesn't work and I tried > every combination of my Ubuntu profile user that I could think of, but > can't login. > > > Recently we changed the first boot experience, removing the default user > (this is insecure if people forget to remove/change the password) and > instead we require that the device is set up manually. This process uses > console-conf and will walk you through setting up a first user (must have a > store account) and any network changes that may be needed. For this you > will need a screen and keyboard attached. We are looking at alternative > options to configure a device for people who do not have a screen and > keyboard attached but that is something for the near future. > > Regards, > Jamie. > > So, my switch is now pretty much a brick. > > Here's the last things on the console I see... > > [ OK ] Started Generate sshd host keys. > Starting OpenBSD Secure Shell server... > [FAILED] Failed to start Run snappy firstboot setup. > See 'systemctl status snapd.firstboot.service' for details. > [ OK ] Started Snappy daemon. > [ OK ] Started Network Service. > Starting Update resolvconf for networkd DNS... > [ OK ] Started OpenBSD Secure Shell server. > [ OK ] Reached target Multi-User System. > [ OK ] Reached target Graphical Interface. > Starting Update UTMP about System Runlevel Changes... > Starting Notify bootloader that boot was successful... > [ OK ] Started Update UTMP about System Runlevel Changes. > [FAILED] Failed to start Notify bootloader that boot was successful. > See 'systemctl status snapd.boot-ok.service' for details. > [ OK ] Started Update resolvconf for networkd DNS. > > > The first 'Failed' has been happening all along. But the second 'Failed' > is new to me. Usually, at this point I see some cloud-init/network-init > messages but that's not happening. > > I can't login, so I can't tell you what 'systemctl status' is for those > errors. > > Any help would be appreciated. > > Thanks, Mike > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From eloy.garcia.pca at gmail.com Wed Sep 7 10:04:45 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Wed, 7 Sep 2016 12:04:45 +0200 Subject: No such file or directory when try to execute gsettings In-Reply-To: <07556db2-d35e-3aa6-e703-0a98ece000aa@ubuntu.com> References: <07556db2-d35e-3aa6-e703-0a98ece000aa@ubuntu.com> Message-ID: Hi Sebastien! First of all, thank you very much for your suggestions :) I get your point and I think this is the real problem. I guess gsettings command line tool from the snap is storing the correct key/value inside the snap, but the "native" dconf database is not getting the new value, because the change wasn't made using "native" gsettings command line tool. I have tried to search dconf database within snap confinement, but I couldn't find it. Where should I search for it to see if the new value for the wallpaper was set there? P.S.: I've been searching within the tree which is in /snap/wallpaperdownloader/current with no luck. Thank you very much again :D 2016-09-06 11:42 GMT+02:00 Sebastien Bacher : > Le 30/08/2016 à 10:15, Eloy García (PC Actual) a écrit : > > Thanks for your answer, but I forgot to say I tried gsettings and > > unity7 interfaces with no success. Nevertheless, I only put those > > interfaces in snapcraft.yml, but I didn't do anyting else > > (configurations, tweaks...). Adding more information, this is the > > script executed when wallpaperdownloader is launched. Maybe it is here > > where I should add some mappings to the "native" environment? > > > Hey again, > > Did you figure it out? The gsettings command talk over dbus to the > service which is in the user session, so the key should be written in > your user db, the confined code in the snap doesn't get back the value > since dconf does that by mapping a file from the user directory so it > might mean your frontend might not get the new value of the key. Can you > check from the outside the value of the key? > > If you use the shared desktop launcher it includes an hack to symlink > the snap dconf db to the real one which should make reading work as well > > Cheers, > > Sebastien Bacher > > -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Wed Sep 7 10:07:08 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 07 Sep 2016 12:07:08 +0200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> Message-ID: <1473242828.24404.7.camel@ubuntu.com> hi, On Mi, 2016-09-07 at 06:03 -0400, MikeB wrote: > Yes, I have gone through the manual setup of the first user > successfully.  However, I've tried every combination of user name and > password I could think of and none of them allow me to login. > > Given that I've set up the first user using the email address I use > for Ubuntu One, what should I use for login credentials for that user > - what is the username I should use and what is the password I should > use? >  console-conf prints exactly what to use on the last screen (the one with the "finish" button). there is no password set, you can not log in locally, only key based ssh logins are set up for this user, there is no password enabled by default. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From michael.hudson at canonical.com Wed Sep 7 10:14:13 2016 From: michael.hudson at canonical.com (Michael Hudson-Doyle) Date: Wed, 7 Sep 2016 22:14:13 +1200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: <1473241967.24404.5.camel@ubuntu.com> References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> <1473241967.24404.5.camel@ubuntu.com> Message-ID: On 7 September 2016 at 21:52, Oliver Grawert wrote: > > depending on the console (console-conf isnt actually great with serial) > you will not see the "please press enter" message that starts the > config tool though. I'd love to know why this isn't working, btw. We're just using agetty's -f option... Cheers, mwh -------------- next part -------------- An HTML attachment was scrubbed... URL: From ondrej.kubik at canonical.com Wed Sep 7 10:26:59 2016 From: ondrej.kubik at canonical.com (Ondrej Kubik) Date: Wed, 7 Sep 2016 11:26:59 +0100 Subject: More Ubuntu-Core boot up problems... In-Reply-To: <1473241967.24404.5.camel@ubuntu.com> References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> <1473241967.24404.5.camel@ubuntu.com> Message-ID: On Wed, Sep 7, 2016 at 10:52 AM, Oliver Grawert wrote: > hi, > On Mi, 2016-09-07 at 10:11 +0100, Victor Palau wrote: > > Hi Mike, > > > > We are seeing a similar issue, my understanding is to create a > > working image right now for the edge channel you require a model > > assertion. However, the changes to upload your key to the store, so > > you can sign your own model assertion and the right tools to then > > build an image from (ubuntu-image) have not quite landed yet. > > > > That might be why firstboot.service is failing, which I believe will > > preclude you from accessing console-conf to be able to set up a user > > to ssh into the box. > > > the two are unrelated ... console-conf starts in any case so you can > always set up an ssh user (as long as you have network access on the > device and a store account with ssh key in launchpad least) > > depending on the console (console-conf isnt actually great with serial) > you will not see the "please press enter" message that starts the > config tool though. if you press enter anyway it will start the config > UI after a while. > > the failing firstboot service is indeed most likely related to a > missing model assertion, but that should not prevent you from having a > booting img or from being able to use console-conf to set up an admin > user. it will (currently) only affect any snap related bits of the > system (install, upgrade, roll back, searching for snaps etc). > > if firstboot would actually block the boot without model assertion, > porting to a new device would be nearly impossible :) > Sadly this is exactly the case. I'm trapped in same situation, using own gadget snap and building image from edge channel. It won't boot as snapd crashes with missing model assertion, you won't get to console-conf at all... So as it stands porting to new device is no go. Ondra > > ciao > oli > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Wed Sep 7 10:32:28 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 07 Sep 2016 12:32:28 +0200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: <65819433-480D-4799-BA30-D65FEBE17F7B@canonical.com> <1473241967.24404.5.camel@ubuntu.com> Message-ID: <1473244348.24404.13.camel@ubuntu.com> hi, On Mi, 2016-09-07 at 11:26 +0100, Ondrej Kubik wrote: >  > > if firstboot would actually block the boot without model assertion, > > porting to a new device would be nearly impossible :) > Sadly this is exactly the case. I'm trapped in same situation, using > own gadget snap and building image from edge channel. It won't boot > as snapd crashes with missing model assertion, you won't get to > console-conf at all... > So as it stands porting to new device is no go. >  console-conf starts in any case ... we are in the middle of porting to ubuntu-image currently and i had 100s of firstboot failures and tons of non-starting snapd issues within the last week, none of them blocked console-conf. if it does not start this must be some other issue, there is definitely no relation between firstboot, snapd and console-conf. as long as a tty can start on your device, console-conf can start too (but as i said, you might not see the message and it can take a long time to start the python interpreter and load all modules on arm devices which means you have to wait a bit after hitting enter til the UI shows up on low level arm boards). ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From matthew.williams at canonical.com Wed Sep 7 10:40:32 2016 From: matthew.williams at canonical.com (Matthew Williams) Date: Wed, 7 Sep 2016 11:40:32 +0100 Subject: Snapping Neovim In-Reply-To: <57C07AFC.2030300@canonical.com> References: <57C07AFC.2030300@canonical.com> Message-ID: I've commented on the bug, but there's a larger question here I can't work out the answer to, which is how would a user expect neovim to be restricted when installed via a snap? It could be reasonable to suggest it can only edit files in $HOME, but that's not useful to sysadmins, so why would a sysadmin use snaps to install neovim? Matty On Fri, Aug 26, 2016 at 6:23 PM, Leo Arias wrote: > On 2016-08-26 10:53, Andreas Hasenack wrote: > > Can you change neovim to look for these dot files elsewhere, like > > $SNAP_USER_DATA (if I didn't typoed the var name)? > > That is right for the config files of Neovim itself. However, I would > like to be able to edit something like ~/.bashrc. We have a bug for that: > > https://bugs.launchpad.net/snappy/+bug/1607067 > > It doesn't have a lot of information, so feel free to expand it in the > comments. > > pura vida. > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mabnhdev at gmail.com Wed Sep 7 10:56:38 2016 From: mabnhdev at gmail.com (MikeB) Date: Wed, 7 Sep 2016 06:56:38 -0400 Subject: More Ubuntu-Core boot up problems... Message-ID: On 2016-09-07 at 03:08:05 -0700, Oliver Graawert wrote: console-conf prints exactly what to use on the last screen (the one > with the "finish" button). there is no password set, you can not log in > locally, only key based ssh logins are set up for this user, there is > no password enabled by default. I was able to login via ssh as you said - thanks! So, the intention is that the 'first user' logs in via ssh and then creates the necessary local users using 'sudo adduser '? However, when I try to add a new user, it fails as follows. I've tried rebooting, etc, but that didn't change anything. sudo adduser admin Adding user `admin' ... Adding new group `admin' (1001) ... groupadd: cannot lock /etc/group; try again later. adduser: `/usr/sbin/groupadd -g 1001 admin' returned error code 10. Exiting. I also have a few follow-up questions to subjects brought up in this thread. This new 'first user experience' caught me by surprise - I don't think I saw anything about this on the snapcraft mailing list. Is there some other list I'm not following that discusses these changes ahead of time? I'm still using ubuntu-device-flash to build my image. Is there any way for me to fix the missing model assertion problem yet? Do I have to wait for ubuntu-image or is there something I can do sooner? Thanks for all the help. Regards, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Wed Sep 7 10:58:29 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 7 Sep 2016 07:58:29 -0300 Subject: Snapping Neovim In-Reply-To: References: <57C07AFC.2030300@canonical.com> Message-ID: They can install it as devmode, and we can also introduce an "editor-support" interface which gives global read/write access to it. Would need to be manually connected for the time being, but assertion control for that is coming very soon. On Wed, Sep 7, 2016 at 7:40 AM, Matthew Williams < matthew.williams at canonical.com> wrote: > I've commented on the bug, but there's a larger question here I can't work > out the answer to, which is how would a user expect neovim to be restricted > when installed via a snap? It could be reasonable to suggest it can only > edit files in $HOME, but that's not useful to sysadmins, so why would a > sysadmin use snaps to install neovim? > > Matty > > On Fri, Aug 26, 2016 at 6:23 PM, Leo Arias > wrote: > >> On 2016-08-26 10:53, Andreas Hasenack wrote: >> > Can you change neovim to look for these dot files elsewhere, like >> > $SNAP_USER_DATA (if I didn't typoed the var name)? >> >> That is right for the config files of Neovim itself. However, I would >> like to be able to edit something like ~/.bashrc. We have a bug for that: >> >> https://bugs.launchpad.net/snappy/+bug/1607067 >> >> It doesn't have a lot of information, so feel free to expand it in the >> comments. >> >> pura vida. >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From sylvain.pineau at canonical.com Wed Sep 7 11:16:36 2016 From: sylvain.pineau at canonical.com (Sylvain Pineau) Date: Wed, 7 Sep 2016 13:16:36 +0200 Subject: /snap/bin not in $PATH on desktop Message-ID: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> Hello, I noticed that I was not able to call other snap commands from my own snap on my desktop. So I tested what was defined using the hello-world.env command. On desktop (with ubuntu-core 16.04.1 rev 352) $ hello-world.env | grep ^PATH= PATH=/usr/sbin:/usr/bin:/sbin:/bin:/usr/games But on a true snappy system (with ubuntu-core 16.04.1 rev 453), I get: $ hello-world.env | grep ^PATH= PATH=/home/ubuntu/bin:/home/ubuntu/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin Is there any reason to not have /snap/bin as part of the $PATH available to snap commands? Sylvain From mabnhdev at gmail.com Wed Sep 7 11:29:11 2016 From: mabnhdev at gmail.com (MikeB) Date: Wed, 7 Sep 2016 07:29:11 -0400 Subject: More Ubuntu-Core boot up problems... Message-ID: On 2016-09-07 at 03:28:35 -0700, Ondrej Kubik wrote: Sadly this is exactly the case. I'm trapped in same situation, using own > gadget snap and building image from edge channel. It won't boot as snapd > crashes with missing model assertion Initially, I didn't see snapd crashing during boot on my newly installed system. However, after I installed the 'hello-world' snap to make sure all was well in snappy land, snapd does indeed crash on subsequent reboots. So, it comes up fine as long as you don't install anything. I guess this adds urgency to my previous question about fixing the 'model assertion' crash. Cheers, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From matthew.williams at canonical.com Wed Sep 7 11:32:59 2016 From: matthew.williams at canonical.com (Matthew Williams) Date: Wed, 7 Sep 2016 12:32:59 +0100 Subject: Snapping Neovim In-Reply-To: References: <57C07AFC.2030300@canonical.com> Message-ID: Hi Gustavo, devmode was what I was thinking, I'd not consider an interface like that, I'll look into it Thanks Matty On Wed, Sep 7, 2016 at 11:58 AM, Gustavo Niemeyer < gustavo.niemeyer at canonical.com> wrote: > They can install it as devmode, and we can also introduce an > "editor-support" interface which gives global read/write access to it. > > Would need to be manually connected for the time being, but assertion > control for that is coming very soon. > > On Wed, Sep 7, 2016 at 7:40 AM, Matthew Williams < > matthew.williams at canonical.com> wrote: > >> I've commented on the bug, but there's a larger question here I can't >> work out the answer to, which is how would a user expect neovim to be >> restricted when installed via a snap? It could be reasonable to suggest it >> can only edit files in $HOME, but that's not useful to sysadmins, so why >> would a sysadmin use snaps to install neovim? >> >> Matty >> >> On Fri, Aug 26, 2016 at 6:23 PM, Leo Arias >> wrote: >> >>> On 2016-08-26 10:53, Andreas Hasenack wrote: >>> > Can you change neovim to look for these dot files elsewhere, like >>> > $SNAP_USER_DATA (if I didn't typoed the var name)? >>> >>> That is right for the config files of Neovim itself. However, I would >>> like to be able to edit something like ~/.bashrc. We have a bug for that: >>> >>> https://bugs.launchpad.net/snappy/+bug/1607067 >>> >>> It doesn't have a lot of information, so feel free to expand it in the >>> comments. >>> >>> pura vida. >>> >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>> an/listinfo/snapcraft >>> >>> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > > -- > gustavo @ http://niemeyer.net > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Wed Sep 7 11:51:45 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 7 Sep 2016 12:51:45 +0100 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Hi, I am having some difficulty building RXTX (get unsatisfied link errors with the new shared objects and jar file). Is there an other alternative to workaround my original problem with the permissions to use lock file? Thanks. On 6 September 2016 at 12:48, Loïc Minier wrote: > Hi Jenny, > > The INSTALL file > suggests that RXTX implements custom locking which requires OS integration; > because you want your app to be the only one accessing the serial port(s), > you should build RXTX with --disable-lockfiles to disable this feature of > the library. > > Cheers, > - Loïc > > On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy > wrote: > >> Hi, >> >> I have a java application running in a .snap package. >> The java application is going to access the serial port and uses the RXTX >> Java library to do this. The RXTX jar is included in the .snap. >> >> When it runs I get the following error : >> >> check_group_uucp(): error testing lock file creation Error >> details:Permission deniedcheck_lock_status: No permission to create lock >> file. >> please see: How can I use Lock Files with rxtx? in INSTALL >> >> >> Anyone have any experience of this ? >> >> Thanks in advance. >> Jenny >> >> -- >> *Jenny Murphy* >> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >> jenny.murphy at episensor.com t | +353 (0) 61 >> 512 511 w | http://www.episensor.com >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > > -- > - Loïc > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From mabnhdev at gmail.com Wed Sep 7 12:54:23 2016 From: mabnhdev at gmail.com (MikeB) Date: Wed, 7 Sep 2016 08:54:23 -0400 Subject: More Ubuntu-Core boot up problems... Message-ID: On 2016-09-07 at 06:56:38 -0400, MikeB wrote: > > However, when I try to add a new user, it fails as follows. I've tried > rebooting, etc, but that didn't change anything. New users have to be added to the extrausers files. sudo adduser --extrausers works for adding a new user. However, the usermod command does not have an '--extrausers' option. All attempts to add my new user to the 'sudo' group fail with form of an error trying to lock '/etc/group'. Oddly, even adduser can't add a user to an existing group. sudo adduser --extrausers foo sudo Adding user `foo' to group `sudo' ... Adding user foo to group sudo gpasswd: cannot lock /etc/group; try again later. adduser: `/usr/bin/gpasswd -a foo sudo' returned error code 1. Exiting. That seems like a bug to me. Can someone suggest a way to add an existing user to the 'sudo' group in Ubuntu-Core? Thanks, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Wed Sep 7 13:01:54 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 07 Sep 2016 15:01:54 +0200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: <1473253314.17795.2.camel@ubuntu.com> hi, Am Mittwoch, den 07.09.2016, 08:54 -0400 schrieb MikeB: > On 2016-09-07 at 06:56:38 -0400, MikeB wrote: > > > > However, when I try to add a new user, it fails as follows. I've > > tried rebooting, etc, but that didn't change anything. > New users have to be added to the extrausers files. > > sudo adduser --extrausers > > works for adding a new user. > > However, the usermod command does not have an '--extrausers' option.  > All attempts to add my new user to the 'sudo' group fail with form of > an error trying to lock '/etc/group'. > > Oddly, even adduser can't add a user to an existing group. > > sudo adduser --extrausers foo sudo > Adding user `foo' to group `sudo' ... > Adding user foo to group sudo > gpasswd: cannot lock /etc/group; try again later. > adduser: `/usr/bin/gpasswd -a foo sudo' returned error code 1. > Exiting. > > That seems like a bug to me. > no, thats a design limitation ... the sudo group has to be in /etc/groups which is in the readonly snap ...  https://bugs.launchpad.net/snappy/+bug/1495059 has a workaround via adding a file to /etc/sudoers.d/ ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From jenny.murphy at episensor.com Wed Sep 7 13:32:49 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 7 Sep 2016 14:32:49 +0100 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Hi all, This is solved . The problem was to do with the Java versions supported by RXTX. Jenny On 7 September 2016 at 12:51, Jenny Murphy wrote: > Hi, > I am having some difficulty building RXTX (get unsatisfied link errors > with the new shared objects and jar file). > Is there an other alternative to workaround my original problem with the > permissions to use lock file? > Thanks. > > > On 6 September 2016 at 12:48, Loïc Minier wrote: > >> Hi Jenny, >> >> The INSTALL file >> suggests that RXTX implements custom locking which requires OS integration; >> because you want your app to be the only one accessing the serial port(s), >> you should build RXTX with --disable-lockfiles to disable this feature of >> the library. >> >> Cheers, >> - Loïc >> >> On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy >> wrote: >> >>> Hi, >>> >>> I have a java application running in a .snap package. >>> The java application is going to access the serial port and uses the >>> RXTX Java library to do this. The RXTX jar is included in the .snap. >>> >>> When it runs I get the following error : >>> >>> check_group_uucp(): error testing lock file creation Error >>> details:Permission deniedcheck_lock_status: No permission to create lock >>> file. >>> please see: How can I use Lock Files with rxtx? in INSTALL >>> >>> >>> Anyone have any experience of this ? >>> >>> Thanks in advance. >>> Jenny >>> >>> -- >>> *Jenny Murphy* >>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>> jenny.murphy at episensor.com t | +353 (0) >>> 61 512 511 w | http://www.episensor.com >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>> an/listinfo/snapcraft >>> >>> >> >> >> -- >> - Loïc >> > > > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | +353 (0) 61 > 512 511 w | http://www.episensor.com > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Wed Sep 7 13:42:13 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 7 Sep 2016 09:42:13 -0400 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: On 07/09/16 06:56, MikeB wrote: > This new 'first user experience' caught me by surprise - I don't think > I saw anything about this on the snapcraft mailing list. Is there > some other list I'm not following that discusses these changes ahead > of time? Do we need a list for Ubuntu Core, the all-snap rendition of Ubuntu? This list is mainly for people making snaps, which might be running on any classic distro, and I would not want folks who's primary interest is a snap running on their preferred distro to have to wade through threads about Ubuntu Core. Mark -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Wed Sep 7 13:54:24 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 7 Sep 2016 10:54:24 -0300 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: Rather than splitting Ubuntu Core, I suggest splitting device building. Ubuntu Core is just another distribution using snaps, so discussing the well-being of snaps and snapd inside it sounds like a welcome topic here. Device building and image tuning is a world on itself, though. On Wed, Sep 7, 2016 at 10:42 AM, Mark Shuttleworth wrote: > On 07/09/16 06:56, MikeB wrote: > > This new 'first user experience' caught me by surprise - I don't think I > saw anything about this on the snapcraft mailing list. Is there some other > list I'm not following that discusses these changes ahead of time? > > > Do we need a list for Ubuntu Core, the all-snap rendition of Ubuntu? This > list is mainly for people making snaps, which might be running on any > classic distro, and I would not want folks who's primary interest is a snap > running on their preferred distro to have to wade through threads about > Ubuntu Core. > > Mark > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Wed Sep 7 14:00:03 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 7 Sep 2016 15:00:03 +0100 Subject: Access denied to /dev/ttyUSB0 Message-ID: Hi, I solved the previous issue regarding the use of the RXTX java library and the application is loading and running without errors. However my java code isn't detecting the port, it should be able to see /dev/ttyUSB0. In the dmesg log I see the following errors : [101043.875139] usb 1-2: new full-speed USB device number 7 using xhci_hcd [101044.006418] usb 1-2: New USB device found, idVendor=10c4, idProduct=ea60 [101044.006441] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [101044.006456] usb 1-2: Product: CP2102 USB to UART Bridge Controller [101044.006470] usb 1-2: Manufacturer: Silicon Labs [101044.006482] usb 1-2: SerialNumber: 0001 [101044.011589] cp210x 1-2:1.0: cp210x converter detected [101044.012037] usb 1-2: cp210x converter now attached to ttyUSB0 [101098.191960] audit_printk_skb: 258 callbacks suppressed [101098.191979] audit: type=1400 audit(1473254849.416:69078): apparmor="DENIED" operation="open" profile="gateway.sideload_main_IdNRFcRcGPGe" name="/dev/ttyUSB0" pid=4875 comm="java" requested_mask="wr" denied_mask="wr" fsuid=0 ouid=0 Is there anyway to solve this. I am working on a Ubuntu 15.04 platform. Thanks again. -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Wed Sep 7 14:01:23 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Wed, 7 Sep 2016 16:01:23 +0200 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Hi, I'm guessing the Java version change solved the rebuilding problem? Cheers, - Loïc Minier On Wed, Sep 7, 2016 at 3:32 PM, Jenny Murphy wrote: > Hi all, > This is solved . The problem was to do with the Java versions supported > by RXTX. > Jenny > > On 7 September 2016 at 12:51, Jenny Murphy > wrote: > >> Hi, >> I am having some difficulty building RXTX (get unsatisfied link errors >> with the new shared objects and jar file). >> Is there an other alternative to workaround my original problem with the >> permissions to use lock file? >> Thanks. >> >> >> On 6 September 2016 at 12:48, Loïc Minier wrote: >> >>> Hi Jenny, >>> >>> The INSTALL file >>> suggests that RXTX implements custom locking which requires OS integration; >>> because you want your app to be the only one accessing the serial port(s), >>> you should build RXTX with --disable-lockfiles to disable this feature of >>> the library. >>> >>> Cheers, >>> - Loïc >>> >>> On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy >> > wrote: >>> >>>> Hi, >>>> >>>> I have a java application running in a .snap package. >>>> The java application is going to access the serial port and uses the >>>> RXTX Java library to do this. The RXTX jar is included in the .snap. >>>> >>>> When it runs I get the following error : >>>> >>>> check_group_uucp(): error testing lock file creation Error >>>> details:Permission deniedcheck_lock_status: No permission to create lock >>>> file. >>>> please see: How can I use Lock Files with rxtx? in INSTALL >>>> >>>> >>>> Anyone have any experience of this ? >>>> >>>> Thanks in advance. >>>> Jenny >>>> >>>> -- >>>> *Jenny Murphy* >>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>> jenny.murphy at episensor.com t | +353 (0) >>>> 61 512 511 w | http://www.episensor.com >>>> >>>> -- >>>> Snapcraft mailing list >>>> Snapcraft at lists.snapcraft.io >>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>>> an/listinfo/snapcraft >>>> >>>> >>> >>> >>> -- >>> - Loïc >>> >> >> >> >> -- >> *Jenny Murphy* >> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >> jenny.murphy at episensor.com t | +353 (0) 61 >> 512 511 w | http://www.episensor.com >> > > > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | +353 (0) 61 > 512 511 w | http://www.episensor.com > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Wed Sep 7 14:04:40 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 7 Sep 2016 15:04:40 +0100 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Hi, No, actually the configure file in RXTX needed to be modified to support java 1.7 : case $JAVA_VERSION in 1.2*|1.3*|1.4*|1.5*) #fix_parameters $JPATH/jre/lib/javax.comm.properties CLASSPATH=".:\$(TOP):\$(TOP)/src:"`find $JPATH/ -name RXTXcomm.jar |head -n1` RXTX_PATH="\$(JPATH)/jre/lib/\$(OS_ARCH)" JHOME=$JPATH/"jre/lib/ext" So the paths were not being set up correctly for me. I changed it to case $JAVA_VERSION in 1.2*|1.3*|1.4*|1.5*|1.7*) Thanks a million for your suggestion. Jenny On 7 September 2016 at 15:01, Loïc Minier wrote: > Hi, > > I'm guessing the Java version change solved the rebuilding problem? > > Cheers, > - Loïc Minier > > On Wed, Sep 7, 2016 at 3:32 PM, Jenny Murphy > wrote: > >> Hi all, >> This is solved . The problem was to do with the Java versions supported >> by RXTX. >> Jenny >> >> On 7 September 2016 at 12:51, Jenny Murphy >> wrote: >> >>> Hi, >>> I am having some difficulty building RXTX (get unsatisfied link errors >>> with the new shared objects and jar file). >>> Is there an other alternative to workaround my original problem with the >>> permissions to use lock file? >>> Thanks. >>> >>> >>> On 6 September 2016 at 12:48, Loïc Minier >>> wrote: >>> >>>> Hi Jenny, >>>> >>>> The INSTALL file >>>> suggests that >>>> RXTX implements custom locking which requires OS integration; because you >>>> want your app to be the only one accessing the serial port(s), you should >>>> build RXTX with --disable-lockfiles to disable this feature of the library. >>>> >>>> Cheers, >>>> - Loïc >>>> >>>> On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy < >>>> jenny.murphy at episensor.com> wrote: >>>> >>>>> Hi, >>>>> >>>>> I have a java application running in a .snap package. >>>>> The java application is going to access the serial port and uses the >>>>> RXTX Java library to do this. The RXTX jar is included in the .snap. >>>>> >>>>> When it runs I get the following error : >>>>> >>>>> check_group_uucp(): error testing lock file creation Error >>>>> details:Permission deniedcheck_lock_status: No permission to create lock >>>>> file. >>>>> please see: How can I use Lock Files with rxtx? in INSTALL >>>>> >>>>> >>>>> Anyone have any experience of this ? >>>>> >>>>> Thanks in advance. >>>>> Jenny >>>>> >>>>> -- >>>>> *Jenny Murphy* >>>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>>> jenny.murphy at episensor.com t | +353 (0) >>>>> 61 512 511 w | http://www.episensor.com >>>>> >>>>> -- >>>>> Snapcraft mailing list >>>>> Snapcraft at lists.snapcraft.io >>>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>>>> an/listinfo/snapcraft >>>>> >>>>> >>>> >>>> >>>> -- >>>> - Loïc >>>> >>> >>> >>> >>> -- >>> *Jenny Murphy* >>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>> jenny.murphy at episensor.com t | +353 (0) >>> 61 512 511 w | http://www.episensor.com >>> >> >> >> >> -- >> *Jenny Murphy* >> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >> jenny.murphy at episensor.com t | +353 (0) 61 >> 512 511 w | http://www.episensor.com >> > > > > -- > - Loïc > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Wed Sep 7 14:05:48 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Wed, 7 Sep 2016 16:05:48 +0200 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: Message-ID: Hi, You need to use: plugs: [serial-port] In your snapcraft.yaml; then you also have to connect your snap to that interface manually as it's not autoconnected; check "snap interfaces" to see which plugs and slots are connected. Cheers, - Loïc Minier On Wed, Sep 7, 2016 at 4:00 PM, Jenny Murphy wrote: > Hi, > I solved the previous issue regarding the use of the RXTX java library > and the application is loading and running without errors. > However my java code isn't detecting the port, it should be able to see > /dev/ttyUSB0. > > In the dmesg log I see the following errors : > > [101043.875139] usb 1-2: new full-speed USB device number 7 using xhci_hcd > [101044.006418] usb 1-2: New USB device found, idVendor=10c4, > idProduct=ea60 > [101044.006441] usb 1-2: New USB device strings: Mfr=1, Product=2, > SerialNumber=3 > [101044.006456] usb 1-2: Product: CP2102 USB to UART Bridge Controller > [101044.006470] usb 1-2: Manufacturer: Silicon Labs > [101044.006482] usb 1-2: SerialNumber: 0001 > [101044.011589] cp210x 1-2:1.0: cp210x converter detected > [101044.012037] usb 1-2: cp210x converter now attached to ttyUSB0 > > [101098.191960] audit_printk_skb: 258 callbacks suppressed > > [101098.191979] audit: type=1400 audit(1473254849.416:69078): > apparmor="DENIED" operation="open" profile="gateway.sideload_main_IdNRFcRcGPGe" > name="/dev/ttyUSB0" pid=4875 comm="java" requested_mask="wr" > denied_mask="wr" fsuid=0 ouid=0 > > > Is there anyway to solve this. I am working on a Ubuntu 15.04 platform. > > Thanks again. > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | +353 (0) 61 > 512 511 w | http://www.episensor.com > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Wed Sep 7 14:06:41 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Wed, 7 Sep 2016 16:06:41 +0200 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Great; now we even have google juice for other people trying to use RXTX with snaps, thanks! :-) On Wed, Sep 7, 2016 at 4:04 PM, Jenny Murphy wrote: > Hi, > No, actually the configure file in RXTX needed to be modified to support > java 1.7 : > > > case $JAVA_VERSION in > 1.2*|1.3*|1.4*|1.5*) > #fix_parameters $JPATH/jre/lib/javax.comm.properties > CLASSPATH=".:\$(TOP):\$(TOP)/src:"`find $JPATH/ -name RXTXcomm.jar |head > -n1` > RXTX_PATH="\$(JPATH)/jre/lib/\$(OS_ARCH)" > JHOME=$JPATH/"jre/lib/ext" > > So the paths were not being set up correctly for me. > > I changed it to > case $JAVA_VERSION in > 1.2*|1.3*|1.4*|1.5*|1.7*) > > Thanks a million for your suggestion. > > Jenny > > > On 7 September 2016 at 15:01, Loïc Minier wrote: > >> Hi, >> >> I'm guessing the Java version change solved the rebuilding problem? >> >> Cheers, >> - Loïc Minier >> >> On Wed, Sep 7, 2016 at 3:32 PM, Jenny Murphy >> wrote: >> >>> Hi all, >>> This is solved . The problem was to do with the Java versions supported >>> by RXTX. >>> Jenny >>> >>> On 7 September 2016 at 12:51, Jenny Murphy >>> wrote: >>> >>>> Hi, >>>> I am having some difficulty building RXTX (get unsatisfied link errors >>>> with the new shared objects and jar file). >>>> Is there an other alternative to workaround my original problem with >>>> the permissions to use lock file? >>>> Thanks. >>>> >>>> >>>> On 6 September 2016 at 12:48, Loïc Minier >>>> wrote: >>>> >>>>> Hi Jenny, >>>>> >>>>> The INSTALL file >>>>> suggests that >>>>> RXTX implements custom locking which requires OS integration; because you >>>>> want your app to be the only one accessing the serial port(s), you should >>>>> build RXTX with --disable-lockfiles to disable this feature of the library. >>>>> >>>>> Cheers, >>>>> - Loïc >>>>> >>>>> On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy < >>>>> jenny.murphy at episensor.com> wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> I have a java application running in a .snap package. >>>>>> The java application is going to access the serial port and uses the >>>>>> RXTX Java library to do this. The RXTX jar is included in the .snap. >>>>>> >>>>>> When it runs I get the following error : >>>>>> >>>>>> check_group_uucp(): error testing lock file creation Error >>>>>> details:Permission deniedcheck_lock_status: No permission to create lock >>>>>> file. >>>>>> please see: How can I use Lock Files with rxtx? in INSTALL >>>>>> >>>>>> >>>>>> Anyone have any experience of this ? >>>>>> >>>>>> Thanks in advance. >>>>>> Jenny >>>>>> >>>>>> -- >>>>>> *Jenny Murphy* >>>>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>>>> jenny.murphy at episensor.com t | +353 >>>>>> (0) 61 512 511 w | http://www.episensor.com >>>>>> >>>>>> -- >>>>>> Snapcraft mailing list >>>>>> Snapcraft at lists.snapcraft.io >>>>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>>>>> an/listinfo/snapcraft >>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> - Loïc >>>>> >>>> >>>> >>>> >>>> -- >>>> *Jenny Murphy* >>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>> jenny.murphy at episensor.com t | +353 (0) >>>> 61 512 511 w | http://www.episensor.com >>>> >>> >>> >>> >>> -- >>> *Jenny Murphy* >>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>> jenny.murphy at episensor.com t | +353 (0) >>> 61 512 511 w | http://www.episensor.com >>> >> >> >> >> -- >> - Loïc >> > > > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | +353 (0) 61 > 512 511 w | http://www.episensor.com > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From zygmunt.krynicki at canonical.com Wed Sep 7 14:20:00 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 7 Sep 2016 16:20:00 +0200 Subject: /snap/bin not in $PATH on desktop In-Reply-To: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> References: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> Message-ID: <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> > On 7 Sep 2016, at 13:16, Sylvain Pineau wrote: > > Hello, > > I noticed that I was not able to call other snap commands from my own snap on my desktop. > So I tested what was defined using the hello-world.env command. > > On desktop (with ubuntu-core 16.04.1 rev 352) > > $ hello-world.env | grep ^PATH= > PATH=/usr/sbin:/usr/bin:/sbin:/bin:/usr/games > > But on a true snappy system (with ubuntu-core 16.04.1 rev 453), I get: > > $ hello-world.env | grep ^PATH= > PATH=/home/ubuntu/bin:/home/ubuntu/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin > > Is there any reason to not have /snap/bin as part of the $PATH available to snap commands? Snaps cannot execute other snaps. The PATH difference is caused by how snap-confine behaves when it runs on classic but in general even if you used an absolute path you would not be able to start any other applications from /snap/bin. Allowing this would create an implicit interface (dependency between your snap and some other, perhaps third party, snap). If you need access to executables that you control you can use the content interface to bind mount another snap into your own snap and execute those commands directly, with the same security profile as the running application. Best regards ZK From zygmunt.krynicki at canonical.com Wed Sep 7 14:24:23 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 7 Sep 2016 16:24:23 +0200 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: Message-ID: <373ED355-C86E-4584-9CB0-F26DF7C3D6B5@canonical.com> > On 7 Sep 2016, at 16:05, Loïc Minier wrote: > > Hi, > > You need to use: > plugs: [serial-port] > I don’t think this will work as this is a 15.04 based system. Jenny, can you confirm that you are using Ubuntu Core 15.04? Best regards ZK From simon.fels at canonical.com Wed Sep 7 14:26:52 2016 From: simon.fels at canonical.com (Simon Fels) Date: Wed, 7 Sep 2016 15:26:52 +0100 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: Message-ID: On 07.09.2016 15:05, Loïc Minier wrote: > Hi, > > You need to use: > > plugs: [serial-port] > > In your snapcraft.yaml; then you also have to connect your snap to that > interface manually as it's not autoconnected; check "snap interfaces" to > see which plugs and slots are connected. That doesn't work anymore. There is no implicit slot for serial-port. Right now the only option to get access to serial ports is via defining a slot on a gadget snap with either a USB product/vendor id pair or a absolute path to the serial node which then the snap can connect to and it will only get access to that particular serial port. You can put something like slots: my-usb-serial-device: interface: serial-port product-id: 0x0 vendor-id: 0x0 into your gadget snap and then connect both the snap and gadget together so your app gets access to that serial node. In a system where no gadget exists you currently don't get access to serial nodes. We need to implement hotplut support to make this slots automatically exported for those devices. Not sure for when this is on the list. regards, Simon > > > On Wed, Sep 7, 2016 at 4:00 PM, Jenny Murphy > wrote: > > Hi, > I solved the previous issue regarding the use of the RXTX java > library and the application is loading and running without errors. > However my java code isn't detecting the port, it should be able to > see /dev/ttyUSB0. > > In the dmesg log I see the following errors : > > [101043.875139] usb 1-2: new full-speed USB device number 7 using > xhci_hcd > [101044.006418] usb 1-2: New USB device found, idVendor=10c4, > idProduct=ea60 > [101044.006441] usb 1-2: New USB device strings: Mfr=1, Product=2, > SerialNumber=3 > [101044.006456] usb 1-2: Product: CP2102 USB to UART Bridge Controller > [101044.006470] usb 1-2: Manufacturer: Silicon Labs > [101044.006482] usb 1-2: SerialNumber: 0001 > [101044.011589] cp210x 1-2:1.0: cp210x converter detected > [101044.012037] usb 1-2: cp210x converter now attached to ttyUSB0 > > [101098.191960] audit_printk_skb: 258 callbacks suppressed > > [101098.191979] audit: type=1400 audit(1473254849.416:69078): > apparmor="DENIED" operation="open" > profile="gateway.sideload_main_IdNRFcRcGPGe" name="/dev/ttyUSB0" > pid=4875 comm="java" requested_mask="wr" denied_mask="wr" fsuid=0 ouid=0 > > > Is there anyway to solve this. I am working on a Ubuntu 15.04 platform. > > Thanks again. > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | > +353 (0) 61 512 511 w | > http://www.episensor.com > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > -- > - Loïc > > From jenny.murphy at episensor.com Wed Sep 7 14:41:10 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 7 Sep 2016 15:41:10 +0100 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: <373ED355-C86E-4584-9CB0-F26DF7C3D6B5@canonical.com> References: <373ED355-C86E-4584-9CB0-F26DF7C3D6B5@canonical.com> Message-ID: Yes I am using Ubuntu 15.04 On 7 Sep 2016 15:24, "Zygmunt Krynicki" wrote: > > > On 7 Sep 2016, at 16:05, Loïc Minier wrote: > > > > Hi, > > > > You need to use: > > plugs: [serial-port] > > > > I don’t think this will work as this is a 15.04 based system. > > Jenny, can you confirm that you are using Ubuntu Core 15.04? > > Best regards > ZK > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Wed Sep 7 15:14:54 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Wed, 7 Sep 2016 17:14:54 +0200 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: <373ED355-C86E-4584-9CB0-F26DF7C3D6B5@canonical.com> Message-ID: Ok, then this should work: sudo snappy hw-assign yoursnap /dev/ttyUSB0 On Wed, Sep 7, 2016 at 4:41 PM, Jenny Murphy wrote: > Yes I am using Ubuntu 15.04 > On 7 Sep 2016 15:24, "Zygmunt Krynicki" > wrote: > >> >> > On 7 Sep 2016, at 16:05, Loïc Minier wrote: >> > >> > Hi, >> > >> > You need to use: >> > plugs: [serial-port] >> > >> >> I don’t think this will work as this is a 15.04 based system. >> >> Jenny, can you confirm that you are using Ubuntu Core 15.04? >> >> Best regards >> ZK >> >> -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From matthew.williams at canonical.com Wed Sep 7 16:32:45 2016 From: matthew.williams at canonical.com (Matthew Williams) Date: Wed, 7 Sep 2016 17:32:45 +0100 Subject: Access to other commands Message-ID: Hi folks, I have a few applications I'm looking to snap, but quite a lot of them shell out to other commands at some point and it's not always practical to include these with my snap, one concrete example is shelling out to /usr/bin/sensible-browser. My questions are: 1) Is there some way I can be specifying a list of commands my snapped app is allowed to call? 2) Should I really be trying to bundle these commands into my snap? Bonus question: Does the fact that I'm asking this question mean there is something fundamental about snaps that I'm not grokking? Cheers Matty -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Wed Sep 7 19:30:11 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 7 Sep 2016 20:30:11 +0100 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: <373ED355-C86E-4584-9CB0-F26DF7C3D6B5@canonical.com> Message-ID: Hi, Thats works great, thanks a million. On 7 September 2016 at 16:14, Loïc Minier wrote: > Ok, then this should work: > sudo snappy hw-assign yoursnap /dev/ttyUSB0 > > > On Wed, Sep 7, 2016 at 4:41 PM, Jenny Murphy > wrote: > >> Yes I am using Ubuntu 15.04 >> On 7 Sep 2016 15:24, "Zygmunt Krynicki" >> wrote: >> >>> >>> > On 7 Sep 2016, at 16:05, Loïc Minier wrote: >>> > >>> > Hi, >>> > >>> > You need to use: >>> > plugs: [serial-port] >>> > >>> >>> I don’t think this will work as this is a 15.04 based system. >>> >>> Jenny, can you confirm that you are using Ubuntu Core 15.04? >>> >>> Best regards >>> ZK >>> >>> > > > -- > - Loïc > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From michael.vogt at canonical.com Wed Sep 7 21:50:54 2016 From: michael.vogt at canonical.com (Michael Vogt) Date: Wed, 7 Sep 2016 23:50:54 +0200 Subject: Ubuntu Core 16 beta images available Message-ID: <20160907215054.GG3871@bod> Ubuntu Core 16 Images ===================== The Ubuntu snappy team is happy to announce the first beta images for Ubuntu Core 16. The images use the snapd package manager to install and update all components of the system including kernel, core, gadget and applications. The images are available for PC (amd64, i386) and Raspberry Pi2 (armhf). More architectures and boards (arm64 dragonboard, pi3) will follow shortly. You can download them at: http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ The images are bootable, the pc image can be booted directly in qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: unxz ubuntu-core-16-pc.img.xz dd if= ubuntu-core-16-pc.img of=/dev/sdXX Where /dev/sdXX is the path of your sd card. After booting the image you can enter your Ubuntu SSO email and it will automatically create a matching user with the right ssh keys. If you do not have an Ubuntu SSO account yet you can create one at: https://login.ubuntu.com/ These images follow the "beta" channel. Enjoy the fresh images! If you find any bugs or issues, please let us know via: https://bugs.launchpad.net/snappy/ Cheers, Michael (on behalf of the snappy team) From gustavo.niemeyer at canonical.com Wed Sep 7 23:00:09 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 7 Sep 2016 20:00:09 -0300 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160907215054.GG3871@bod> References: <20160907215054.GG3871@bod> Message-ID: Thanks so much and congratulations to everybody who worked hard to put in place this initial public release of Ubuntu Core, a minimalist Linux distribution that uses snaps as its building blocks. Special thanks as well to Michael, Oliver Grawert, Martin Pitt, and others who've put all the bits in right offsets to have a working image today. Over the next few weeks we'll be including a small set of pending features and polishing this image until we have a stable golden release. Please remember that despite some of those conversations being about Ubuntu Core, at its core we have snaps and snapd still. The vast majority of features we talk about for Ubuntu Core benefits all the distributions in which snapd works today. Hold on tight and join us in this exciting trip. On Wed, Sep 7, 2016 at 6:50 PM, Michael Vogt wrote: > Ubuntu Core 16 Images > ===================== > > The Ubuntu snappy team is happy to announce the first beta images for > Ubuntu Core 16. The images use the snapd package manager to install > and update all components of the system including kernel, core, gadget > and applications. > > The images are available for PC (amd64, i386) and Raspberry Pi2 > (armhf). More architectures and boards (arm64 dragonboard, pi3) will > follow shortly. You can download them at: > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > > The images are bootable, the pc image can be booted directly in > qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: > > unxz ubuntu-core-16-pc.img.xz > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > Where /dev/sdXX is the path of your sd card. > > After booting the image you can enter your Ubuntu SSO email and it > will automatically create a matching user with the right ssh keys. If > you do not have an Ubuntu SSO account yet you can create one at: > > https://login.ubuntu.com/ > > These images follow the "beta" channel. > > Enjoy the fresh images! If you find any bugs or issues, please let us > know via: > > https://bugs.launchpad.net/snappy/ > > Cheers, > Michael (on behalf of the snappy team) > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.wakeling at webdrake.net Wed Sep 7 23:10:06 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Thu, 8 Sep 2016 01:10:06 +0200 Subject: Working LDC snap [was: Re: Snapping LDC (LLVM-based D compiler)] In-Reply-To: <17152bf2-210c-f0d3-6c69-5cb0e9bf0300@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> <17152bf2-210c-f0d3-6c69-5cb0e9bf0300@webdrake.net> Message-ID: On 05/09/16 00:25, Joseph Rushton Wakeling wrote: > One problem fixed (the inclusion of `build-essential` as a build dependency > ensures the snap will build in a `cleanbuild` environment), but one remains: it > looks like the `ldc-config` part (which manually replaces the incorrectly > auto-generated ldc2.conf with a correct alternative, using the `dump` plugin) > can sometimes fail. > > There doesn't seem to be any reason for it, just with some builds I end up with > a wrong (auto-generated) ldc2.conf in the snap package, and some I end up with > the correct one copied from the ldc-config directory. > > Any ideas what's up? I'm guessing bug either with the dump plugin or with the > handling of the `snap` section of the `ldc` part ... ? Looks like it because of only filtering out the ldc part's etc/ldc2.conf file at the `snap` step. This allowed a race condition where if ldc was staged after ldc-config, its etc/ldc2.conf would overwrite the desired one, and the `snap` filter was powerless to distinguish this. Replacing the `snap` exclusion filter with a `stage` exclusion filter solved things: parts: ldc: source: git://github.com/ldc-developers/ldc.git source-tag: v1.0.0 plugin: cmake stage: - -etc/ldc2.conf build-packages: - build-essential - ldc - llvm-dev - libconfig++-dev - libcurl4-gnutls-dev - libedit-dev - zlib1g-dev ldc-config: plugin: dump source: ldc-config organize: ldc2.conf: etc/ldc2.conf From ehbello at gmail.com Wed Sep 7 23:29:03 2016 From: ehbello at gmail.com (=?UTF-8?Q?Enrique_Hern=C3=A1ndez_Bello?=) Date: Thu, 8 Sep 2016 00:29:03 +0100 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: Message-ID: Hello, taking advantage of this thread, what is the right way to grant access of a /dev/ttyS0 to our snap in Ubuntu Core 16? Regards. 2016-09-07 15:26 GMT+01:00 Simon Fels : > On 07.09.2016 15:05, Loïc Minier wrote: >> Hi, >> >> You need to use: >> >> plugs: [serial-port] >> >> In your snapcraft.yaml; then you also have to connect your snap to that >> interface manually as it's not autoconnected; check "snap interfaces" to >> see which plugs and slots are connected. > > That doesn't work anymore. There is no implicit slot for serial-port. > Right now the only option to get access to serial ports is via defining > a slot on a gadget snap with either a USB product/vendor id pair or a > absolute path to the serial node which then the snap can connect to and > it will only get access to that particular serial port. > > You can put something like > > slots: > my-usb-serial-device: > interface: serial-port > product-id: 0x0 > vendor-id: 0x0 > > into your gadget snap and then connect both the snap and gadget together > so your app gets access to that serial node. > > In a system where no gadget exists you currently don't get access to > serial nodes. We need to implement hotplut support to make this slots > automatically exported for those devices. Not sure for when this is on > the list. > > regards, > Simon > >> >> >> On Wed, Sep 7, 2016 at 4:00 PM, Jenny Murphy > > wrote: >> >> Hi, >> I solved the previous issue regarding the use of the RXTX java >> library and the application is loading and running without errors. >> However my java code isn't detecting the port, it should be able to >> see /dev/ttyUSB0. >> >> In the dmesg log I see the following errors : >> >> [101043.875139] usb 1-2: new full-speed USB device number 7 using >> xhci_hcd >> [101044.006418] usb 1-2: New USB device found, idVendor=10c4, >> idProduct=ea60 >> [101044.006441] usb 1-2: New USB device strings: Mfr=1, Product=2, >> SerialNumber=3 >> [101044.006456] usb 1-2: Product: CP2102 USB to UART Bridge Controller >> [101044.006470] usb 1-2: Manufacturer: Silicon Labs >> [101044.006482] usb 1-2: SerialNumber: 0001 >> [101044.011589] cp210x 1-2:1.0: cp210x converter detected >> [101044.012037] usb 1-2: cp210x converter now attached to ttyUSB0 >> >> [101098.191960] audit_printk_skb: 258 callbacks suppressed >> >> [101098.191979] audit: type=1400 audit(1473254849.416:69078): >> apparmor="DENIED" operation="open" >> profile="gateway.sideload_main_IdNRFcRcGPGe" name="/dev/ttyUSB0" >> pid=4875 comm="java" requested_mask="wr" denied_mask="wr" fsuid=0 ouid=0 >> >> >> Is there anyway to solve this. I am working on a Ubuntu 15.04 platform. >> >> Thanks again. >> >> -- >> *Jenny Murphy* >> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >> jenny.murphy at episensor.com t | >> +353 (0) 61 512 511 w | >> http://www.episensor.com >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> >> >> >> >> -- >> - Loïc >> >> > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From gustavo.niemeyer at canonical.com Wed Sep 7 23:34:14 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 7 Sep 2016 20:34:14 -0300 Subject: Access denied to /dev/ttyUSB0 In-Reply-To: References: Message-ID: We've just been polishing this process in the last couple of weeks. You just need to ask for a serial-port plug in your snap. There's a step missing for that to work well, but that's on our end. We need to expose serial-port slots in the gadget or core for serial-port-0 (S0) and serial-port-1 (S1) if those are available. In the near future, these slots will show up automatically based on hotplug support. Then, you can run something like: snap connect yoursnap:serial-port ubuntu-core:serial-port-1 On Wed, Sep 7, 2016 at 8:29 PM, Enrique Hernández Bello wrote: > Hello, > > taking advantage of this thread, what is the right way to grant access > of a /dev/ttyS0 to our snap in Ubuntu Core 16? > > Regards. > > 2016-09-07 15:26 GMT+01:00 Simon Fels : > > On 07.09.2016 15:05, Loïc Minier wrote: > >> Hi, > >> > >> You need to use: > >> > >> plugs: [serial-port] > >> > >> In your snapcraft.yaml; then you also have to connect your snap to that > >> interface manually as it's not autoconnected; check "snap interfaces" to > >> see which plugs and slots are connected. > > > > That doesn't work anymore. There is no implicit slot for serial-port. > > Right now the only option to get access to serial ports is via defining > > a slot on a gadget snap with either a USB product/vendor id pair or a > > absolute path to the serial node which then the snap can connect to and > > it will only get access to that particular serial port. > > > > You can put something like > > > > slots: > > my-usb-serial-device: > > interface: serial-port > > product-id: 0x0 > > vendor-id: 0x0 > > > > into your gadget snap and then connect both the snap and gadget together > > so your app gets access to that serial node. > > > > In a system where no gadget exists you currently don't get access to > > serial nodes. We need to implement hotplut support to make this slots > > automatically exported for those devices. Not sure for when this is on > > the list. > > > > regards, > > Simon > > > >> > >> > >> On Wed, Sep 7, 2016 at 4:00 PM, Jenny Murphy < > jenny.murphy at episensor.com > >> > wrote: > >> > >> Hi, > >> I solved the previous issue regarding the use of the RXTX java > >> library and the application is loading and running without errors. > >> However my java code isn't detecting the port, it should be able to > >> see /dev/ttyUSB0. > >> > >> In the dmesg log I see the following errors : > >> > >> [101043.875139] usb 1-2: new full-speed USB device number 7 using > >> xhci_hcd > >> [101044.006418] usb 1-2: New USB device found, idVendor=10c4, > >> idProduct=ea60 > >> [101044.006441] usb 1-2: New USB device strings: Mfr=1, Product=2, > >> SerialNumber=3 > >> [101044.006456] usb 1-2: Product: CP2102 USB to UART Bridge > Controller > >> [101044.006470] usb 1-2: Manufacturer: Silicon Labs > >> [101044.006482] usb 1-2: SerialNumber: 0001 > >> [101044.011589] cp210x 1-2:1.0: cp210x converter detected > >> [101044.012037] usb 1-2: cp210x converter now attached to ttyUSB0 > >> > >> [101098.191960] audit_printk_skb: 258 callbacks suppressed > >> > >> [101098.191979] audit: type=1400 audit(1473254849.416:69078): > >> apparmor="DENIED" operation="open" > >> profile="gateway.sideload_main_IdNRFcRcGPGe" name="/dev/ttyUSB0" > >> pid=4875 comm="java" requested_mask="wr" denied_mask="wr" fsuid=0 > ouid=0 > >> > >> > >> Is there anyway to solve this. I am working on a Ubuntu 15.04 > platform. > >> > >> Thanks again. > >> > >> -- > >> *Jenny Murphy* > >> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > >> jenny.murphy at episensor.com t | > >> +353 (0) 61 512 511 w | > >> http://www.episensor.com > >> > >> -- > >> Snapcraft mailing list > >> Snapcraft at lists.snapcraft.io > >> Modify settings or unsubscribe at: > >> https://lists.ubuntu.com/mailman/listinfo/snapcraft > >> > >> > >> > >> > >> > >> -- > >> - Loïc > >> > >> > > > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Thu Sep 8 00:42:21 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 7 Sep 2016 21:42:21 -0300 Subject: New spread, new features, new 14.04 image Message-ID: Hello all, Tomorrow there's a new spread coming with a couple of changes, one of them incompatible on the adhoc backend. Before jumping into those, please note there's a new image available for our tests: ubuntu-14.04-64 The first change is on using images that require username/password. That's often required to test setups with custom images more easily (anyone said Ubuntu Core? :). Spread used to login into the image as the provided user and then use sudo to setup root access via ssh. Now spread won't touch the system configuration in those cases and will instead use sudo itself to run all test scripts. This second one breaks compatibility with the previous adhoc backend, so please read through if you're using it. The AdHoc backend has changed the way it communicates information back to Spread on allocation and discarding of systems to be more flexible. Now the address allocated is reported with: ADDRESS addr[:port] and you can exit with a nice error message with one of: ERROR $msg FATAL $msg That's supported in any of the scripts (so far we have two: allocate, and discard). More details on both changes is available at: https://github.com/snapcore/spread#adhoc https://github.com/snapcore/spread#passwords These features are available on the master branch, but not yet release to Travis or to the spread snap. I'll update those tomorrow during my day so I'm around to sort any issues. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Thu Sep 8 01:31:36 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 7 Sep 2016 21:31:36 -0400 Subject: Access to other commands In-Reply-To: References: Message-ID: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> On 07/09/16 12:32, Matthew Williams wrote: > I have a few applications I'm looking to snap, but quite a lot of them > shell out to other commands at some point and it's not always > practical to include these with my snap, one concrete example is > shelling out to /usr/bin/sensible-browser. Something like sensible-browser is probably an interface we should have. > > My questions are: > > 1) Is there some way I can be specifying a list of commands my snapped > app is allowed to call? You could. But I think there is a class of things that should be allowed to integrate with the classic shell environment, which means they can shell out to lots of things. > 2) Should I really be trying to bundle these commands into my snap? Only if it is a tight and bounded set. > Bonus question: Does the fact that I'm asking this question mean there > is something fundamental about snaps that I'm not grokking? No, we are teasing around the edges of this problem, and your comments are a useful extra data point. Thank you! Mark From alejandro.vera at gmail.com Thu Sep 8 01:33:43 2016 From: alejandro.vera at gmail.com (Alejandro Vera) Date: Wed, 7 Sep 2016 22:33:43 -0300 Subject: /snap/bin not in $PATH on desktop In-Reply-To: <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> References: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> Message-ID: So how do you resolve the problem of using multiple apps? I am developing a code editor. But i need a code beautifier. How can I call a external beautifier from my snap? On Wed, Sep 7, 2016 at 11:20 AM, Zygmunt Krynicki < zygmunt.krynicki at canonical.com> wrote: > > > On 7 Sep 2016, at 13:16, Sylvain Pineau > wrote: > > > > Hello, > > > > I noticed that I was not able to call other snap commands from my own > snap on my desktop. > > So I tested what was defined using the hello-world.env command. > > > > On desktop (with ubuntu-core 16.04.1 rev 352) > > > > $ hello-world.env | grep ^PATH= > > PATH=/usr/sbin:/usr/bin:/sbin:/bin:/usr/games > > > > But on a true snappy system (with ubuntu-core 16.04.1 rev 453), I get: > > > > $ hello-world.env | grep ^PATH= > > PATH=/home/ubuntu/bin:/home/ubuntu/.local/bin:/usr/local/ > sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/ > games:/usr/local/games:/snap/bin > > > > Is there any reason to not have /snap/bin as part of the $PATH available > to snap commands? > > Snaps cannot execute other snaps. The PATH difference is caused by how > snap-confine behaves when it runs on classic but in general even if you > used an absolute path you would not be able to start any other applications > from /snap/bin. Allowing this would create an implicit interface > (dependency between your snap and some other, perhaps third party, snap). > > If you need access to executables that you control you can use the content > interface to bind mount another snap into your own snap and execute those > commands directly, with the same security profile as the running > application. > > Best regards > ZK > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- Alejandro Vera http://www.recicleta.cl -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Thu Sep 8 02:11:15 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 7 Sep 2016 23:11:15 -0300 Subject: Access to other commands In-Reply-To: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> Message-ID: On Wed, Sep 7, 2016 at 10:31 PM, Mark Shuttleworth wrote: > On 07/09/16 12:32, Matthew Williams wrote: > > I have a few applications I'm looking to snap, but quite a lot of them > > shell out to other commands at some point and it's not always > > practical to include these with my snap, one concrete example is > > shelling out to /usr/bin/sensible-browser. > We have something slightly better than this. Any snap can call xdg-open $url and have that URL being sent to a safe launcher over dbus. This is the exact line: bus-send --print-reply --session --dest=com.canonical.SafeLauncher / com.canonical.SafeLauncher.OpenURL string:"$1" Because this is always safe to do and offered over a common mechanism, we don't need an interface to mediate it. If something is listening, a browser is fired. We need to make sensible-browser call the same so applications just work. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From oliver.ries at canonical.com Thu Sep 8 02:13:53 2016 From: oliver.ries at canonical.com (Oliver Ries) Date: Wed, 7 Sep 2016 20:13:53 -0600 Subject: Ubuntu Core 16 beta images available In-Reply-To: References: <20160907215054.GG3871@bod> Message-ID: On Wed, Sep 7, 2016 at 5:00 PM, Gustavo Niemeyer < gustavo.niemeyer at canonical.com> wrote: > > Thanks so much and congratulations to everybody who worked hard to put in > place this initial public release of Ubuntu Core, a minimalist Linux > distribution that uses snaps as its building blocks. Special thanks as well > to Michael, Oliver Grawert, Martin Pitt, and others who've put all the bits > in right offsets to have a working image today. > > Over the next few weeks we'll be including a small set of pending features > and polishing this image until we have a stable golden release. > > Please remember that despite some of those conversations being about > Ubuntu Core, at its core we have snaps and snapd still. The vast majority > of features we talk about for Ubuntu Core benefits all the distributions in > which snapd works today. > > Hold on tight and join us in this exciting trip. > also my thank you to everybody involved in getting us to this important milestone, which will be the foundation of the Ubuntu Core 16 release. I sure hope you guys feel proud about your work! Olli -------------- next part -------------- An HTML attachment was scrubbed... URL: From eloy.garcia.pca at gmail.com Thu Sep 8 07:17:07 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Thu, 8 Sep 2016 09:17:07 +0200 Subject: Access to other commands In-Reply-To: References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> Message-ID: Hi all. I currently have several problems with my snap package for wallpaperdownloader application, a java-based piece of software that downloads wallpapers from the Internet, and maybe this is the solution for some of them. In this application, I execute some Linux command tools such as xdg-open and gsettings. They work flawlessly within a "native" environment, but when I try to execute them inside the snap package, they simply don't work. Gustavo, when you suggest to use the line: bus-send --print-reply --session --dest=com.canonical.SafeLauncher / com.canonical.SafeLauncher.OpenURL string:"$1" could you give an example of that? I mean, if I want to, for example, open a browser using "xdg-open https://www.google.com" executing this command from my Java code, what would be the line I should use instead?. This implies to modify my java code too and I had to do the distinction between the application executed in a native environment and in a snap confinement. This wouldn't be the desired way of packaging the app. Would be possible to use some kind of interface to have access to the "native" command line tools installed in the system? This way, source code would be agnostic (I mean, I didn't have to tweak the application depending on the environment executed) and it would be only a matter of snapcraft.yaml configuration. Thanks in advance :) Best, Eloy 2016-09-08 4:11 GMT+02:00 Gustavo Niemeyer : > > > On Wed, Sep 7, 2016 at 10:31 PM, Mark Shuttleworth > wrote: > >> On 07/09/16 12:32, Matthew Williams wrote: >> > I have a few applications I'm looking to snap, but quite a lot of them >> > shell out to other commands at some point and it's not always >> > practical to include these with my snap, one concrete example is >> > shelling out to /usr/bin/sensible-browser. >> > > We have something slightly better than this. Any snap can call xdg-open > $url and have that URL being sent to a safe launcher over dbus. > > This is the exact line: > > bus-send --print-reply --session --dest=com.canonical.SafeLauncher / > com.canonical.SafeLauncher.OpenURL string:"$1" > > Because this is always safe to do and offered over a common mechanism, we > don't need an interface to mediate it. If something is listening, a browser > is fired. > > We need to make sensible-browser call the same so applications just work. > > > gustavo @ http://niemeyer.net > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From zygmunt.krynicki at canonical.com Thu Sep 8 07:30:22 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Thu, 8 Sep 2016 09:30:22 +0200 Subject: Access to other commands In-Reply-To: References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> Message-ID: <4A752850-14A6-42D6-9D78-68071497F9E3@canonical.com> > On 8 Sep 2016, at 09:17, Eloy García (PC Actual) wrote: > > Hi all. > > I currently have several problems with my snap package for wallpaperdownloader application, a java-based piece of software that downloads wallpapers from the Internet, and maybe this is the solution for some of them. > In this application, I execute some Linux command tools such as xdg-open and gsettings. They work flawlessly within a "native" environment, but when I try to execute them inside the snap package, they simply don't work. > > Gustavo, when you suggest to use the line: bus-send --print-reply --session --dest=com.canonical.SafeLauncher / com.canonical.SafeLauncher.OpenURL string:"$1" > could you give an example of that? I mean, if I want to, for example, open a browser using "xdg-open https://www.google.com" executing this command from my Java code, what would be the line I should use instead?. This implies to modify my java code too and I had to do the distinction between the application executed in a native environment and in a snap confinement. This wouldn't be the desired way of packaging the app. Would be possible to use some kind of interface to have access to the "native" command line tools installed in the system? This way, source code would be agnostic (I mean, I didn't have to tweak the application depending on the environment executed) and it would be only a matter of snapcraft.yaml configuration. > I believe there’s a special version of xdg-open that does exactly what that bus-send line above is doing. https://github.com/snapcore/snapd-xdg-open/blob/master/src/xdg-open.c You can add a part to your snap that includes this executable so that all the other code in your snap can remain the same. Best regards ZK From seb128 at ubuntu.com Thu Sep 8 07:58:54 2016 From: seb128 at ubuntu.com (Sebastien Bacher) Date: Thu, 8 Sep 2016 09:58:54 +0200 Subject: Access to other commands In-Reply-To: <4A752850-14A6-42D6-9D78-68071497F9E3@canonical.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <4A752850-14A6-42D6-9D78-68071497F9E3@canonical.com> Message-ID: <314141e2-d21c-64ca-9a17-153d94e57822@ubuntu.com> Le 08/09/2016 à 09:30, Zygmunt Krynicki a écrit : > I believe there’s a special version of xdg-open that does exactly what that bus-send line above is doing. > > https://github.com/snapcore/snapd-xdg-open/blob/master/src/xdg-open.c > > You can add a part to your snap that includes this executable so that all the other code in your snap can remain the same. That shouldn't be needed, we have a custom xdg-open wrapper shipped as /usr/local/bin/xdg-open in the ubuntu-core snap [1] (not sure if the version on the stable channel includes it now, it has been outdated for a while so we might need to wait for a refresh there) snapd-xdg-open needs to be installed on the system though for the service to work and we don't do that by default yet Cheers, Sebastien Bacher [1] http://bazaar.launchpad.net/~ubuntu-core-dev/livecd-rootfs/trunk/view/head:/live-build/ubuntu-core/hooks/500-create-xdg-wrapper.binary#L10 From michael.vogt at canonical.com Thu Sep 8 08:16:07 2016 From: michael.vogt at canonical.com (Michael Vogt) Date: Thu, 8 Sep 2016 10:16:07 +0200 Subject: New snapd 2.14.2 release available Message-ID: <20160908081607.GF3863@bod> Hi, we are happy to announce that the 2.14.2 release of snapd is released in Ubuntu 16.04 and Fedora 24 (COPR). Other distributions will follow soon. We also released a new version of the "core" snap that contains this version of snapd. Some highlights: - When installing snaps without assertions, please use the --force-dangerous option - Use lazy umount on removal which fixes some removal delays we had - Disable the "reexec" feature, you can enable it via SNAP_REEXEC=1 - New `snap download` command to download snaps - Allow install and remove by revision - Tons of image releated fixes and improvements - New interfaces: + fwupd for firmware upgrades + upower-observe + screen-inhibit-control + Updates to default policy, browser-support, bluetooth-control, x11 Please let us know if you notice any issues. Cheers, Michael From jamie.bennett at canonical.com Thu Sep 8 08:25:34 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Thu, 8 Sep 2016 09:25:34 +0100 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160907215054.GG3871@bod> References: <20160907215054.GG3871@bod> Message-ID: <20160908082534.GE15065@phoenix> On 07/09/16 at 11:50pm, Michael Vogt wrote: > Ubuntu Core 16 Images > ===================== > > The Ubuntu snappy team is happy to announce the first beta images for > Ubuntu Core 16. The images use the snapd package manager to install > and update all components of the system including kernel, core, gadget > and applications. Great news and well done all round. To squeeze so much goodness into snapd, snap-confine, snapcraft, store, ubuntu-image, and other components that make this possible is really an achievement. > The images are available for PC (amd64, i386) and Raspberry Pi2 > (armhf). More architectures and boards (arm64 dragonboard, pi3) will > follow shortly. You can download them at: > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > > The images are bootable, the pc image can be booted directly in > qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: > > unxz ubuntu-core-16-pc.img.xz > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > Where /dev/sdXX is the path of your sd card. > > After booting the image you can enter your Ubuntu SSO email and it > will automatically create a matching user with the right ssh keys. If > you do not have an Ubuntu SSO account yet you can create one at: > > https://login.ubuntu.com/ > > These images follow the "beta" channel. > > Enjoy the fresh images! If you find any bugs or issues, please let us > know via: > > https://bugs.launchpad.net/snappy/ > > Cheers, > Michael (on behalf of the snappy team) > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From jenny.murphy at episensor.com Thu Sep 8 09:07:38 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Thu, 8 Sep 2016 10:07:38 +0100 Subject: Access to a Serial Port using Java RXTX library In-Reply-To: References: Message-ID: Just one final update on this. For 64 bit architectures it is very important to build the sources from the rxtx-2.2pre2 release. Using the older 2.1-7r2 version you will run into the following problem when serial communication starts : # A fatal error has been detected by the Java Runtime Environment: # SIGSEGV (0xb) at pc=0x00007fb8b4caec4b, pid=5401, tid=140431283873536 # Problematic frame: # C [librxtxSerial.so+0x6c4b] read_byte_array+0x3b But it is solved in rxtx-2.2pre2. My serial comms are working great now. On 7 September 2016 at 15:06, Loïc Minier wrote: > Great; now we even have google juice for other people trying to use RXTX > with snaps, thanks! :-) > > On Wed, Sep 7, 2016 at 4:04 PM, Jenny Murphy > wrote: > >> Hi, >> No, actually the configure file in RXTX needed to be modified to support >> java 1.7 : >> >> >> case $JAVA_VERSION in >> 1.2*|1.3*|1.4*|1.5*) >> #fix_parameters $JPATH/jre/lib/javax.comm.properties >> CLASSPATH=".:\$(TOP):\$(TOP)/src:"`find $JPATH/ -name RXTXcomm.jar |head >> -n1` >> RXTX_PATH="\$(JPATH)/jre/lib/\$(OS_ARCH)" >> JHOME=$JPATH/"jre/lib/ext" >> >> So the paths were not being set up correctly for me. >> >> I changed it to >> case $JAVA_VERSION in >> 1.2*|1.3*|1.4*|1.5*|1.7*) >> >> Thanks a million for your suggestion. >> >> Jenny >> >> >> On 7 September 2016 at 15:01, Loïc Minier wrote: >> >>> Hi, >>> >>> I'm guessing the Java version change solved the rebuilding problem? >>> >>> Cheers, >>> - Loïc Minier >>> >>> On Wed, Sep 7, 2016 at 3:32 PM, Jenny Murphy >> > wrote: >>> >>>> Hi all, >>>> This is solved . The problem was to do with the Java versions >>>> supported by RXTX. >>>> Jenny >>>> >>>> On 7 September 2016 at 12:51, Jenny Murphy >>>> wrote: >>>> >>>>> Hi, >>>>> I am having some difficulty building RXTX (get unsatisfied link >>>>> errors with the new shared objects and jar file). >>>>> Is there an other alternative to workaround my original problem with >>>>> the permissions to use lock file? >>>>> Thanks. >>>>> >>>>> >>>>> On 6 September 2016 at 12:48, Loïc Minier >>>>> wrote: >>>>> >>>>>> Hi Jenny, >>>>>> >>>>>> The INSTALL file >>>>>> suggests >>>>>> that RXTX implements custom locking which requires OS integration; because >>>>>> you want your app to be the only one accessing the serial port(s), you >>>>>> should build RXTX with --disable-lockfiles to disable this feature of the >>>>>> library. >>>>>> >>>>>> Cheers, >>>>>> - Loïc >>>>>> >>>>>> On Tue, Sep 6, 2016 at 1:34 PM, Jenny Murphy < >>>>>> jenny.murphy at episensor.com> wrote: >>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> I have a java application running in a .snap package. >>>>>>> The java application is going to access the serial port and uses the >>>>>>> RXTX Java library to do this. The RXTX jar is included in the .snap. >>>>>>> >>>>>>> When it runs I get the following error : >>>>>>> >>>>>>> check_group_uucp(): error testing lock file creation Error >>>>>>> details:Permission deniedcheck_lock_status: No permission to create lock >>>>>>> file. >>>>>>> please see: How can I use Lock Files with rxtx? in INSTALL >>>>>>> >>>>>>> >>>>>>> Anyone have any experience of this ? >>>>>>> >>>>>>> Thanks in advance. >>>>>>> Jenny >>>>>>> >>>>>>> -- >>>>>>> *Jenny Murphy* >>>>>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>>>>> jenny.murphy at episensor.com t | +353 >>>>>>> (0) 61 512 511 w | http://www.episensor.com >>>>>>> >>>>>>> -- >>>>>>> Snapcraft mailing list >>>>>>> Snapcraft at lists.snapcraft.io >>>>>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>>>>>> an/listinfo/snapcraft >>>>>>> >>>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> - Loïc >>>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> *Jenny Murphy* >>>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>>> jenny.murphy at episensor.com t | +353 (0) >>>>> 61 512 511 w | http://www.episensor.com >>>>> >>>> >>>> >>>> >>>> -- >>>> *Jenny Murphy* >>>> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >>>> jenny.murphy at episensor.com t | +353 (0) >>>> 61 512 511 w | http://www.episensor.com >>>> >>> >>> >>> >>> -- >>> - Loïc >>> >> >> >> >> -- >> *Jenny Murphy* >> *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* >> jenny.murphy at episensor.com t | +353 (0) 61 >> 512 511 w | http://www.episensor.com >> > > > > -- > - Loïc > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.lenton at canonical.com Thu Sep 8 09:43:44 2016 From: john.lenton at canonical.com (John Lenton) Date: Thu, 8 Sep 2016 10:43:44 +0100 Subject: Access to other commands In-Reply-To: References: Message-ID: On 7 September 2016 at 17:32, Matthew Williams wrote: > one concrete example is shelling out to /usr/bin/sensible-browser. note that sensible-browser will probably only work on debian and derivatives. The better cross-distro way would be to call xdg-open, and we already have that in place AFAIK. From gustavo at niemeyer.net Thu Sep 8 12:13:18 2016 From: gustavo at niemeyer.net (Gustavo Niemeyer) Date: Thu, 8 Sep 2016 09:13:18 -0300 Subject: Access to other commands In-Reply-To: References: Message-ID: If Debian and Ubuntu both have sensible-browser, quite a bit of people will think it's standard and end up relying on it. We can have both, as it's super cheap on our end. On Thu, Sep 8, 2016 at 6:43 AM, John Lenton wrote: > On 7 September 2016 at 17:32, Matthew Williams > wrote: > > one concrete example is shelling out to /usr/bin/sensible-browser. > > note that sensible-browser will probably only work on debian and > derivatives. The better cross-distro way would be to call xdg-open, > and we already have that in place AFAIK. > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Thu Sep 8 12:25:07 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 8 Sep 2016 09:25:07 -0300 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: Alright, I'm moving ahead with this. My proposal is devicecraft at lists.snapcraft.io. If you don't like it please speak up soon. :) On Wed, Sep 7, 2016 at 10:54 AM, Gustavo Niemeyer < gustavo.niemeyer at canonical.com> wrote: > > Rather than splitting Ubuntu Core, I suggest splitting device building. > Ubuntu Core is just another distribution using snaps, so discussing the > well-being of snaps and snapd inside it sounds like a welcome topic here. > Device building and image tuning is a world on itself, though. > > > On Wed, Sep 7, 2016 at 10:42 AM, Mark Shuttleworth > wrote: > >> On 07/09/16 06:56, MikeB wrote: >> >> This new 'first user experience' caught me by surprise - I don't think I >> saw anything about this on the snapcraft mailing list. Is there some other >> list I'm not following that discusses these changes ahead of time? >> >> >> Do we need a list for Ubuntu Core, the all-snap rendition of Ubuntu? This >> list is mainly for people making snaps, which might be running on any >> classic distro, and I would not want folks who's primary interest is a snap >> running on their preferred distro to have to wade through threads about >> Ubuntu Core. >> >> Mark >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > > -- > gustavo @ http://niemeyer.net > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From mabnhdev at gmail.com Thu Sep 8 12:28:31 2016 From: mabnhdev at gmail.com (MikeB) Date: Thu, 8 Sep 2016 08:28:31 -0400 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: Works for me. Like the name! I hope you'll also echo the list to https://www.mail-archive.com like snapcraft - that's my preferred way of navigating the mailing lists. Cheers, Mike On Thu, Sep 8, 2016 at 8:25 AM, Gustavo Niemeyer < gustavo.niemeyer at canonical.com> wrote: > > Alright, I'm moving ahead with this. > > My proposal is devicecraft at lists.snapcraft.io. If you don't like it > please speak up soon. :) > > On Wed, Sep 7, 2016 at 10:54 AM, Gustavo Niemeyer < > gustavo.niemeyer at canonical.com> wrote: > >> >> Rather than splitting Ubuntu Core, I suggest splitting device building. >> Ubuntu Core is just another distribution using snaps, so discussing the >> well-being of snaps and snapd inside it sounds like a welcome topic here. >> Device building and image tuning is a world on itself, though. >> >> >> On Wed, Sep 7, 2016 at 10:42 AM, Mark Shuttleworth >> wrote: >> >>> On 07/09/16 06:56, MikeB wrote: >>> >>> This new 'first user experience' caught me by surprise - I don't think I >>> saw anything about this on the snapcraft mailing list. Is there some other >>> list I'm not following that discusses these changes ahead of time? >>> >>> >>> Do we need a list for Ubuntu Core, the all-snap rendition of Ubuntu? >>> This list is mainly for people making snaps, which might be running on any >>> classic distro, and I would not want folks who's primary interest is a snap >>> running on their preferred distro to have to wade through threads about >>> Ubuntu Core. >>> >>> Mark >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>> an/listinfo/snapcraft >>> >>> >> >> >> -- >> gustavo @ http://niemeyer.net >> > > > > -- > gustavo @ http://niemeyer.net > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Thu Sep 8 12:28:38 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 8 Sep 2016 09:28:38 -0300 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: Actually, I don't like it myself.. sorry. I'll go for the simpler devices at lists.snapcraft.io if there are no complaints soon. On Thu, Sep 8, 2016 at 9:25 AM, Gustavo Niemeyer < gustavo.niemeyer at canonical.com> wrote: > > Alright, I'm moving ahead with this. > > My proposal is devicecraft at lists.snapcraft.io. If you don't like it > please speak up soon. :) > > On Wed, Sep 7, 2016 at 10:54 AM, Gustavo Niemeyer < > gustavo.niemeyer at canonical.com> wrote: > >> >> Rather than splitting Ubuntu Core, I suggest splitting device building. >> Ubuntu Core is just another distribution using snaps, so discussing the >> well-being of snaps and snapd inside it sounds like a welcome topic here. >> Device building and image tuning is a world on itself, though. >> >> >> On Wed, Sep 7, 2016 at 10:42 AM, Mark Shuttleworth >> wrote: >> >>> On 07/09/16 06:56, MikeB wrote: >>> >>> This new 'first user experience' caught me by surprise - I don't think I >>> saw anything about this on the snapcraft mailing list. Is there some other >>> list I'm not following that discusses these changes ahead of time? >>> >>> >>> Do we need a list for Ubuntu Core, the all-snap rendition of Ubuntu? >>> This list is mainly for people making snaps, which might be running on any >>> classic distro, and I would not want folks who's primary interest is a snap >>> running on their preferred distro to have to wade through threads about >>> Ubuntu Core. >>> >>> Mark >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>> an/listinfo/snapcraft >>> >>> >> >> >> -- >> gustavo @ http://niemeyer.net >> > > > > -- > gustavo @ http://niemeyer.net > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Thu Sep 8 12:29:01 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 08 Sep 2016 14:29:01 +0200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: Message-ID: <1473337741.24404.19.camel@ubuntu.com> hi, On Do, 2016-09-08 at 09:25 -0300, Gustavo Niemeyer wrote: > > Alright, I'm moving ahead with this. > > My proposal is devicecraft at lists.snapcraft.io. If you don't like it > please speak up soon. :) >  how about just "devices@" ... this feels all like redhats "*kit" initiative from a few years ago (device-kit, console-kit ... foo-kit, bar-kit) also, currently we dont really have many device related incoming questions on this list, does it really already justify the fragmentation as long as the throughput is that low ?  ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From gustavo.niemeyer at canonical.com Thu Sep 8 12:34:49 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 8 Sep 2016 09:34:49 -0300 Subject: More Ubuntu-Core boot up problems... In-Reply-To: <1473337741.24404.19.camel@ubuntu.com> References: <1473337741.24404.19.camel@ubuntu.com> Message-ID: On Thu, Sep 8, 2016 at 9:29 AM, Oliver Grawert wrote: > hi, > On Do, 2016-09-08 at 09:25 -0300, Gustavo Niemeyer wrote: > > > > Alright, I'm moving ahead with this. > > > > My proposal is devicecraft at lists.snapcraft.io. If you don't like it > > please speak up soon. :) > > > how about just "devices@" ... this feels all like redhats "*kit" > initiative from a few years ago (device-kit, console-kit ... foo-kit, > bar-kit) > > also, currently we dont really have many device related incoming > questions on this list, does it really already justify the > fragmentation as long as the throughput is that low ? > > ciao > oli > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Thu Sep 8 12:36:43 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 8 Sep 2016 09:36:43 -0300 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: <1473337741.24404.19.camel@ubuntu.com> Message-ID: Strange.. my email client just did something silly and discarded my message. Trying again: Yeah, devicecraft is a good name but also awkward to type, and makes snapcraft less interesting as you point out. The fragmentation seems justified as the subjects are quite unrelated. Mixing makes a bad experience on both ends. On Thu, Sep 8, 2016 at 9:34 AM, Gustavo Niemeyer < gustavo.niemeyer at canonical.com> wrote: > > On Thu, Sep 8, 2016 at 9:29 AM, Oliver Grawert wrote: > >> hi, >> On Do, 2016-09-08 at 09:25 -0300, Gustavo Niemeyer wrote: >> > >> > Alright, I'm moving ahead with this. >> > >> > My proposal is devicecraft at lists.snapcraft.io. If you don't like it >> > please speak up soon. :) >> > >> how about just "devices@" ... this feels all like redhats "*kit" >> initiative from a few years ago (device-kit, console-kit ... foo-kit, >> bar-kit) >> >> also, currently we dont really have many device related incoming >> questions on this list, does it really already justify the >> fragmentation as long as the throughput is that low ? >> >> ciao >> oli >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > > -- > gustavo @ http://niemeyer.net > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Thu Sep 8 12:45:41 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 08 Sep 2016 14:45:41 +0200 Subject: More Ubuntu-Core boot up problems... In-Reply-To: References: <1473337741.24404.19.camel@ubuntu.com> Message-ID: <1473338741.24404.24.camel@ubuntu.com> hi, On Do, 2016-09-08 at 09:36 -0300, Gustavo Niemeyer wrote: > Strange.. my email client just did something silly and discarded my > message. Trying again: > > Yeah, devicecraft is a good name but also awkward to type, and makes > snapcraft less interesting as you point out. > > The fragmentation seems justified as the subjects are quite > unrelated. Mixing makes a bad experience on both ends. > i'm not sure how unrelated the subjects really are, for this month i see exactly three small image related threads (plus michaels announce mail) of which only one exploded (this one) due to discussing the creation of a new ML :) looking at the august archive the amount of device related questions is about the same ... and most questions are actually snap/snapd related ... i dont think this justifies singeling out the users of images yet, there is enough overlap. we should definitely do this if there is enough traffic justifying it, but currently this seems like we will just create a desert full of tumbleweed which wont really help promotion either. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From seb128 at ubuntu.com Thu Sep 8 13:38:18 2016 From: seb128 at ubuntu.com (Sebastien Bacher) Date: Thu, 8 Sep 2016 15:38:18 +0200 Subject: No such file or directory when try to execute gsettings In-Reply-To: References: <07556db2-d35e-3aa6-e703-0a98ece000aa@ubuntu.com> Message-ID: <7472fdcf-abdd-5faa-448f-17ba05a73482@ubuntu.com> Le 07/09/2016 à 12:04, Eloy García (PC Actual) a écrit : > > > I get your point and I think this is the real problem. I guess > gsettings command line tool from the snap is storing the correct > key/value inside the snap, but the "native" dconf database is not > getting the new value, because the change wasn't made using "native" > gsettings command line tool. No, it's the other way around. Writting to dconf goes through a proxy service and the client talks to the server using dbus, so with the gsettings interface your snap can contact the desktop service and the write is done in the real-session's dconf database. Reading is done by opening a file in the user directory though and that doesn't work "out of the box" from inside the snaps because your code is going to see the private-snap-dir and not the real system one. If you use the desktop common part it has a hack for that though https://github.com/ubuntu/snapcraft-desktop-helpers/blob/master/gtk/launcher-specific#L79 > I have tried to search dconf database within snap confinement, but I > couldn't find it. Where should I search for it to see if the new value > for the wallpaper was set there? You should rather look on your desktop using e.g dconf-editor Cheers, Sebastien Bacher From olivier.tilloy at canonical.com Thu Sep 8 14:28:07 2016 From: olivier.tilloy at canonical.com (Olivier Tilloy) Date: Thu, 8 Sep 2016 16:28:07 +0200 Subject: Setting release version in snapcraft build In-Reply-To: References: Message-ID: On Thu, Sep 1, 2016 at 7:40 PM, Sergio Schvezov wrote: > Readding the list > > > El 01/09/16 a las 14:27, Casey Marshall escribió: > > On Thu, Sep 1, 2016 at 12:15 PM, Sergio Schvezov > wrote: >> >> El 01/09/16 a las 14:05, Casey Marshall escribió: >> >> When automating the build process for snaps, I'd like to be able to >> provide the release version as an argument to snapcraft, which could then be >> used as a variable in the snapcraft.yaml. >> >> For example, say I'd like to release "foo" version "1.2.3". I'd then like >> to build the snap with version: 1.2.3, and use the git tag "v1.2.3" in its >> source. For example: >> >> name: foo >> version: ${release-version} >> ... >> parts: >> foo: >> source: git at ... >> source-tag: v${release-version} >> >> >> In 2.16 we are introducing a variable to set this one, it comes from the >> `version` defined above. > > > That certainly helps, thanks for this! > >> >> >> >> The alternatives at the moment seem to be: >> >> 1. Parse the version out of snapcraft.yaml, using that as the >> authoritative release version string. Workable for new projects, but a tough >> sell for existing projects with an established release process. >> >> 2. Generate the snapcraft.yaml from a template (jinja, for example), which >> works but is kind of awkward. >> >> So, how about a ${release-version} variable in snapcraft.yaml, which you >> could set on the snapcraft command-line with a --release-version flag? This >> would be darn useful for Jenkins CI scripts that build snaps. >> >> >> What would the version be set to during CI? The version from a snappy >> point of view is just a friendly name, the architecture only cares about >> revisions. > > > Well, I'm thinking of the case where CI knows the release version up front, > and wants to set that on the snap it's building. > > For example, let's say I've added a snapcraft.yaml to my source tree. I > manage my releases by tagging them with a version tag, and then I run a > Jenkins job with the release version as a parameter. > > This release version is used to check out the source tree at that release > version tag, build, run tests, and then build the snap. Jenkins knows the > release version it's building, and should be able to set the version in the > snap. If I can't, I either have to try to the snapcraft.yaml contents in > sync with my release tags, or modify the snapcraft.yaml contents at build > time. > > So instead, I'd want my Jenkins job to be able to set the version on the > command line, something like `snapcraft --version ${RELEASE_VERSION}`. This > avoids fragile repetition of release versions (among the tag & snap), or the > need for generating/substituting snapcraft.yaml contents in the job. > > > I think it would be good to behave like dch does and have a `snapcraft > set-version ` command instead. > > I let this sit for a bit and see what others think. > > In this scenario you'd do > > snapcraft set-version > snapcraft That would work for me too. Is there a bug report I can subscribe to to follow progress/implementation? From jenny.murphy at episensor.com Thu Sep 8 19:05:59 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Thu, 8 Sep 2016 20:05:59 +0100 Subject: Auto start of snap application on boot up Message-ID: Hi, I would like my snap application to start automatically when the system boots. On a non snappy platform I would add the start line to /etc/rc.local, but I see that this is not possible on Ubuntu 15.04 with Snappy. So can someone please advise the alternative? Thanks. -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.manrique at canonical.com Thu Sep 8 19:10:58 2016 From: daniel.manrique at canonical.com (Daniel Manrique) Date: Thu, 8 Sep 2016 15:10:58 -0400 Subject: Auto start of snap application on boot up In-Reply-To: References: Message-ID: On Thu, Sep 8, 2016 at 3:05 PM, Jenny Murphy wrote: > Hi, > I would like my snap application to start automatically when the system > boots. On a non snappy platform I would add the start line to /etc/rc.local, > but I see that this is not possible on Ubuntu 15.04 with Snappy. > So can someone please advise the alternative? > Thanks. Hello, I start a daemon like this: apps: my-nice-daemon: command: bin/my-nice-daemon-server daemon: simple plugs: [network-bind] my-nice-daemon-server is a simple shell wrapper that sets things up and starts the daemon. This will start when the system boots (or when the snap is installed, so snap install your-snap will get the service running with no extra work). Another example is here: https://github.com/snapcore/snapcraft/blob/master/demos/webcam-webui/snapcraft.yaml Cheers! - Daniel > > -- > Jenny Murphy > EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland > jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | > http://www.episensor.com > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > From sergio.schvezov at canonical.com Thu Sep 8 19:44:07 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Thu, 8 Sep 2016 16:44:07 -0300 Subject: Auto start of snap application on boot up In-Reply-To: References: Message-ID: <06ce813a-b5cc-67c0-7a74-e383c75343e7@canonical.com> El 08/09/16 a las 16:10, Daniel Manrique escribió: > On Thu, Sep 8, 2016 at 3:05 PM, Jenny Murphy wrote: >> Hi, >> I would like my snap application to start automatically when the system >> boots. On a non snappy platform I would add the start line to /etc/rc.local, >> but I see that this is not possible on Ubuntu 15.04 with Snappy. >> So can someone please advise the alternative? >> Thanks. > Hello, > > I start a daemon like this: This is 16.04 terminology... > > apps: > my-nice-daemon: > command: bin/my-nice-daemon-server > daemon: simple > plugs: [network-bind] > > > my-nice-daemon-server is a simple shell wrapper that sets things up > and starts the daemon. > > This will start when the system boots (or when the snap is installed, > so snap install your-snap will get the service running with no extra > work). You want something like https://github.com/snapcore/snapcraft/blob/1.x/examples/shout/snapcraft.yaml So services: : start: description: There are many more keywords that can be used here which are described in https://github.com/snapcore/snapd/blob/15.04/docs/meta.md (and might also be in that manual link I shared with you) Cheers Sergio From sylvain.pineau at canonical.com Fri Sep 9 07:41:42 2016 From: sylvain.pineau at canonical.com (Sylvain Pineau) Date: Fri, 9 Sep 2016 09:41:42 +0200 Subject: /snap/bin not in $PATH on desktop In-Reply-To: <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> References: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> Message-ID: <8917eb02-cc1b-8e0c-bada-c60b51b2064a@canonical.com> On 07/09/2016 16:20, Zygmunt Krynicki wrote: >> On 7 Sep 2016, at 13:16, Sylvain Pineau wrote: >> >> Hello, >> >> I noticed that I was not able to call other snap commands from my own snap on my desktop. >> So I tested what was defined using the hello-world.env command. >> >> On desktop (with ubuntu-core 16.04.1 rev 352) >> >> $ hello-world.env | grep ^PATH= >> PATH=/usr/sbin:/usr/bin:/sbin:/bin:/usr/games >> >> But on a true snappy system (with ubuntu-core 16.04.1 rev 453), I get: >> >> $ hello-world.env | grep ^PATH= >> PATH=/home/ubuntu/bin:/home/ubuntu/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin >> >> Is there any reason to not have /snap/bin as part of the $PATH available to snap commands? > Snaps cannot execute other snaps. The PATH difference is caused by how snap-confine behaves when it runs on classic but in general even if you used an absolute path you would not be able to start any other applications from /snap/bin. Allowing this would create an implicit interface (dependency between your snap and some other, perhaps third party, snap). > > If you need access to executables that you control you can use the content interface to bind mount another snap into your own snap and execute those commands directly, with the same security profile as the running application. > > Best regards > ZK > Let me add some context to my initial post. Basically we need to test other snaps (commands) from our checkbox test runner (a different snap). The issue I mentioned above was the ability to run docker commands from the checkbox snap. Both were installed with --devmode. But I'm not the owner of the docker snap so using the content interface won't work for us. Let's consider we're just the owner of the checkbox snap. We could add /snap/bin to our own wrappers but it seems to be a bad practice. 1. What other solutions could we use? 2. On non-classic environments, can we rely on $PATH having /snap/bin in the future? Sylvain From alan.pope at canonical.com Fri Sep 9 10:58:37 2016 From: alan.pope at canonical.com (Alan Pope) Date: Fri, 9 Sep 2016 11:58:37 +0100 Subject: 'Unconfined' apps Message-ID: Hi, This is an architectural snappy question where I have one use case, but have seen others mention similar issues which may be related. Perhaps they could speak up also with their requirements. With regards to https://code.launchpad.net/~popey/ubuntu-terminal-app/add-snapcraft-config/+merge/305206 http://people.canonical.com/~alan/ubuntu-terminal-app_0.7.207_amd64.snap I made the above merge and snap to test out the phone terminal app on the desktop as a snap, for possible inclusion in the store. The goal being that people can install it on a Unity8 snap-only system. But, it's a bit useless in its current form, due in part to our confinement and store policies. In the click store (on the phone) the app is unconfined, so can access files/programs outside of the click. If I set confinement to be 'strict' then I can put it in the stable store, but you can't actually run any non-built-in things (like ssh, top), making it unusable for most people. If I make it use the 'devmode' confinement policy then it (as I understand it) *cannot* go into the stable store (by policy), but can execute external commands in the core. However, it can't be used to launch other executables in other snaps, making it somewhat useless on a snap-only system with other tools installed. I don't believe this to be unique to this terminal, nor desktop/graphical apps, other snap-packaged terminals (and file managers & other system level things) may have the same issue. How do we we resolve this? Do we request a security exception & code audit? Is there some other planned interface for these kinds of 'expert' apps which need to reach outside of their confinement? Cheers, -- Alan Pope Community Manager Canonical - Ubuntu Engineering and Services +44 (0) 7973 620 164 alan.pope at canonical.com http://ubuntu.com/ From jamie at canonical.com Fri Sep 9 13:25:49 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Fri, 09 Sep 2016 08:25:49 -0500 Subject: /snap/bin not in $PATH on desktop In-Reply-To: <8917eb02-cc1b-8e0c-bada-c60b51b2064a@canonical.com> References: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> <8917eb02-cc1b-8e0c-bada-c60b51b2064a@canonical.com> Message-ID: <1473427549.25902.39.camel@canonical.com> On Fri, 2016-09-09 at 09:41 +0200, Sylvain Pineau wrote: > On 07/09/2016 16:20, Zygmunt Krynicki wrote: > > > > > > > > On 7 Sep 2016, at 13:16, Sylvain Pineau > > > wrote: > > > > > > Hello, > > > > > > I noticed that I was not able to call other snap commands from my own snap > > > on my desktop. > > > So I tested what was defined using the hello-world.env command. > > > > > > On desktop (with ubuntu-core 16.04.1 rev 352) > > > > > > $ hello-world.env | grep ^PATH= > > > PATH=/usr/sbin:/usr/bin:/sbin:/bin:/usr/games > > > > > > But on a true snappy system (with ubuntu-core 16.04.1 rev 453), I get: > > > > > > $ hello-world.env | grep ^PATH= > > > PATH=/home/ubuntu/bin:/home/ubuntu/.local/bin:/usr/local/sbin:/usr/local/b > > > in:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin > > > > > > Is there any reason to not have /snap/bin as part of the $PATH available > > > to snap commands? > > Snaps cannot execute other snaps. The PATH difference is caused by how snap- > > confine behaves when it runs on classic but in general even if you used an > > absolute path you would not be able to start any other applications from > > /snap/bin. Allowing this would create an implicit interface (dependency > > between your snap and some other, perhaps third party, snap). > > > > If you need access to executables that you control you can use the content > > interface to bind mount another snap into your own snap and execute those > > commands directly, with the same security profile as the running > > application. > > > > Best regards > > ZK > > > Let me add some context to my initial post. > Basically we need to test other snaps (commands) from our checkbox test  > runner (a different snap). > The issue I mentioned above was the ability to run docker commands from  > the checkbox snap. > Both were installed with --devmode. > But I'm not the owner of the docker snap so using the content interface  > won't work for us. > Let's consider we're just the owner of the checkbox snap. > We could add /snap/bin to our own wrappers but it seems to be a bad  > practice. Since snaps aren't expected to be able to launch programs from /snap/bin for the reasons already mentioned, I would argue that snapd should strip /snap/bin from the PATH for launched commands to avoid confusion. checkbox and things that launch other snaps' commands via devmode are the outliers and it seems perfectly reasonable to me for them to add /snap/bin to their PATH via wrappers. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From fcole90 at gmail.com Fri Sep 9 17:06:05 2016 From: fcole90 at gmail.com (Fabio Colella) Date: Fri, 9 Sep 2016 19:06:05 +0200 Subject: 'Unconfined' apps In-Reply-To: References: Message-ID: Hi, I agree with Alan. Some other snaps that could need something like this could be app launchers (e.g. Whisker Menu) and desktop environments. Cheers On 9 September 2016 at 12:58, Alan Pope wrote: > Hi, > > This is an architectural snappy question where I have one use case, > but have seen others mention similar issues which may be related. > Perhaps they could speak up also with their requirements. > > With regards to > https://code.launchpad.net/~popey/ubuntu-terminal-app/add- > snapcraft-config/+merge/305206 > > http://people.canonical.com/~alan/ubuntu-terminal-app_0.7.207_amd64.snap > > I made the above merge and snap to test out the phone terminal app on > the desktop as a snap, for possible inclusion in the store. The goal > being that people can install it on a Unity8 snap-only system. > > But, it's a bit useless in its current form, due in part to our > confinement and store policies. In the click store (on the phone) the > app is unconfined, so can access files/programs outside of the click. > > If I set confinement to be 'strict' then I can put it in the stable > store, but you can't actually run any non-built-in things (like ssh, > top), making it unusable for most people. > > If I make it use the 'devmode' confinement policy then it (as I > understand it) *cannot* go into the stable store (by policy), but can > execute external commands in the core. However, it can't be used to > launch other executables in other snaps, making it somewhat useless on > a snap-only system with other tools installed. > > I don't believe this to be unique to this terminal, nor > desktop/graphical apps, other snap-packaged terminals (and file > managers & other system level things) may have the same issue. > > How do we we resolve this? Do we request a security exception & code audit? > Is there some other planned interface for these kinds of 'expert' apps > which need to reach outside of their confinement? > > Cheers, > -- > Alan Pope > Community Manager > > Canonical - Ubuntu Engineering and Services > +44 (0) 7973 620 164 > alan.pope at canonical.com > http://ubuntu.com/ > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jb at kdenlive.org Sat Sep 10 14:43:50 2016 From: jb at kdenlive.org (Jean-Baptiste Mardelle) Date: Sat, 10 Sep 2016 16:43:50 +0200 Subject: Accessing dbus (KDE Application) Message-ID: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> Hi all, Working on a snap package for Kdenlive (video editing app based on KDE frameworks), i encounter one problem with restrictions of the "strict" mode. My snap package runs fine when using devmode, but crashes on startup when using the strict mode with the following message: "Couldn't register name 'org.kde.kdenlive-17772' with DBUS - another process owns it already!" Kdenlive (and many KDE Applications) create a DBUS entry on startup: org.kde.appName-$PID In our case, we use dbus to inform the main application of the progress of video rendering tasks, which are performed by a command line application. Is there anything I can do to allow my application to register and access its dbus service ? Thanks, Jean-Baptiste Mardelle Kdenlive maintainer From mhall119 at ubuntu.com Sat Sep 10 15:38:30 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Sat, 10 Sep 2016 11:38:30 -0400 Subject: Accessing dbus (KDE Application) In-Reply-To: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> Message-ID: <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> There's an open bug report for this: https://bugs.launchpad.net/snappy/+bug/1590679 There is work going on to implement a snappy interface that will allow apps like kdenlive to claim their own dbus service name. Michael Hall mhall119 at ubuntu.com On 09/10/2016 10:43 AM, Jean-Baptiste Mardelle wrote: > Hi all, > > Working on a snap package for Kdenlive (video editing app based on KDE > frameworks), i encounter one problem with restrictions of the "strict" > mode. > My snap package runs fine when using devmode, but crashes on startup > when using the strict mode with the following message: > "Couldn't register name 'org.kde.kdenlive-17772' with DBUS - another > process owns it already!" > > Kdenlive (and many KDE Applications) create a DBUS entry on startup: > org.kde.appName-$PID > > In our case, we use dbus to inform the main application of the progress > of video rendering tasks, which are performed by a command line > application. > > Is there anything I can do to allow my application to register and > access its dbus service ? > > Thanks, > > Jean-Baptiste Mardelle > Kdenlive maintainer > From mark at ubuntu.com Sun Sep 11 10:50:29 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 11 Sep 2016 11:50:29 +0100 Subject: Accessing dbus (KDE Application) In-Reply-To: <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> Message-ID: On 10/09/16 16:38, Michael Hall wrote: > There is work going on to implement a snappy interface that will allow > apps like kdenlive to claim their own dbus service name. Can we get a 'blunt' version of that interface which allows access to all dbus services, to start with, quickly? Mark From gustavo.niemeyer at canonical.com Sun Sep 11 17:31:28 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Sun, 11 Sep 2016 14:31:28 -0300 Subject: Accessing dbus (KDE Application) In-Reply-To: References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> Message-ID: We already have that interface pretty much done. There were just details to sort out, which got postponed because we got "distracted" with the Ubuntu Core deadlines. The PR is still open, and still needs those details sorted. On Sep 11, 2016 11:26 AM, "Mark Shuttleworth" wrote: > On 10/09/16 16:38, Michael Hall wrote: > > There is work going on to implement a snappy interface that will allow > > apps like kdenlive to claim their own dbus service name. > > Can we get a 'blunt' version of that interface which allows access to > all dbus services, to start with, quickly? > > Mark > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From xiaoguo.liu at canonical.com Mon Sep 12 01:41:33 2016 From: xiaoguo.liu at canonical.com (XiaoGuo Liu) Date: Mon, 12 Sep 2016 09:41:33 +0800 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160908082534.GE15065@phoenix> References: <20160907215054.GG3871@bod> <20160908082534.GE15065@phoenix> Message-ID: Hi, I have a QualCom dragonboard 410. I have flashed the Ubuntu Core image to the SD card, and I also set the S6 switch to boot from the SD card according to the instruction at: https://github.com/96boards/documentation/wiki/Dragonboard-410c-Installation-Guide-for-Linux-and-Android#installing-image-using-an-sd-card-image When booting into the Ubuntu core, it needs to have the network access to complete the installation. The board does not provide an Ethernet port use. It provide the WLAN. How can we configure the WLAN to make the network working to complete the installation? It would be good to have a complete installation guide at our developer website. Thanks & best regards, XiaoGuo On Thu, Sep 8, 2016 at 4:25 PM, Jamie Bennett wrote: > On 07/09/16 at 11:50pm, Michael Vogt wrote: > > Ubuntu Core 16 Images > > ===================== > > > > The Ubuntu snappy team is happy to announce the first beta images for > > Ubuntu Core 16. The images use the snapd package manager to install > > and update all components of the system including kernel, core, gadget > > and applications. > > Great news and well done all round. To squeeze so much goodness into snapd, > snap-confine, snapcraft, store, ubuntu-image, and other components that > make > this possible is really an achievement. > > > The images are available for PC (amd64, i386) and Raspberry Pi2 > > (armhf). More architectures and boards (arm64 dragonboard, pi3) will > > follow shortly. You can download them at: > > > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > > > > The images are bootable, the pc image can be booted directly in > > qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: > > > > unxz ubuntu-core-16-pc.img.xz > > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > > > Where /dev/sdXX is the path of your sd card. > > > > After booting the image you can enter your Ubuntu SSO email and it > > will automatically create a matching user with the right ssh keys. If > > you do not have an Ubuntu SSO account yet you can create one at: > > > > https://login.ubuntu.com/ > > > > These images follow the "beta" channel. > > > > Enjoy the fresh images! If you find any bugs or issues, please let us > > know via: > > > > https://bugs.launchpad.net/snappy/ > > > > Cheers, > > Michael (on behalf of the snappy team) > > > > > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- XiaoGuo, Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Sun Sep 11 21:35:35 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 11 Sep 2016 22:35:35 +0100 Subject: /snap/bin not in $PATH on desktop In-Reply-To: <1473427549.25902.39.camel@canonical.com> References: <236d9d2e-561f-67ea-df44-35815c492592@canonical.com> <185A4BD6-D5F5-4462-8938-424CB6F3AF58@canonical.com> <8917eb02-cc1b-8e0c-bada-c60b51b2064a@canonical.com> <1473427549.25902.39.camel@canonical.com> Message-ID: <6b39e9f5-5fc1-b135-5025-33f851484440@ubuntu.com> On 09/09/16 14:25, Jamie Strandboge wrote: > Since snaps aren't expected to be able to launch programs from > /snap/bin for the > reasons already mentioned, I would argue that snapd should strip /snap/bin from > the PATH for launched commands to avoid confusion. checkbox and things that > launch other snaps' commands via devmode are the outliers and it seems perfectly > reasonable to me for them to add /snap/bin to their PATH via wrappers. Jamie, I disagree with the blanket statement that 'snaps aren't expected to'... We have interfaces specifically to offer snaps the ability to do anything. Whether a snap is *allowed* to use an interface is a different question. But if by definition snaps are just packages, and packages are software that might do anything, then snaps must at least in principle be able to do anything. For example, a chef snap should definitely be able to do what Sylvain is asking, on Classic Ubuntu. Mark -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From mark at ubuntu.com Sun Sep 11 22:15:58 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 11 Sep 2016 23:15:58 +0100 Subject: (Not entirely) working LDC snap In-Reply-To: <3a91ee5a-1279-a33e-b833-63a719c1cf00@webdrake.net> References: <5acf380b-fe37-68df-9eba-15ccbbefe8a9@webdrake.net> <0cbbc740-1f7a-d6a7-ac8a-bfe97b6ee0de@webdrake.net> <3a91ee5a-1279-a33e-b833-63a719c1cf00@webdrake.net> Message-ID: On 06/09/16 23:21, Joseph Rushton Wakeling wrote: > IOW instead of requesting access to one particular interface, snaps > could request access to collections of services, and other snaps could > add individual services to a given collection. In a sense the "set of slots that have a matching interface to your plug" is the collection (though some of those slots are generative - you ask for a disk, they give you a disk but keep offering disks until they have no more to offer). I like the metaphor though, let's see if we can work it into the story. Mark From mark at ubuntu.com Sun Sep 11 22:19:58 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 11 Sep 2016 23:19:58 +0100 Subject: Snapping Neovim In-Reply-To: References: <57C07AFC.2030300@canonical.com> Message-ID: On 07/09/16 11:58, Gustavo Niemeyer wrote: > They can install it as devmode, and we can also introduce an > "editor-support" interface which gives global read/write access to it. I like the idea of this, specifically excluding global execute (which something like a chef snap or juju agent snap would want). Mark From lorn.potter at canonical.com Mon Sep 12 08:13:27 2016 From: lorn.potter at canonical.com (Lorn Potter) Date: Mon, 12 Sep 2016 18:13:27 +1000 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160907215054.GG3871@bod> References: <20160907215054.GG3871@bod> Message-ID: <6b885c96-9d5c-e96e-272c-7859116021b6@canonical.com> On 08/09/16 07:50, Michael Vogt wrote: > More architectures and boards (arm64 dragonboard, pi3) will > follow shortly. You can download them at: > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ I tried the rpi3 image, when I tried unxz, I got this: unxz: ubuntu-core-16-rpi3.img.xz: Unexpected end of input md5sum was the same as reported on the server. -- Software Engineer System Enablement, Canonical Ltd QtSystemInfo, QtSensors maintainer From xiaoguo.liu at canonical.com Mon Sep 12 08:26:29 2016 From: xiaoguo.liu at canonical.com (XiaoGuo Liu) Date: Mon, 12 Sep 2016 16:26:29 +0800 Subject: Ubuntu Core 16 beta images available In-Reply-To: <6b885c96-9d5c-e96e-272c-7859116021b6@canonical.com> References: <20160907215054.GG3871@bod> <6b885c96-9d5c-e96e-272c-7859116021b6@canonical.com> Message-ID: Hi Lorn, I tried it, and I got the same result. However, the image worked on my Pi3 device :) Best regards, XiaoGuo On Mon, Sep 12, 2016 at 4:13 PM, Lorn Potter wrote: > > > On 08/09/16 07:50, Michael Vogt wrote: > > More architectures and boards (arm64 dragonboard, pi3) will > > follow shortly. You can download them at: > > > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > > I tried the rpi3 image, when I tried unxz, I got this: > unxz: ubuntu-core-16-rpi3.img.xz: Unexpected end of input > > md5sum was the same as reported on the server. > > > -- > Software Engineer > System Enablement, Canonical Ltd > QtSystemInfo, QtSensors maintainer > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- XiaoGuo, Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie.bennett at canonical.com Mon Sep 12 08:46:27 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Mon, 12 Sep 2016 09:46:27 +0100 Subject: Ubuntu Core 16 beta images available In-Reply-To: References: <20160907215054.GG3871@bod> <20160908082534.GE15065@phoenix> Message-ID: <3FDA1A21-F9BA-4AB5-B426-AB6F6C868FC5@canonical.com> > On 12 Sep 2016, at 02:41, XiaoGuo Liu wrote: > > Hi, > > I have a QualCom dragonboard 410. I have flashed the Ubuntu Core image to the SD card, and I also set the S6 switch to boot from the SD card according to the instruction at: > > https://github.com/96boards/documentation/wiki/Dragonboard-410c-Installation-Guide-for-Linux-and-Android#installing-image-using-an-sd-card-image > > When booting into the Ubuntu core, it needs to have the network access to complete the installation. The board does not provide an Ethernet port use. It provide the WLAN. How can we configure the WLAN to make the network working to complete the installation? It would be good to have a complete installation guide at our developer website. WiFI network support on the Dragonboard is not working at the moment due to the changes we made to the first boot experience (console-conf). If you have a USB network adapter then this should work for now. In the near future we will have console-conf talk over WiFi for devices without an ethernet adapter. > Thanks & best regards, > XiaoGuo > > On Thu, Sep 8, 2016 at 4:25 PM, Jamie Bennett > wrote: > On 07/09/16 at 11:50pm, Michael Vogt wrote: > > Ubuntu Core 16 Images > > ===================== > > > > The Ubuntu snappy team is happy to announce the first beta images for > > Ubuntu Core 16. The images use the snapd package manager to install > > and update all components of the system including kernel, core, gadget > > and applications. > > Great news and well done all round. To squeeze so much goodness into snapd, > snap-confine, snapcraft, store, ubuntu-image, and other components that make > this possible is really an achievement. > > > The images are available for PC (amd64, i386) and Raspberry Pi2 > > (armhf). More architectures and boards (arm64 dragonboard, pi3) will > > follow shortly. You can download them at: > > > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > > > > The images are bootable, the pc image can be booted directly in > > qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: > > > > unxz ubuntu-core-16-pc.img.xz > > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > > > Where /dev/sdXX is the path of your sd card. > > > > After booting the image you can enter your Ubuntu SSO email and it > > will automatically create a matching user with the right ssh keys. If > > you do not have an Ubuntu SSO account yet you can create one at: > > > > https://login.ubuntu.com/ > > > > These images follow the "beta" channel. > > > > Enjoy the fresh images! If you find any bugs or issues, please let us > > know via: > > > > https://bugs.launchpad.net/snappy/ > > > > Cheers, > > Michael (on behalf of the snappy team) > > > > > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > -- > XiaoGuo, Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Mon Sep 12 09:12:29 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 12 Sep 2016 11:12:29 +0200 Subject: Accessing dbus (KDE Application) In-Reply-To: References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> Message-ID: <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> Hello, On 11.09.2016 19:31, Gustavo Niemeyer wrote: > We already have that interface pretty much done. There were just details > to sort out, which got postponed because we got "distracted" with the > Ubuntu Core deadlines. The PR is still open, and still needs those > details sorted. does anyone have a link to the PR? If it's working well-enough already, I could build a package from it and give it to Jean-Baptiste for testing. Have a great day, Daniel From seb128 at ubuntu.com Mon Sep 12 09:16:51 2016 From: seb128 at ubuntu.com (Sebastien Bacher) Date: Mon, 12 Sep 2016 11:16:51 +0200 Subject: Accessing dbus (KDE Application) In-Reply-To: <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> Message-ID: <30fb12e8-d9f2-ae30-c222-1ac1bf47eacc@ubuntu.com> Le 12/09/2016 à 11:12, Daniel Holbach a écrit : > does anyone have a link to the PR? If it's working well-enough already, > I could build a package from it and give it to Jean-Baptiste for testing. I think that's this one https://github.com/snapcore/snapd/pull/1613 Cheers, Sebastien Bacher From daniel.holbach at canonical.com Mon Sep 12 14:30:47 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 12 Sep 2016 16:30:47 +0200 Subject: Accessing dbus (KDE Application) In-Reply-To: <30fb12e8-d9f2-ae30-c222-1ac1bf47eacc@ubuntu.com> References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> <30fb12e8-d9f2-ae30-c222-1ac1bf47eacc@ubuntu.com> Message-ID: Hey hey, On 12.09.2016 11:16, Sebastien Bacher wrote: > Le 12/09/2016 à 11:12, Daniel Holbach a écrit : >> does anyone have a link to the PR? If it's working well-enough already, >> I could build a package from it and give it to Jean-Baptiste for testing. > > I think that's this one > > https://github.com/snapcore/snapd/pull/1613 while this is being worked out, I guess you can publish your snap (in devmode) in the 'beta' or 'edge' channel to get some testing going: http://snapcraft.io/docs/build-snaps/publish Have a great day, Daniel From loic.minier at ubuntu.com Mon Sep 12 15:00:08 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Mon, 12 Sep 2016 17:00:08 +0200 Subject: Build snaps in container In-Reply-To: <43331857-caf2-1558-c14c-0defec52d6b5@ubuntu.com> References: <43331857-caf2-1558-c14c-0defec52d6b5@ubuntu.com> Message-ID: Hi folks, Could we also get the docker container snapcore/snapcraft built for ARM? (I'm using classic mode for now though.) Thanks! - Loïc Minier On Tue, Aug 23, 2016 at 2:48 PM, Sergio Schvezov < sergio.schvezov at canonical.com> wrote: > > > El 23/08/16 a las 07:33, Loïc Minier escribió: > > Hi, > > > > On Tue, Aug 16, 2016 at 3:39 PM, Matt Bruzek > > > > wrote: > > > > It is also available on docker hub: > > https://hub.docker.com/r/jujusolutions/snapbox/ > > > > > > > > Thanks for sharing! Didier and Sergio had already published Docker > > images with snapcraft: > > https://hub.docker.com/r/didrocks/snapcraft/ > > https://hub.docker.com/r/sergiusens/snapcraft/ > > I guess we should consolidate these and have some kind of official > > Ubuntu namespace and images. > > I discussed with Evan and we are going to move the Dockerfile to an > official location like snapcore/snapcraft and have it updated regularly. > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Mon Sep 12 15:42:27 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 12 Sep 2016 17:42:27 +0200 Subject: Accessing dbus (KDE Application) In-Reply-To: References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> <30fb12e8-d9f2-ae30-c222-1ac1bf47eacc@ubuntu.com> Message-ID: <962e6d91-7341-9aaa-57ce-627d5865c9bd@canonical.com> Hello again, On 12.09.2016 16:30, Daniel Holbach wrote: > On 12.09.2016 11:16, Sebastien Bacher wrote: >> Le 12/09/2016 à 11:12, Daniel Holbach a écrit : >>> does anyone have a link to the PR? If it's working well-enough already, >>> I could build a package from it and give it to Jean-Baptiste for testing. >> >> I think that's this one >> >> https://github.com/snapcore/snapd/pull/1613 > > while this is being worked out, I guess you can publish your snap (in > devmode) in the 'beta' or 'edge' channel to get some testing going: > > http://snapcraft.io/docs/build-snaps/publish This might actually be the best course of action for now and is exactly what devmode was made for. :-) I asked around earlier: the dbus interface is high priority. So having your feedback and kdenlive as a great use-case is very valuable. Thanks a lot! Have a great day, Daniel From daniel at bowlhat.net Mon Sep 12 20:13:54 2016 From: daniel at bowlhat.net (Daniel Llewellyn) Date: Mon, 12 Sep 2016 20:13:54 +0000 Subject: Accessing dbus (KDE Application) In-Reply-To: <962e6d91-7341-9aaa-57ce-627d5865c9bd@canonical.com> References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> <30fb12e8-d9f2-ae30-c222-1ac1bf47eacc@ubuntu.com> <962e6d91-7341-9aaa-57ce-627d5865c9bd@canonical.com> Message-ID: An informational: I just ran-up against this issue while trying to snap corebird, which wants to reserve the dbus name "org.baedert.corebird". This would fall afoul of the option to limit names to being under org.gnome namespace. Dan. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Mon Sep 12 22:19:55 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Mon, 12 Sep 2016 15:19:55 -0700 Subject: Accessing dbus (KDE Application) In-Reply-To: References: <95aba172-d5a7-4744-b1cf-cc0c09c26b3b@kdenlive.org> <772e31f7-034b-e313-e19b-5bd5c2c00915@ubuntu.com> <5f16df03-12af-4865-024d-f73196fc9d9b@canonical.com> <30fb12e8-d9f2-ae30-c222-1ac1bf47eacc@ubuntu.com> <962e6d91-7341-9aaa-57ce-627d5865c9bd@canonical.com> Message-ID: On 12/09/16 13:13, Daniel Llewellyn wrote: > I just ran-up against this issue while trying to snap corebird, which > wants to reserve the dbus name "org.baedert.corebird". This would fall > afoul of the option to limit names to being under org.gnome namespace. The more-sophisticated interface in development would allow you to designate the namespaces to bind, not be tied to org.gnome. That said, we could probably *trivially* do an org.gnome-specific interface if that would unblock people and be considered safer than a * binding. Mark From yc.cheng at canonical.com Tue Sep 13 02:31:00 2016 From: yc.cheng at canonical.com (YC Cheng) Date: Tue, 13 Sep 2016 10:31:00 +0800 Subject: Build snaps in container In-Reply-To: References: <43331857-caf2-1558-c14c-0defec52d6b5@ubuntu.com> Message-ID: Hi Loic, Maybe you already knew that https://hub.docker.com/r/vicamo/ubuntu/tags/ have both xenial-arm64 and xenial-armhf tag. YC 2016-09-12 23:00 GMT+08:00 Loïc Minier : > Hi folks, > > Could we also get the docker container snapcore/snapcraft built for ARM? > (I'm using classic mode for now though.) > > Thanks! > - Loïc Minier > > On Tue, Aug 23, 2016 at 2:48 PM, Sergio Schvezov < > sergio.schvezov at canonical.com> wrote: > >> >> >> El 23/08/16 a las 07:33, Loïc Minier escribió: >> > Hi, >> > >> > On Tue, Aug 16, 2016 at 3:39 PM, Matt Bruzek >> > > >> wrote: >> > >> > It is also available on docker hub: >> > https://hub.docker.com/r/jujusolutions/snapbox/ >> > >> > >> > >> > Thanks for sharing! Didier and Sergio had already published Docker >> > images with snapcraft: >> > https://hub.docker.com/r/didrocks/snapcraft/ >> > https://hub.docker.com/r/sergiusens/snapcraft/ >> > I guess we should consolidate these and have some kind of official >> > Ubuntu namespace and images. >> >> I discussed with Evan and we are going to move the Dockerfile to an >> official location like snapcore/snapcraft and have it updated regularly. >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > > -- > - Loïc > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Tue Sep 13 08:14:14 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Tue, 13 Sep 2016 10:14:14 +0200 Subject: Build snaps in container In-Reply-To: References: <43331857-caf2-1558-c14c-0defec52d6b5@ubuntu.com> Message-ID: Are these Ubuntu images? I'm usually running armhf/ubuntu which I believe is the one officially maintained by Canonical. But given we propose using snapcraft from docker as an officially supported image too, I figured it would be nice to have it on armhf too for people building on rpi2/3. Cheers, - Loïc On Tue, Sep 13, 2016 at 4:31 AM, YC Cheng wrote: > Hi Loic, > > Maybe you already knew that https://hub.docker.com/r/vicamo/ubuntu/tags/ > have both > > xenial-arm64 and xenial-armhf tag. > > YC > > 2016-09-12 23:00 GMT+08:00 Loïc Minier : > >> Hi folks, >> >> Could we also get the docker container snapcore/snapcraft built for ARM? >> (I'm using classic mode for now though.) >> >> Thanks! >> - Loïc Minier >> >> On Tue, Aug 23, 2016 at 2:48 PM, Sergio Schvezov < >> sergio.schvezov at canonical.com> wrote: >> >>> >>> >>> El 23/08/16 a las 07:33, Loïc Minier escribió: >>> > Hi, >>> > >>> > On Tue, Aug 16, 2016 at 3:39 PM, Matt Bruzek >>> > > >>> wrote: >>> > >>> > It is also available on docker hub: >>> > https://hub.docker.com/r/jujusolutions/snapbox/ >>> > >>> > >>> > >>> > Thanks for sharing! Didier and Sergio had already published Docker >>> > images with snapcraft: >>> > https://hub.docker.com/r/didrocks/snapcraft/ >>> > https://hub.docker.com/r/sergiusens/snapcraft/ >>> > I guess we should consolidate these and have some kind of official >>> > Ubuntu namespace and images. >>> >>> I discussed with Evan and we are going to move the Dockerfile to an >>> official location like snapcore/snapcraft and have it updated regularly. >>> >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>> an/listinfo/snapcraft >>> >>> >> >> >> -- >> - Loïc >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From yc.cheng at canonical.com Tue Sep 13 08:19:48 2016 From: yc.cheng at canonical.com (YC Cheng) Date: Tue, 13 Sep 2016 16:19:48 +0800 Subject: Build snaps in container In-Reply-To: References: <43331857-caf2-1558-c14c-0defec52d6b5@ubuntu.com> Message-ID: Hi Loic, Those are ubuntu. They are not official. Both armhf and arm64 are there. YC 2016-09-13 16:14 GMT+08:00 Loïc Minier : > Are these Ubuntu images? I'm usually running armhf/ubuntu which I believe > is the one officially maintained by Canonical. But given we propose using > snapcraft from docker as an officially supported image too, I figured it > would be nice to have it on armhf too for people building on rpi2/3. > > Cheers, > - Loïc > > On Tue, Sep 13, 2016 at 4:31 AM, YC Cheng wrote: > >> Hi Loic, >> >> Maybe you already knew that https://hub.docker.com/r/vicamo/ubuntu/tags/ >> have both >> >> xenial-arm64 and xenial-armhf tag. >> >> YC >> >> 2016-09-12 23:00 GMT+08:00 Loïc Minier : >> >>> Hi folks, >>> >>> Could we also get the docker container snapcore/snapcraft built for ARM? >>> (I'm using classic mode for now though.) >>> >>> Thanks! >>> - Loïc Minier >>> >>> On Tue, Aug 23, 2016 at 2:48 PM, Sergio Schvezov < >>> sergio.schvezov at canonical.com> wrote: >>> >>>> >>>> >>>> El 23/08/16 a las 07:33, Loïc Minier escribió: >>>> > Hi, >>>> > >>>> > On Tue, Aug 16, 2016 at 3:39 PM, Matt Bruzek >>>> > > >>>> wrote: >>>> > >>>> > It is also available on docker hub: >>>> > https://hub.docker.com/r/jujusolutions/snapbox/ >>>> > >>>> > >>>> > >>>> > Thanks for sharing! Didier and Sergio had already published Docker >>>> > images with snapcraft: >>>> > https://hub.docker.com/r/didrocks/snapcraft/ >>>> > https://hub.docker.com/r/sergiusens/snapcraft/ >>>> > I guess we should consolidate these and have some kind of official >>>> > Ubuntu namespace and images. >>>> >>>> I discussed with Evan and we are going to move the Dockerfile to an >>>> official location like snapcore/snapcraft and have it updated regularly. >>>> >>>> >>>> -- >>>> Snapcraft mailing list >>>> Snapcraft at lists.snapcraft.io >>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>>> an/listinfo/snapcraft >>>> >>>> >>> >>> >>> -- >>> - Loïc >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >>> an/listinfo/snapcraft >>> >>> >> > > > -- > - Loïc > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim.sueberkrueb at web.de Tue Sep 13 12:01:21 2016 From: tim.sueberkrueb at web.de (=?UTF-8?B?VGltIFPDvGJlcmtyw7xi?=) Date: Tue, 13 Sep 2016 14:01:21 +0200 Subject: Problems running Qt/QML + Oxide snap Message-ID: <67acca39-2842-1cdd-b751-1b3bba835a2f@web.de> Hi everyone, today I was trying to snap an application originally created for Ubuntu phone. The app uses Qt5/QML, Ubuntu UI Toolkit, Oxide and Ubuntu Web. Source: https://github.com/tim-sueberkrueb/crazy-mark snapcraft.yaml: http://paste.ubuntu.com/23173301/ 1) Running the application using NVIDIA drivers fails, which is a known bug (https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1575532). 2) Running the application using Nouveau drivers fails with the following console output: http://paste.ubuntu.com/23173243/ snappy-debug.security scanlog: http://paste.ubuntu.com/23173251/ The application was installed with --devmode and --force-dangerous. I have two questions: 1) Are there any news about the NVDIA driver problem? 2) Am I doing something wrong (regarding Nouveau)? Is this a bug? Can I work around the problem? Thanks in advance! All the best, Tim -------------- next part -------------- An HTML attachment was scrubbed... URL: From penk.chen at canonical.com Tue Sep 13 12:35:25 2016 From: penk.chen at canonical.com (Penk Chen) Date: Tue, 13 Sep 2016 20:35:25 +0800 Subject: Problems running Qt/QML + Oxide snap In-Reply-To: <67acca39-2842-1cdd-b751-1b3bba835a2f@web.de> References: <67acca39-2842-1cdd-b751-1b3bba835a2f@web.de> Message-ID: Hi Tim, I'm working on the same idea of snap without any display server, you may try adding this to your wrapper script to disable oxide's sandbox: export OXIDE_NO_SANDBOX=1 And later on you may have a look of the browser-support interface with 'allow-sandbox: false'. Best, penk On Tue, Sep 13, 2016 at 8:01 PM, Tim Süberkrüb wrote: > Hi everyone, > > today I was trying to snap an application originally created for Ubuntu > phone. > The app uses Qt5/QML, Ubuntu UI Toolkit, Oxide and Ubuntu Web. > > Source: https://github.com/tim-sueberkrueb/crazy-mark > snapcraft.yaml: http://paste.ubuntu.com/23173301/ > > 1) Running the application using NVIDIA drivers fails, which is a known > bug (https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1575532). > 2) Running the application using Nouveau drivers fails with the following > console output: http://paste.ubuntu.com/23173243/ > > snappy-debug.security scanlog: http://paste.ubuntu.com/23173251/ > The application was installed with --devmode and --force-dangerous. > > I have two questions: > 1) Are there any news about the NVDIA driver problem? > 2) Am I doing something wrong (regarding Nouveau)? Is this a bug? Can I > work around the problem? > > Thanks in advance! > > All the best, > Tim > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim.sueberkrueb at web.de Tue Sep 13 15:40:20 2016 From: tim.sueberkrueb at web.de (=?UTF-8?B?VGltIFPDvGJlcmtyw7xi?=) Date: Tue, 13 Sep 2016 17:40:20 +0200 Subject: Problems running Qt/QML + Oxide snap In-Reply-To: References: <67acca39-2842-1cdd-b751-1b3bba835a2f@web.de> Message-ID: Hi Penk, thanks very much for your reply. I will try it out as soon as possible :) Have a nice day, Tim On 13.09.2016 14:35, Penk Chen wrote: > Hi Tim, > > I'm working on the same idea of snap without any display server, you > may try adding this to your wrapper script to disable oxide's sandbox: > > export OXIDE_NO_SANDBOX=1 > > And later on you may have a look of the browser-support interface with > 'allow-sandbox: false'. > > Best, > penk > > On Tue, Sep 13, 2016 at 8:01 PM, Tim Süberkrüb > wrote: > > Hi everyone, > > today I was trying to snap an application originally created for > Ubuntu phone. > The app uses Qt5/QML, Ubuntu UI Toolkit, Oxide and Ubuntu Web. > > Source: https://github.com/tim-sueberkrueb/crazy-mark > > snapcraft.yaml: http://paste.ubuntu.com/23173301/ > > > 1) Running the application using NVIDIA drivers fails, which is a > known bug > (https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1575532 > ). > 2) Running the application using Nouveau drivers fails with the > following console output: http://paste.ubuntu.com/23173243/ > > > snappy-debug.security scanlog: http://paste.ubuntu.com/23173251/ > > The application was installed with --devmode and --force-dangerous. > > I have two questions: > 1) Are there any news about the NVDIA driver problem? > 2) Am I doing something wrong (regarding Nouveau)? Is this a bug? > Can I work around the problem? > > Thanks in advance! > > All the best, > Tim > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.barth at canonical.com Tue Sep 13 16:09:12 2016 From: david.barth at canonical.com (David Barth) Date: Tue, 13 Sep 2016 18:09:12 +0200 Subject: Problems running Qt/QML + Oxide snap In-Reply-To: References: <67acca39-2842-1cdd-b751-1b3bba835a2f@web.de> Message-ID: Hey Tim, You may also want to look at this branch which has a few changes required to get webbrowser-app working as a snap. A typical Qt/QML app using the Oxide webiew : https://code.launchpad.net/~osomon/webbrowser-app/snap On Tue, Sep 13, 2016 at 5:40 PM, Tim Süberkrüb wrote: > Hi Penk, > > thanks very much for your reply. I will try it out as soon as possible :) > > Have a nice day, > > Tim > > On 13.09.2016 14:35, Penk Chen wrote: > > Hi Tim, > > I'm working on the same idea of snap without any display server, you may > try adding this to your wrapper script to disable oxide's sandbox: > > export OXIDE_NO_SANDBOX=1 > > And later on you may have a look of the browser-support interface with > 'allow-sandbox: false'. > > Best, > penk > > On Tue, Sep 13, 2016 at 8:01 PM, Tim Süberkrüb > wrote: > >> Hi everyone, >> >> today I was trying to snap an application originally created for Ubuntu >> phone. >> The app uses Qt5/QML, Ubuntu UI Toolkit, Oxide and Ubuntu Web. >> >> Source: https://github.com/tim-sueberkrueb/crazy-mark >> snapcraft.yaml: http://paste.ubuntu.com/23173301/ >> >> 1) Running the application using NVIDIA drivers fails, which is a known >> bug (https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1575532). >> 2) Running the application using Nouveau drivers fails with the following >> console output: http://paste.ubuntu.com/23173243/ >> >> snappy-debug.security scanlog: http://paste.ubuntu.com/23173251/ >> The application was installed with --devmode and --force-dangerous. >> >> I have two questions: >> 1) Are there any news about the NVDIA driver problem? >> 2) Am I doing something wrong (regarding Nouveau)? Is this a bug? Can I >> work around the problem? >> >> Thanks in advance! >> >> All the best, >> Tim >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> >> > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.park at canonical.com Wed Sep 14 03:08:04 2016 From: robert.park at canonical.com (Robert Park) Date: Tue, 13 Sep 2016 20:08:04 -0700 Subject: First time snapcraft user having issues Message-ID: Hi all, So I want to snap up a script that needs root and does administrative stuff on the host system. I realize this goes against snaps' goals of being all secure and confined, is there any way I can list a plug in my 'plugs:' that would give me root on the users computer, perhaps giving the user a giant warning that they're installing something that can break their system? Basically I don't want any confinement in my snap at all. Also, is there a way to get a list of possible plugs? The documentation only hints that network and network-bind exist, I found some others by googling other people's snapcraft.yamls, but there doesn't seem to be a way to query what the valid set of possible plugs are (yeah, I suppose you could put anything in there, but I mean I need to query what sockets exist that would make sense to put as my plugs). Thanks. -- robru From sergio.schvezov at canonical.com Wed Sep 14 03:15:21 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 14 Sep 2016 00:15:21 -0300 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: <21b3c556-0edd-678f-1d37-69bd43a67f4c@canonical.com> El 14/09/16 a las 00:08, Robert Park escribió: > Hi all, > > So I want to snap up a script that needs root and does administrative > stuff on the host system. I realize this goes against snaps' goals of > being all secure and confined, is there any way I can list a plug in > my 'plugs:' that would give me root on the users computer, perhaps > giving the user a giant warning that they're installing something that > can break their system? Basically I don't want any confinement in my > snap at all. You can run as root (just use `sudo`), maybe just use `--devmode` initially to get started. > Also, is there a way to get a list of possible plugs? The > documentation only hints that network and network-bind exist, I found > some others by googling other people's snapcraft.yamls, but there > doesn't seem to be a way to query what the valid set of possible plugs > are (yeah, I suppose you could put anything in there, but I mean I > need to query what sockets exist that would make sense to put as my > plugs). Running `snap interfaces` on a system with snapd installed will give you a nice list; the only other way I know is to look at https://github.com/snapcore/snapd/tree/master/interfaces/builtin From david.calle at canonical.com Wed Sep 14 05:36:34 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Wed, 14 Sep 2016 07:36:34 +0200 Subject: First time snapcraft user having issues In-Reply-To: <21b3c556-0edd-678f-1d37-69bd43a67f4c@canonical.com> References: <21b3c556-0edd-678f-1d37-69bd43a67f4c@canonical.com> Message-ID: <3f16e214-9846-a93d-fe5d-58005e78b6ca@canonical.com> On 14/09/2016 05:15, Sergio Schvezov wrote: > El 14/09/16 a las 00:08, Robert Park escribió: > >> Hi all, >> >> So I want to snap up a script that needs root and does administrative >> stuff on the host system. I realize this goes against snaps' goals of >> being all secure and confined, is there any way I can list a plug in >> my 'plugs:' that would give me root on the users computer, perhaps >> giving the user a giant warning that they're installing something that >> can break their system? Basically I don't want any confinement in my >> snap at all. > > You can run as root (just use `sudo`), maybe just use `--devmode` > initially to get started. > >> Also, is there a way to get a list of possible plugs? The >> documentation only hints that network and network-bind exist, I found >> some others by googling other people's snapcraft.yamls, but there >> doesn't seem to be a way to query what the valid set of possible plugs >> are (yeah, I suppose you could put anything in there, but I mean I >> need to query what sockets exist that would make sense to put as my >> plugs). > > Running `snap interfaces` on a system with snapd installed will give > you a nice list; the only other way I know is to look at > https://github.com/snapcore/snapd/tree/master/interfaces/builtin > > Hi Robru and welcome to snaps! - The Interfaces reference on snapcraft.io will give you the complete list of what's available in xenial-update's snapd. - If you are building the latest snapd or using a daily version: the reference is on GitHub https://github.com/snapcore/snapd/blob/master/docs/interfaces.md Cheers, David -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.park at canonical.com Wed Sep 14 06:11:56 2016 From: robert.park at canonical.com (Robert Park) Date: Tue, 13 Sep 2016 23:11:56 -0700 Subject: First time snapcraft user having issues In-Reply-To: <21b3c556-0edd-678f-1d37-69bd43a67f4c@canonical.com> References: <21b3c556-0edd-678f-1d37-69bd43a67f4c@canonical.com> Message-ID: On Sep 13, 2016 8:16 PM, "Sergio Schvezov" wrote: > > El 14/09/16 a las 00:08, Robert Park escribió: > > >> Hi all, >> >> So I want to snap up a script that needs root and does administrative >> stuff on the host system. I realize this goes against snaps' goals of >> being all secure and confined, is there any way I can list a plug in >> my 'plugs:' that would give me root on the users computer, perhaps >> giving the user a giant warning that they're installing something that >> can break their system? Basically I don't want any confinement in my >> snap at all. > > > You can run as root (just use `sudo`), maybe just use `--devmode` initially to get started. You mean use sudo in my script? Or have the user run my script under sudo? Either way I can't seem to run apt-get. Even bundling apt in my snap doesn't work. I really need to get out of the confinement and run my script as root on the host. > > >> Also, is there a way to get a list of possible plugs? The >> documentation only hints that network and network-bind exist, I found >> some others by googling other people's snapcraft.yamls, but there >> doesn't seem to be a way to query what the valid set of possible plugs >> are (yeah, I suppose you could put anything in there, but I mean I >> need to query what sockets exist that would make sense to put as my >> plugs). > > > Running `snap interfaces` on a system with snapd installed will give you a nice list; the only other way I know is to look at https://github.com/snapcore/snapd/tree/master/interfaces/builtin Thanks, I'll try that. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Wed Sep 14 08:56:05 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 14 Sep 2016 09:56:05 +0100 Subject: Auto start of snap application on boot up In-Reply-To: <06ce813a-b5cc-67c0-7a74-e383c75343e7@canonical.com> References: <06ce813a-b5cc-67c0-7a74-e383c75343e7@canonical.com> Message-ID: Hi, I need to assign /dev / ttyUSB0 to my snap and I want to run my snaps a service. So is it legal to add the snappy hw-assign command to the wrapper script ? Thanks. El 08/09/16 a las 16:10, Daniel Manrique escribió: > On Thu, Sep 8, 2016 at 3:05 PM, Jenny Murphy > wrote: > >> Hi, >> I would like my snap application to start automatically when the system >> boots. On a non snappy platform I would add the start line to >> /etc/rc.local, >> but I see that this is not possible on Ubuntu 15.04 with Snappy. >> So can someone please advise the alternative? >> Thanks. >> > Hello, > > I start a daemon like this: > This is 16.04 terminology... > > apps: > my-nice-daemon: > command: bin/my-nice-daemon-server > daemon: simple > plugs: [network-bind] > > > my-nice-daemon-server is a simple shell wrapper that sets things up > and starts the daemon. > > This will start when the system boots (or when the snap is installed, > so snap install your-snap will get the service running with no extra > work). > You want something like https://github.com/snapcore/sn apcraft/blob/1.x/examples/shout/snapcraft.yaml So services: : start: description: There are many more keywords that can be used here which are described in https://github.com/snapcore/snapd/blob/15.04/docs/meta.md (and might also be in that manual link I shared with you) Cheers Sergio -- Snapcraft mailing list Snapcraft at lists.snapcraft.io Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm an/listinfo/snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Wed Sep 14 12:29:19 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 14 Sep 2016 14:29:19 +0200 Subject: ubuntu-core image hygiene Message-ID: <1473856159.17795.120.camel@ubuntu.com> hi, Tl;dr;  - managing seeds for ubuntu-core changed from task to meta package - x86 based ubuntu-core snaps lost multiple megabytes of fat ...  And here the details:  With the switch to ubuntu-image and the new gadget handling that comes along with this, we were now able to drop all bits of bootloaders we shipped inside the rootfs (bootloader binaries (specifically grub) are actually shipped inside the gadget now and do not need to be copied from the rootfs into the boot partition any more like we did in the past with ubuntu-device-flash). While working on dropping grub from ubuntu-core one large problem arose ... our seeds used tasks instead of a meta package. Since we can not change the seed itself after a release is out (snappy is based on xenial) this makes it impossible to actually drop packages ... at the same time the only way to add new packages is to add very ugly hacks to livecd-rootfs to import them directly during the rootfs build. Since the classic ubuntu has always had similar issues with the above when doing point releases, it was actually completely switched from tasks to meta packages (relatively recently) ... i did the same for ubuntu-core now so that package addition/removal gets easier and to be in sync with the main distro. If you need to add/remove a package in the ubuntu-core snap, please use the ubuntu-core-meta package now (currently in the snappy-dev/image PPA, it will be SRUed into xenial-updates proper before GA) I am happy to report that thanks to the switch i was also able to drop about 50 lines of awful hacks from the build system that injected packages at build time in an undocumented way :) Now to the more exciting bits ...  Dropping all grub bits from the amd64 ubuntu-core snap got us from a 75.1MB package size down to 66.9MB ... Likewise the i386 snap went from 72.1MB down to 64.9MB I know that a bunch of people will actually celebrate that massive size reduction (hey manik :) ). Note that the arm based snaps did not change and are still at sizes around 55MB ... ciao oli  -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From kyle.fazzari at canonical.com Wed Sep 14 14:53:40 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Wed, 14 Sep 2016 07:53:40 -0700 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: On 09/13/2016 08:08 PM, Robert Park wrote: > Hi all, > > So I want to snap up a script that needs root and does administrative > stuff on the host system. I realize this goes against snaps' goals of > being all secure and confined, is there any way I can list a plug in > my 'plugs:' that would give me root on the users computer, perhaps > giving the user a giant warning that they're installing something that > can break their system? Basically I don't want any confinement in my > snap at all. Yeah, installing with --devmode disables confinement (but enables warnings). Make sure the "confinement" property in the snapcraft.yaml is set to devmode as well, which will make it impossible to install without --devmode. -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From akash.chandrashekar at canonical.com Wed Sep 14 15:08:11 2016 From: akash.chandrashekar at canonical.com (Akash Chandrashekar) Date: Wed, 14 Sep 2016 08:08:11 -0700 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: Additionally if you want to figure out things like how apparmor blocks certain things and resolve them you can follow /var/log/syslog. Regards, Akash On Sep 14, 2016 7:54 AM, "Kyle Fazzari" wrote: > > On 09/13/2016 08:08 PM, Robert Park wrote: > > Hi all, > > > > So I want to snap up a script that needs root and does administrative > > stuff on the host system. I realize this goes against snaps' goals of > > being all secure and confined, is there any way I can list a plug in > > my 'plugs:' that would give me root on the users computer, perhaps > > giving the user a giant warning that they're installing something that > > can break their system? Basically I don't want any confinement in my > > snap at all. > > Yeah, installing with --devmode disables confinement (but enables > warnings). Make sure the "confinement" property in the snapcraft.yaml is > set to devmode as well, which will make it impossible to install without > --devmode. > > -- > Kyle Fazzari (kyrofa) > Software Engineer > Canonical Ltd. > kyle at canonical.com > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Wed Sep 14 15:13:15 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 14 Sep 2016 12:13:15 -0300 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: <35d35653-05ec-347a-105e-ede36b9b9b60@canonical.com> El 14/09/16 a las 12:08, Akash Chandrashekar escribió: > > Additionally if you want to figure out things like how apparmor blocks > certain things and resolve them you can follow /var/log/syslog. > Preferably `snap install snappy-debug` and use that :-) http://askubuntu.com/questions/783979/how-do-i-debug-snaps/784113#784113 From robert.park at canonical.com Wed Sep 14 15:47:43 2016 From: robert.park at canonical.com (Robert Park) Date: Wed, 14 Sep 2016 08:47:43 -0700 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari wrote: > Yeah, installing with --devmode disables confinement (but enables > warnings). Make sure the "confinement" property in the snapcraft.yaml is > set to devmode as well, which will make it impossible to install without > --devmode. Is it possible to distribute devmode snaps through the snap store? -- robru From evan.dandrea at canonical.com Wed Sep 14 16:23:01 2016 From: evan.dandrea at canonical.com (Evan Dandrea) Date: Wed, 14 Sep 2016 16:23:01 +0000 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: On Wed, 14 Sep 2016 at 10:49 Robert Park wrote: > On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari > wrote: > > Yeah, installing with --devmode disables confinement (but enables > > warnings). Make sure the "confinement" property in the snapcraft.yaml is > > set to devmode as well, which will make it impossible to install without > > --devmode. > > Is it possible to distribute devmode snaps through the snap store? > Yes, but you cannot publish them to the stable channel. They will not appear in snap find results. -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.park at canonical.com Wed Sep 14 17:11:01 2016 From: robert.park at canonical.com (Robert Park) Date: Wed, 14 Sep 2016 10:11:01 -0700 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari wrote: > On 09/13/2016 08:08 PM, Robert Park wrote: >> So I want to snap up a script that needs root and does administrative >> stuff on the host system. I realize this goes against snaps' goals of >> being all secure and confined, is there any way I can list a plug in >> my 'plugs:' that would give me root on the users computer, perhaps >> giving the user a giant warning that they're installing something that >> can break their system? Basically I don't want any confinement in my >> snap at all. > > Yeah, installing with --devmode disables confinement (but enables > warnings). Make sure the "confinement" property in the snapcraft.yaml is > set to devmode as well, which will make it impossible to install without > --devmode. Even with "confinement: devmode" and "sudo snap install --force-dangerous --devmode *.snap" I still can't run commands from the host system: /snap/bileto/x1/bin/bileto: 77: /snap/bileto/x1/bin/bileto: curl: not found + add-apt-repository ppa:ci-train-ppa-service/ /snap/bileto/x1/bin/bileto: 82: /snap/bileto/x1/bin/bileto: add-apt-repository: not found From kyle.fazzari at canonical.com Wed Sep 14 17:18:36 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Wed, 14 Sep 2016 10:18:36 -0700 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: <1237829d-7822-6bf4-6c95-ee27ab7e0311@canonical.com> On 09/14/2016 10:11 AM, Robert Park wrote: > On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari > wrote: >> On 09/13/2016 08:08 PM, Robert Park wrote: >>> So I want to snap up a script that needs root and does administrative >>> stuff on the host system. I realize this goes against snaps' goals of >>> being all secure and confined, is there any way I can list a plug in >>> my 'plugs:' that would give me root on the users computer, perhaps >>> giving the user a giant warning that they're installing something that >>> can break their system? Basically I don't want any confinement in my >>> snap at all. >> Yeah, installing with --devmode disables confinement (but enables >> warnings). Make sure the "confinement" property in the snapcraft.yaml is >> set to devmode as well, which will make it impossible to install without >> --devmode. > Even with "confinement: devmode" and "sudo snap install > --force-dangerous --devmode *.snap" I still can't run commands from > the host system: > > /snap/bileto/x1/bin/bileto: 77: /snap/bileto/x1/bin/bileto: curl: not found > + add-apt-repository ppa:ci-train-ppa-service/ > /snap/bileto/x1/bin/bileto: 82: /snap/bileto/x1/bin/bileto: > add-apt-repository: not found Yeah your PATH probably doesn't include those. The execution environment for a snap (even a devmode snap) is the core snap, which doesn't include e.g. curl. The classic system on which you're presumably running is mounted into var/lib/snapd/hostfs, I _think_ (zygmunt can answer with more certainty). -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From sergio.schvezov at canonical.com Wed Sep 14 17:18:54 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 14 Sep 2016 14:18:54 -0300 Subject: First time snapcraft user having issues In-Reply-To: References: Message-ID: El 14/09/16 a las 14:11, Robert Park escribió: > On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari > wrote: >> On 09/13/2016 08:08 PM, Robert Park wrote: >>> So I want to snap up a script that needs root and does administrative >>> stuff on the host system. I realize this goes against snaps' goals of >>> being all secure and confined, is there any way I can list a plug in >>> my 'plugs:' that would give me root on the users computer, perhaps >>> giving the user a giant warning that they're installing something that >>> can break their system? Basically I don't want any confinement in my >>> snap at all. >> Yeah, installing with --devmode disables confinement (but enables >> warnings). Make sure the "confinement" property in the snapcraft.yaml is >> set to devmode as well, which will make it impossible to install without >> --devmode. > Even with "confinement: devmode" and "sudo snap install > --force-dangerous --devmode *.snap" I still can't run commands from > the host system: > > /snap/bileto/x1/bin/bileto: 77: /snap/bileto/x1/bin/bileto: curl: not found > + add-apt-repository ppa:ci-train-ppa-service/ > /snap/bileto/x1/bin/bileto: 82: /snap/bileto/x1/bin/bileto: > add-apt-repository: not found Snaps live on top of the core os, you can run most things on there; you indeed would need some sort of `classic` access interface to run stuff from you workstation which would need support for all the classic distros and well, it might not work on Ubuntu Core. From kyle.fazzari at canonical.com Wed Sep 14 17:36:19 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Wed, 14 Sep 2016 10:36:19 -0700 Subject: First time snapcraft user having issues In-Reply-To: <1237829d-7822-6bf4-6c95-ee27ab7e0311@canonical.com> References: <1237829d-7822-6bf4-6c95-ee27ab7e0311@canonical.com> Message-ID: <9e13526c-0d6a-1c1a-e5a0-330d3832a499@canonical.com> On 09/14/2016 10:18 AM, Kyle Fazzari wrote: > > On 09/14/2016 10:11 AM, Robert Park wrote: >> On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari >> wrote: >>> On 09/13/2016 08:08 PM, Robert Park wrote: >>>> So I want to snap up a script that needs root and does administrative >>>> stuff on the host system. I realize this goes against snaps' goals of >>>> being all secure and confined, is there any way I can list a plug in >>>> my 'plugs:' that would give me root on the users computer, perhaps >>>> giving the user a giant warning that they're installing something that >>>> can break their system? Basically I don't want any confinement in my >>>> snap at all. >>> Yeah, installing with --devmode disables confinement (but enables >>> warnings). Make sure the "confinement" property in the snapcraft.yaml is >>> set to devmode as well, which will make it impossible to install without >>> --devmode. >> Even with "confinement: devmode" and "sudo snap install >> --force-dangerous --devmode *.snap" I still can't run commands from >> the host system: >> >> /snap/bileto/x1/bin/bileto: 77: /snap/bileto/x1/bin/bileto: curl: not found >> + add-apt-repository ppa:ci-train-ppa-service/ >> /snap/bileto/x1/bin/bileto: 82: /snap/bileto/x1/bin/bileto: >> add-apt-repository: not found > Yeah your PATH probably doesn't include those. The execution environment > for a snap (even a devmode snap) is the core snap, which doesn't include > e.g. curl. The classic system on which you're presumably running is > mounted into var/lib/snapd/hostfs, I _think_ (zygmunt can answer with > more certainty). (Oops, I missed a leading slash on the /var) -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From robert.park at canonical.com Wed Sep 14 19:06:50 2016 From: robert.park at canonical.com (Robert Park) Date: Wed, 14 Sep 2016 12:06:50 -0700 Subject: First time snapcraft user having issues In-Reply-To: <9e13526c-0d6a-1c1a-e5a0-330d3832a499@canonical.com> References: <1237829d-7822-6bf4-6c95-ee27ab7e0311@canonical.com> <9e13526c-0d6a-1c1a-e5a0-330d3832a499@canonical.com> Message-ID: Fun times, gentlemen: HOST="/var/lib/snapd/hostfs" # Break out of snap chroot if [ -x "$HOST/usr/bin/sudo" ]; then app="$(mktemp $HOST/tmp/bileto-app-XXX.sh)" $HOST/usr/bin/sudo cp --verbose --force $0 $app inchroot="$(echo $app | sed "s#$HOST##")" $HOST/usr/bin/sudo $HOST/usr/sbin/chroot $HOST sh $inchroot "$@" retval="$?" rm --force $app exit $retval fi On Wed, Sep 14, 2016 at 10:36 AM, Kyle Fazzari wrote: > On 09/14/2016 10:18 AM, Kyle Fazzari wrote: >> >> On 09/14/2016 10:11 AM, Robert Park wrote: >>> On Wed, Sep 14, 2016 at 7:53 AM, Kyle Fazzari >>> wrote: >>>> On 09/13/2016 08:08 PM, Robert Park wrote: >>>>> So I want to snap up a script that needs root and does administrative >>>>> stuff on the host system. I realize this goes against snaps' goals of >>>>> being all secure and confined, is there any way I can list a plug in >>>>> my 'plugs:' that would give me root on the users computer, perhaps >>>>> giving the user a giant warning that they're installing something that >>>>> can break their system? Basically I don't want any confinement in my >>>>> snap at all. >>>> Yeah, installing with --devmode disables confinement (but enables >>>> warnings). Make sure the "confinement" property in the snapcraft.yaml is >>>> set to devmode as well, which will make it impossible to install without >>>> --devmode. >>> Even with "confinement: devmode" and "sudo snap install >>> --force-dangerous --devmode *.snap" I still can't run commands from >>> the host system: >>> >>> /snap/bileto/x1/bin/bileto: 77: /snap/bileto/x1/bin/bileto: curl: not found >>> + add-apt-repository ppa:ci-train-ppa-service/ >>> /snap/bileto/x1/bin/bileto: 82: /snap/bileto/x1/bin/bileto: >>> add-apt-repository: not found >> Yeah your PATH probably doesn't include those. The execution environment >> for a snap (even a devmode snap) is the core snap, which doesn't include >> e.g. curl. The classic system on which you're presumably running is >> mounted into var/lib/snapd/hostfs, I _think_ (zygmunt can answer with >> more certainty). > > (Oops, I missed a leading slash on the /var) > > -- > Kyle Fazzari (kyrofa) > Software Engineer > Canonical Ltd. > kyle at canonical.com > > -- robru From leo.arias at canonical.com Wed Sep 14 23:43:10 2016 From: leo.arias at canonical.com (Leo Arias) Date: Wed, 14 Sep 2016 17:43:10 -0600 Subject: Access to other commands In-Reply-To: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> Message-ID: <57D9E08E.40807@canonical.com> On 2016-09-07 19:31, Mark Shuttleworth wrote: >> 1) Is there some way I can be specifying a list of commands my snapped >> > app is allowed to call? > You could. But I think there is a class of things that should be allowed > to integrate with the classic shell environment, which means they can > shell out to lots of things. What about --devmode? Should devmode allow calls to all binaries in the $PATH? -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 837 bytes Desc: OpenPGP digital signature URL: From mark at ubuntu.com Wed Sep 14 23:52:15 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 14 Sep 2016 16:52:15 -0700 Subject: ubuntu-core image hygiene In-Reply-To: <1473856159.17795.120.camel@ubuntu.com> References: <1473856159.17795.120.camel@ubuntu.com> Message-ID: <31078ffc-daad-23bb-1bbf-2fc00eba4057@ubuntu.com> On 14/09/16 05:29, Oliver Grawert wrote: > I am happy to report that thanks to the switch i was also able to drop > about 50 lines of awful hacks from the build system that injected > packages at build time in an undocumented way :) > > Now to the more exciting bits ... > > Dropping all grub bits from the amd64 ubuntu-core snap got us from a > 75.1MB package size down to 66.9MB ... > > Likewise the i386 snap went from 72.1MB down to 64.9MB > > I know that a bunch of people will actually celebrate that massive size > reduction (hey manik :) ). Cleaner, faster, better. Thanks Ogra :) Mark -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From manik at canonical.com Wed Sep 14 23:59:20 2016 From: manik at canonical.com (Manik Taneja) Date: Wed, 14 Sep 2016 16:59:20 -0700 Subject: ubuntu-core image hygiene In-Reply-To: <1473856159.17795.120.camel@ubuntu.com> References: <1473856159.17795.120.camel@ubuntu.com> Message-ID: On Wed, Sep 14, 2016 at 5:29 AM, Oliver Grawert wrote: > hi, > > Tl;dr; > > - managing seeds for ubuntu-core changed from task to meta package > - x86 based ubuntu-core snaps lost multiple megabytes of fat ... > > And here the details: > > With the switch to ubuntu-image and the new gadget handling that comes > along with this, we were now able to drop all bits of bootloaders we > shipped inside the rootfs (bootloader binaries (specifically grub) are > actually shipped inside the gadget now and do not need to be copied > from the rootfs into the boot partition any more like we did in the > past with ubuntu-device-flash). > > While working on dropping grub from ubuntu-core one large problem arose > ... our seeds used tasks instead of a meta package. Since we can not > change the seed itself after a release is out (snappy is based on > xenial) this makes it impossible to actually drop packages ... at the > same time the only way to add new packages is to add very ugly hacks to > livecd-rootfs to import them directly during the rootfs build. > > Since the classic ubuntu has always had similar issues with the above > when doing point releases, it was actually completely switched from > tasks to meta packages (relatively recently) ... i did the same for > ubuntu-core now so that package addition/removal gets easier and to be > in sync with the main distro. > > If you need to add/remove a package in the ubuntu-core snap, please use > the ubuntu-core-meta package now (currently in the snappy-dev/image > PPA, it will be SRUed into xenial-updates proper before GA) > > I am happy to report that thanks to the switch i was also able to drop > about 50 lines of awful hacks from the build system that injected > packages at build time in an undocumented way :) > > Now to the more exciting bits ... > > Dropping all grub bits from the amd64 ubuntu-core snap got us from a > 75.1MB package size down to 66.9MB ... > > Likewise the i386 snap went from 72.1MB down to 64.9MB > > I know that a bunch of people will actually celebrate that massive size > reduction (hey manik :) ). > woot woot ;) a big thank you for cleaning this up. what's causing the difference still between the 2 architectures? /manik -------------- next part -------------- An HTML attachment was scrubbed... URL: From stuart.bishop at canonical.com Thu Sep 15 00:50:02 2016 From: stuart.bishop at canonical.com (Stuart Bishop) Date: Wed, 14 Sep 2016 17:50:02 -0700 Subject: Snap layer for Juju reactive charms Message-ID: Hi. I've created a Snap layer for writing Juju reactive charms, allowing you to easily install snaps and keep them updated. At deployment time, the user may opt to upload the snaps as Juju resources allowing them to deploy in network restricted environments. The default is for snaps to be pulled and refreshed from the snap store. The layer is registered at interfaces.juju.solutions and is available for use now. Code is available on https://launchpad.net/layer-snap and https://github.com/stub42/layer-snap A simple example of a charm using this layer is available at https://git.launchpad.net/~stub/+git/rocketchat (Rocketchat, not production ready due to absolutely zero configuration options and no HA). # Snap layer The Snap layer for Juju enables layered charms to more easily deal with snap packages in a simple and efficient manner. It provides consistent configuration for users, allowing them the choice of pulling snaps from the main snap store, or uploading them as Juju resources for deploys in environments with limited network access. ## Configuration To have the Snap layer install snaps automatically, declare the snaps in layer.yaml: ```yaml includes: - layer:basic - layer:snap options: snap: telegraf: channel: stable devmode: false jailmode: false force_dangerous: false revision: null ``` In addition, you should declare Juju resource slots for the snaps. This allows operators to have snaps distributed from their Juju controller node rather than the snap store, and is necessary for when your charm is deployed in network restricted environments. ```yaml resources: telegraf: type: file filename: telegraf.snap description: Telegraf snap ``` With the Juju resource defined, the operator may deploy your charm using locally provided snaps instead of the snap store: ```sh juju deploy --resource telegraf=telegraf_0_19.snap cs:telegraf ``` If your charm needs to control installation, update and removal of snaps itself then do not declare the snaps in layer.yaml. Instead, use the API provided by the `charms.layer.snap` Python package. ## Usage If you have defined your snaps in layer.yaml for automatic installation and updates, there is nothing further to do. ### API If your charm need to control installation, update and removal of snaps itself, the following methods are available via the `charms.layer.snap` package:: * `install(snapname, **args)`. Install the snap from the corresponding Juju resource (using --force-dangerous implicitly). If the resource is not available, download and install from the Snap store using the provided keyword arguments. * `refresh(snapname, **args)`. Update the snap. If the snap was installed from a local resource then the resource is checked for updates and the snap updated if the snap or arguments have changed. If the snap was installed from the Snap store, `snap refresh` is run to update the snap. * `remove(snapname)`. The snap is removed. Keyword arguments correspond to the layer.yaml options and snap command line options. See the snap command line documentation for authorative details on what these options do: * `channel` (str) * `devmode` (boolean) * `jailmode` (boolean) * `force_dangerous` (boolean) * `revision` (str) ## Support This layer is maintained on Launchpad by Stuart Bishop (stuart.bishop at canonical.com). Code is available using git at git+ssh://git.launchpad.net/layer-snap. Bug reports can be made at https://bugs.launchpad.net/layer-snap. Queries and comments can be made on the Juju mailing list, Juju IRC channels, or at https://answers.launchpad.net/layer-snap. -- Stuart Bishop -------------- next part -------------- An HTML attachment was scrubbed... URL: From michael.vogt at canonical.com Thu Sep 15 08:47:17 2016 From: michael.vogt at canonical.com (Michael Vogt) Date: Thu, 15 Sep 2016 10:47:17 +0200 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160907215054.GG3871@bod> References: <20160907215054.GG3871@bod> Message-ID: <20160915084717.GA27970@bod> Hi, a quick followup with some suggestions how to improve the user experience when working with the amd64 and pi2 images. On Wed, Sep 07, 2016 at 11:50:54PM +0200, Michael Vogt wrote: [..] > The images are bootable, the pc image can be booted directly in > qemu-kvm or virtualenv. Some more details about running the images inside qemu-kvm. When running the images in qemu-kvm it is helpful to use the "-redir" feature of qemu-kvm. E.g.: $ kvm -m 1500 -redir tcp:10022::22 ubuntu-core-16-amd64.img The message from console-conf is a bit misleading in this setup. It will say "ssh USER at 10.0.2.15". However due to the way that qemu-kvm user networking works, you will actually have to run: $ ssh -p 10022 USER at localhost to ssh into the images. We should probably include it in the instructions for images on snapcraft.io. > The pi2 image can be written to a sdcard via: > > unxz ubuntu-core-16-pc.img.xz > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > Where /dev/sdXX is the path of your sd card. [..] An alternative way to write the image is to use "go-dd": $ sudo snap install --devmode --beta godd $ sudo /snap/bin/godd ubuntu-core-16-pi2.img [this will print a message what devices are removable] $ sudo /snap/bin/godd ubuntu-core-16-pi2.img /dev/sdXX The advantage of godd is that it will not write to devices that are still mounted and that it can help detecting what removable devices are available (and it shows a progressbar). Cheers, Michael From ogra at ubuntu.com Thu Sep 15 10:52:32 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 15 Sep 2016 12:52:32 +0200 Subject: ubuntu-core image hygiene In-Reply-To: References: <1473856159.17795.120.camel@ubuntu.com> Message-ID: <1473936752.6549.7.camel@ubuntu.com> hi, On Mi, 2016-09-14 at 16:59 -0700, Manik Taneja wrote: >  > woot woot ;) a big thank you for cleaning this up. what's causing the > difference  > still between the 2 architectures? this is a matter of how the compilers assemble the binaries... 64bit binaries are usually a bit bigger than 32bit ones ...  along with that the dependencies per-arch can indeed slightly differ, a dependency on arch X might not be available or even needed on arch Y (the x86 arches are both above 60MB while the arm arches both are below ... ppc64el and s390x are somewhere in between ... ) for the dependency bit i can provide actual diffs once [1] is fixed ... ciao oli [1] https://bugs.launchpad.net/bugs/1608432 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From mark at ubuntu.com Thu Sep 15 11:56:59 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 15 Sep 2016 04:56:59 -0700 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160915084717.GA27970@bod> References: <20160907215054.GG3871@bod> <20160915084717.GA27970@bod> Message-ID: <2a5e9f5e-a569-7ed9-e965-169f51374e06@ubuntu.com> On 15/09/16 01:47, Michael Vogt wrote: > An alternative way to write the image is to use "go-dd": > $ sudo snap install --devmode --beta godd > $ sudo /snap/bin/godd ubuntu-core-16-pi2.img > [this will print a message what devices are removable] > > $ sudo /snap/bin/godd ubuntu-core-16-pi2.img /dev/sdXX > > The advantage of godd is that it will not write to devices that are > still mounted and that it can help detecting what removable devices > are available (and it shows a progressbar). Nice to see new software coming on board as a snap :) Mark From zygmunt.krynicki at canonical.com Thu Sep 15 12:00:48 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Thu, 15 Sep 2016 14:00:48 +0200 Subject: Access to other commands In-Reply-To: <57D9E08E.40807@canonical.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> Message-ID: <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> > On 15 Sep 2016, at 01:43, Leo Arias wrote: > > On 2016-09-07 19:31, Mark Shuttleworth wrote: >>> 1) Is there some way I can be specifying a list of commands my snapped >>>> app is allowed to call? >> You could. But I think there is a class of things that should be allowed >> to integrate with the classic shell environment, which means they can >> shell out to lots of things. > > What about --devmode? Should devmode allow calls to all binaries in the > $PATH? As discussed a few times this is technically challenging to do. All of “classic” is visible from /var/lib/snapd/hostfs/ but there is no guarantee that you can run them in any way. They may require the classic dynamic linker, the classic runtime libraries and the classic filesystem layout that are all lost when snap-confine sets up the execution environment. If there’s desire to run executables from the outside we could look for solutions but this is not as simple as “just use devmode” Thanks ZK From andreas at canonical.com Thu Sep 15 12:04:35 2016 From: andreas at canonical.com (Andreas Hasenack) Date: Thu, 15 Sep 2016 09:04:35 -0300 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160907215054.GG3871@bod> References: <20160907215054.GG3871@bod> Message-ID: Does anybody know if there will be pandaboard (omap4) images? On Sep 7, 2016 18:52, "Michael Vogt" wrote: > Ubuntu Core 16 Images > ===================== > > The Ubuntu snappy team is happy to announce the first beta images for > Ubuntu Core 16. The images use the snapd package manager to install > and update all components of the system including kernel, core, gadget > and applications. > > The images are available for PC (amd64, i386) and Raspberry Pi2 > (armhf). More architectures and boards (arm64 dragonboard, pi3) will > follow shortly. You can download them at: > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > > The images are bootable, the pc image can be booted directly in > qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: > > unxz ubuntu-core-16-pc.img.xz > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > Where /dev/sdXX is the path of your sd card. > > After booting the image you can enter your Ubuntu SSO email and it > will automatically create a matching user with the right ssh keys. If > you do not have an Ubuntu SSO account yet you can create one at: > > https://login.ubuntu.com/ > > These images follow the "beta" channel. > > Enjoy the fresh images! If you find any bugs or issues, please let us > know via: > > https://bugs.launchpad.net/snappy/ > > Cheers, > Michael (on behalf of the snappy team) > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Thu Sep 15 12:12:56 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 15 Sep 2016 05:12:56 -0700 Subject: Access to other commands In-Reply-To: <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> Message-ID: <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> On 15/09/16 05:00, Zygmunt Krynicki wrote: >> On 15 Sep 2016, at 01:43, Leo Arias wrote: >> >> On 2016-09-07 19:31, Mark Shuttleworth wrote: >>>> 1) Is there some way I can be specifying a list of commands my snapped >>>>> app is allowed to call? >>> You could. But I think there is a class of things that should be allowed >>> to integrate with the classic shell environment, which means they can >>> shell out to lots of things. >> What about --devmode? Should devmode allow calls to all binaries in the >> $PATH? > As discussed a few times this is technically challenging to do. > > All of “classic” is visible from /var/lib/snapd/hostfs/ but there is no guarantee that you can run them in any way. They may require the classic dynamic linker, the classic runtime libraries and the classic filesystem layout that are all lost when snap-confine sets up the execution environment. If there’s desire to run executables from the outside we could look for solutions but this is not as simple as “just use devmode” I think this is a topic for the next snapfest community event, in October/November. Call it "snapping CLI utilities". Mark From ogra at ubuntu.com Thu Sep 15 12:35:45 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 15 Sep 2016 14:35:45 +0200 Subject: Ubuntu Core 16 beta images available In-Reply-To: References: <20160907215054.GG3871@bod> Message-ID: <1473942945.6549.11.camel@ubuntu.com> hi, On Do, 2016-09-15 at 09:04 -0300, Andreas Hasenack wrote: > Does anybody know if there will be pandaboard (omap4) images? > i dont think pandaboard is a valid target anymore, the production of the board stopped in 2012, at the same time TI laid off everyone involved with omap4 and the panda board on the software and hardware side. there is no community support around it either anymore (if you go to the #pnadaboard IRC channel on freenode you will be told to find other hardware) if you feel like rolling a kernel and gadget snap nobody will stop you though, but in general the pandaboard is dead beef nowadays. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From ehbello at gmail.com Thu Sep 15 13:13:17 2016 From: ehbello at gmail.com (=?UTF-8?Q?Enrique_Hern=C3=A1ndez_Bello?=) Date: Thu, 15 Sep 2016 14:13:17 +0100 Subject: Ubuntu Core 16 beta images available In-Reply-To: <1473942945.6549.11.camel@ubuntu.com> References: <20160907215054.GG3871@bod> <1473942945.6549.11.camel@ubuntu.com> Message-ID: Hello, I'm working on that, based in a previous work of Gumstix guys. You can check my GitHub repos for the Gadget[1] snap and a working Kernel[2] for this board. Currently all seems to work succesfully except the graphics support (need a valid omap_dri driver for kernel) and maybe the bluetooth, that I can't tested it. Look at README.md files to get instructions to build the image. Any help is welcome. [1] https://github.com/ehbello/armhf-snaps [2] https://github.com/ehbello/linux-panda-snap 2016-09-15 13:35 GMT+01:00 Oliver Grawert : > hi, > On Do, 2016-09-15 at 09:04 -0300, Andreas Hasenack wrote: >> Does anybody know if there will be pandaboard (omap4) images? >> > i dont think pandaboard is a valid target anymore, the production of > the board stopped in 2012, at the same time TI laid off everyone > involved with omap4 and the panda board on the software and hardware > side. > > there is no community support around it either anymore (if you go to > the #pnadaboard IRC channel on freenode you will be told to find other > hardware) > > if you feel like rolling a kernel and gadget snap nobody will stop you > though, but in general the pandaboard is dead beef nowadays. > > ciao > oli > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > From andreas at canonical.com Thu Sep 15 13:26:41 2016 From: andreas at canonical.com (Andreas Hasenack) Date: Thu, 15 Sep 2016 10:26:41 -0300 Subject: Ubuntu Core 16 beta images available In-Reply-To: References: <20160907215054.GG3871@bod> <1473942945.6549.11.camel@ubuntu.com> Message-ID: On Thu, Sep 15, 2016 at 10:13 AM, Enrique Hernández Bello wrote: > Hello, > > I'm working on that, based in a previous work of Gumstix guys. > > You can check my GitHub repos for the Gadget[1] snap and a working > Kernel[2] for this board. > > Currently all seems to work succesfully except the graphics support > (need a valid omap_dri driver for kernel) and maybe the bluetooth, > that I can't tested it. > Thanks! I don't care about graphic drivers or bt for this board, so it should be fine. I'll give it a whirl -------------- next part -------------- An HTML attachment was scrubbed... URL: From jim.hodapp at canonical.com Thu Sep 15 13:30:42 2016 From: jim.hodapp at canonical.com (Jim Hodapp) Date: Thu, 15 Sep 2016 09:30:42 -0400 Subject: Ubuntu Core 16 beta images available In-Reply-To: <20160915084717.GA27970@bod> References: <20160907215054.GG3871@bod> <20160915084717.GA27970@bod> Message-ID: On Thu, Sep 15, 2016 at 4:47 AM, Michael Vogt wrote: > Hi, > > a quick followup with some suggestions how to improve the user > experience when working with the amd64 and pi2 images. > > On Wed, Sep 07, 2016 at 11:50:54PM +0200, Michael Vogt wrote: > [..] > > The images are bootable, the pc image can be booted directly in > > qemu-kvm or virtualenv. > > Some more details about running the images inside qemu-kvm. > > When running the images in qemu-kvm it is helpful to use the "-redir" > feature of qemu-kvm. E.g.: > > $ kvm -m 1500 -redir tcp:10022::22 ubuntu-core-16-amd64.img > > The message from console-conf is a bit misleading in this setup. It > will say "ssh USER at 10.0.2.15". However due to the way that qemu-kvm > user networking works, you will actually have to run: > > $ ssh -p 10022 USER at localhost > > to ssh into the images. We should probably include it in the > instructions for images on snapcraft.io. > I think this is a great idea. I think it'd be very useful to have this information on snapcraft.io, or at the very least to start a wiki page with this info. > > > The pi2 image can be written to a sdcard via: > > > > unxz ubuntu-core-16-pc.img.xz > > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > > > > Where /dev/sdXX is the path of your sd card. > [..] > > An alternative way to write the image is to use "go-dd": > > $ sudo snap install --devmode --beta godd > $ sudo /snap/bin/godd ubuntu-core-16-pi2.img > [this will print a message what devices are removable] > > $ sudo /snap/bin/godd ubuntu-core-16-pi2.img /dev/sdXX > > The advantage of godd is that it will not write to devices that are > still mounted and that it can help detecting what removable devices > are available (and it shows a progressbar). > > Cheers, > Michael > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Thu Sep 15 13:41:19 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 15 Sep 2016 15:41:19 +0200 Subject: Ubuntu Core 16 beta images available In-Reply-To: References: <20160907215054.GG3871@bod> <1473942945.6549.11.camel@ubuntu.com> Message-ID: <1473946879.6549.14.camel@ubuntu.com> hi, On Do, 2016-09-15 at 14:13 +0100, Enrique Hernández Bello wrote: > Hello, > > I'm working on that, based in a previous work of Gumstix guys. > > You can check my GitHub repos for the Gadget[1] snap and a working > Kernel[2] for this board. ... > > [1] https://github.com/ehbello/armhf-snaps > wow, this is impressive i even see asserions and all ... you should blog about it so we can share it on social media ! ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From robert.park at canonical.com Thu Sep 15 15:11:07 2016 From: robert.park at canonical.com (Robert Park) Date: Thu, 15 Sep 2016 08:11:07 -0700 Subject: Access to other commands In-Reply-To: <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> Message-ID: On Thu, Sep 15, 2016 at 5:00 AM, Zygmunt Krynicki wrote: > As discussed a few times this is technically challenging to do. > > All of “classic” is visible from /var/lib/snapd/hostfs/ but there is no guarantee that you can run them in any way. They may require the classic dynamic linker, the classic runtime libraries and the classic filesystem layout that are all lost when snap-confine sets up the execution environment. If there’s desire to run executables from the outside we could look for solutions but this is not as simple as “just use devmode” Yes, see my other thread where I have a script that needs to add a PPA, even running in devmode and chrooting back to /var/lib/snapd/hostfs, everything is still broken. Would be great if snap offered some alternative for shell scripts that need to just run unrestricted on the host system. -- robru From mark at ubuntu.com Thu Sep 15 16:11:17 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 15 Sep 2016 09:11:17 -0700 Subject: Access to other commands In-Reply-To: References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> Message-ID: On 15/09/16 08:11, Robert Park wrote: > On Thu, Sep 15, 2016 at 5:00 AM, Zygmunt Krynicki > wrote: >> As discussed a few times this is technically challenging to do. >> >> All of “classic” is visible from /var/lib/snapd/hostfs/ but there is no guarantee that you can run them in any way. They may require the classic dynamic linker, the classic runtime libraries and the classic filesystem layout that are all lost when snap-confine sets up the execution environment. If there’s desire to run executables from the outside we could look for solutions but this is not as simple as “just use devmode” > Yes, see my other thread where I have a script that needs to add a > PPA, even running in devmode and chrooting back to > /var/lib/snapd/hostfs, everything is still broken. Would be great if > snap offered some alternative for shell scripts that need to just run > unrestricted on the host system. Right, I think this is a discussion about snapping up CLI utilities. It would be great to have an awk-of-the-day :) Mark From jamie.bennett at canonical.com Thu Sep 15 17:57:47 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Thu, 15 Sep 2016 18:57:47 +0100 Subject: Ubuntu Core 16 beta images available In-Reply-To: <1473946879.6549.14.camel@ubuntu.com> References: <20160907215054.GG3871@bod> <1473942945.6549.11.camel@ubuntu.com> <1473946879.6549.14.camel@ubuntu.com> Message-ID: <60C3F8EA-330A-4BBA-8070-4291139752CB@canonical.com> > On 15 Sep 2016, at 14:41, Oliver Grawert wrote: > > hi, > On Do, 2016-09-15 at 14:13 +0100, Enrique Hernández Bello wrote: >> Hello, >> >> I'm working on that, based in a previous work of Gumstix guys. >> >> You can check my GitHub repos for the Gadget[1] snap and a working >> Kernel[2] for this board. > ... >> >> [1] https://github.com/ehbello/armhf-snaps >> > > wow, this is impressive i even see asserions and all ... you should > blog about it so we can share it on social media ! +1, definitely. > ciao > oli-- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From paintmecyanide at gmail.com Thu Sep 15 19:09:25 2016 From: paintmecyanide at gmail.com (Michael) Date: Thu, 15 Sep 2016 15:09:25 -0400 Subject: Ubuntu Core 16 beta images available Message-ID: Hello, The RPi3 image has been deleted. Was this intentional? I have the file, but if this was intentionally taken down, I probably don't want to flash it to my card. I did get the "unxz: ubuntu-core-16-pi3.img.xz: Unexpected end of input" error noted by others when I extracted it. Thanks >Ubuntu Core 16 Images >===================== > >The Ubuntu snappy team is happy to announce the first beta images for >Ubuntu Core 16. The images use the snapd package manager to install >and update all components of the system including kernel, core, gadget >and applications. > >The images are available for PC (amd64, i386) and Raspberry Pi2 >(armhf). More architectures and boards (arm64 dragonboard, pi3) will >follow shortly. You can download them at: > > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ > >The images are bootable, the pc image can be booted directly in >qemu-kvm or virtualenv. The pi2 image can be written to a sdcard via: > > unxz ubuntu-core-16-pc.img.xz > dd if= ubuntu-core-16-pc.img of=/dev/sdXX > >Where /dev/sdXX is the path of your sd card. > >After booting the image you can enter your Ubuntu SSO email and it >will automatically create a matching user with the right ssh keys. If >you do not have an Ubuntu SSO account yet you can create one at: > > https://login.ubuntu.com/ > >These images follow the "beta" channel. > >Enjoy the fresh images! If you find any bugs or issues, please let us >know via: > > https://bugs.launchpad.net/snappy/ > >Cheers, > Michael (on behalf of the snappy team) > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.chiluk at canonical.com Thu Sep 15 21:10:32 2016 From: dave.chiluk at canonical.com (Dave Chiluk) Date: Thu, 15 Sep 2016 16:10:32 -0500 Subject: How do you add config files - "file" can't find it's configuration files? Message-ID: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> Hey guys, I'm trying to snap up file. I thought it would be simple. This is what I have for my snapcraft.yaml. vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv name: file # the name of the snap version: 0 # the version of the snap summary: Unix file utility # 79 char long summary description: Unix file utility # a longer description for the snap confinement: devmode # use "strict" to enforce system access only via declared interfaces apps: file: command: usr/bin/file parts: file: # Replace with a part name of your liking # Get more information about plugins by running # snapcraft help plugins # and more information about the available plugins # by running # snapcraft list-plugins plugin: nil stage-packages: - file - libmagic1 filesets: libmagic1: - etc/magic* - usr/* file: - usr/bin/file snap: - $libmagic1 - $file ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ However when I go to run file out of my snap I get " chiluk at localhost:~$ file /home/chiluk/file_0_amd64.snap /etc/magic, 0: Warning: using regular magic file `/usr/share/misc/magic' file: could not find any valid magic files! " Both of these files are included with my snap, but not being found when running "file /home/chiluk/ Feel free to post links to documentation/examples as searches for ubuntu snapcraft add config file have been useless. I'm mostly trying to understand / learn so simply fixing this yaml is not what I'm looking for. Also once file is capable of finding it's configs, how do you make them customizable such that root could modify them? Thanks, Dave. From spencertparkin at gmail.com Thu Sep 15 21:28:31 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Thu, 15 Sep 2016 15:28:31 -0600 Subject: candidate/release Message-ID: Hi, I'm embarrassed to say that while, supposedly, I have published a candidate and a stable release, I have no idea how to "go get them" the same way that, say, a user would. I run "snap find ", but it doesn't find it, and I can't find it on any store front page. Perhaps there's still some vetting to be done. That's fine. If I've pushed up a beta or candidate, though, shouldn't I be able to do something with it? Is all this documented somewhere in a completely obvious place that I have completely overlooked? Here's hoping my first post to the mailing list doesn't immediately alienate me from everyone... Thanks, --Spencer -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Thu Sep 15 21:37:00 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 15 Sep 2016 14:37:00 -0700 Subject: candidate/release In-Reply-To: References: Message-ID: On 15/09/16 14:28, Spencer Parkin wrote: > I'm embarrassed to say that while, supposedly, I have published a > candidate and a stable release, I have no idea how to "go get them" > the same way that, say, a user would. I run "snap find > ", but it doesn't find it, and I can't find it on any > store front page. Perhaps there's still some vetting to be done. > That's fine. If I've pushed up a beta or candidate, though, shouldn't > I be able to do something with it? Is all this documented somewhere > in a completely obvious place that I have completely overlooked? Try 'sudo snap install --edge ' Have you done "snapcraft release" or just "snapcraft push"? Or snapcraft push --release=? > Here's hoping my first post to the mailing list doesn't immediately > alienate me from everyone... You're crazy, get out of my office :) Of course not, this is the place for questions, and if the docs aren't found then they are not completely obvious by definition, so thanks for at least getting this on the Googlebot's radar. Mark From sergio.schvezov at canonical.com Thu Sep 15 22:56:30 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Thu, 15 Sep 2016 19:56:30 -0300 Subject: candidate/release In-Reply-To: References: Message-ID: <850eef05-d314-4306-88e6-b5844f7fe6f1@canonical.com> El jueves, 15 de septiembre de 2016 18h'37:00 ART, Mark Shuttleworth escribió: > On 15/09/16 14:28, Spencer Parkin wrote: >> I'm embarrassed to say that while, supposedly, I have published a >> candidate and a stable release, I have no idea how to "go get them" >> the same way that, say, a user would. I run "snap find >> ", but it doesn't find it, and I can't find it on any >> store front page. Perhaps there's still some vetting to be done. >> That's fine. If I've pushed up a beta or candidate, though, shouldn't >> I be able to do something with it? Is all this documented somewhere >> in a completely obvious place that I have completely overlooked? > > Try 'sudo snap install --edge ' > > Have you done "snapcraft release" or just "snapcraft push"? Or snapcraft > push --release=? Just to expand a bit, if you did `snapcraft release` or `snapcraft push --release` you should see a channel mapping for the current state of your snap. If you did see this, and the output is correct then you did indeed do the right thing. As a side note, we are working on some new APIs to get to query the status of your snap on the store as well. Evan might be able to figure out what is going on store side if needed. Cheers Sergio -- Enviado con Dekko desde mi dispositivo Ubuntu From spencertparkin at gmail.com Fri Sep 16 02:16:39 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Thu, 15 Sep 2016 20:16:39 -0600 Subject: candidate/release In-Reply-To: <850eef05-d314-4306-88e6-b5844f7fe6f1@canonical.com> References: <850eef05-d314-4306-88e6-b5844f7fe6f1@canonical.com> Message-ID: Thanks all for your help. I followed the following instructions... https://developer.ubuntu.com/en/snappy/build-apps/upload-your-snap/ ...then, seeing the snap passed with no warnings or errors, pushed the "Publish" button at... https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/2/ It's been published now for 2 days. Maybe the system is still churning on it. I suppose, too, there's a way to do the publishing from the command line instead as some of you have pointed out. Reading through "snapcraft -h", I see that the upload option is deprecated. Perhaps the first URL I cited above should be revised to reflect this. Anyhow, it should already be published, but all variants of "sudo snap install [myprojname]" have failed with "snap not found." I've gone ahead and tried... $ snapcraft push [mysnapfile] This chugs for a while during the upload, then reports that an error occurred when trying to analyze the snap. Maybe I'll try rebuilding the snap and pushing it again later when I get some time. I'm running version 2.16 of snapcraft, which I believe is the latest version. Oh wait...no...version 2.17 is the latest. I did... $ snapcraft release [myproj] 2 stable ...and that appeared to work, or so it said. Oh, I got it to work now! I see what I did wrong. I had revision 1 released to the masses, which had errors. But revision 2 didn't, so I guess now the "snap" command could find it in the store. If you try to release a snap that had errors, it won't get found or be installable. Hurrah! That worked! Okay, thanks guys. My crappy app is now live. ;) This is truly a momentous day for us all. On Thu, Sep 15, 2016 at 4:56 PM, Sergio Schvezov < sergio.schvezov at canonical.com> wrote: > El jueves, 15 de septiembre de 2016 18h'37:00 ART, Mark Shuttleworth < > mark at ubuntu.com> escribió: > >> On 15/09/16 14:28, Spencer Parkin wrote: >> >>> I'm embarrassed to say that while, supposedly, I have published a >>> candidate and a stable release, I have no idea how to "go get them" >>> the same way that, say, a user would. I run "snap find >>> ", but it doesn't find it, and I can't find it on any >>> store front page. Perhaps there's still some vetting to be done. That's >>> fine. If I've pushed up a beta or candidate, though, shouldn't >>> I be able to do something with it? Is all this documented somewhere >>> in a completely obvious place that I have completely overlooked? >>> >> >> Try 'sudo snap install --edge ' >> >> Have you done "snapcraft release" or just "snapcraft push"? Or snapcraft >> push --release=? >> > > Just to expand a bit, if you did `snapcraft release` or `snapcraft push > --release` you should see a channel mapping for the current state of your > snap. If you did see this, and the output is correct then you did indeed do > the right thing. > > As a side note, we are working on some new APIs to get to query the status > of your snap on the store as well. > > Evan might be able to figure out what is going on store side if needed. > > Cheers > Sergio > > > > -- > Enviado con Dekko desde mi dispositivo Ubuntu > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mhall119 at ubuntu.com Fri Sep 16 02:32:37 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Thu, 15 Sep 2016 22:32:37 -0400 Subject: candidate/release In-Reply-To: References: <850eef05-d314-4306-88e6-b5844f7fe6f1@canonical.com> Message-ID: <1cf64e82-f068-a0e7-64c5-b6d808f354c4@ubuntu.com> If an app fails the automated review, it won't be published at all, which is what happened here. In the overview page for your app (https://myapps.developer.ubuntu.com/dev/click-apps/5902/) you will see a red exclamation mark next to revision #1, which indicates that it had some issues and was not published. If you click on that (https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/1/) and scroll down to the review and comment history, it will tell you what the problem was (and Jamie gave advice on how to correct it). Since this was manual feedback from a human reviewer, I'm not sure how we can get this across to the snapcraft commandline interface. Perhaps the new APIs sergio mentioned for querying status can include them. Michael Hall mhall119 at ubuntu.com On 09/15/2016 10:16 PM, Spencer Parkin wrote: > Thanks all for your help. I followed the following instructions... > > https://developer.ubuntu.com/en/snappy/build-apps/upload-your-snap/ > > ...then, seeing the snap passed with no warnings or errors, pushed the > "Publish" button at... > > https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/2/ > > It's been published now for 2 days. Maybe the system is still churning > on it. I suppose, too, there's a way to do the publishing from the > command line instead as some of you have pointed out. > > Reading through "snapcraft -h", I see that the upload option is > deprecated. Perhaps the first URL I cited above should be revised to > reflect this. > > Anyhow, it should already be published, but all variants of "sudo snap > install [myprojname]" have failed with "snap not found." > > I've gone ahead and tried... > > $ snapcraft push [mysnapfile] > > This chugs for a while during the upload, then reports that an error > occurred when trying to analyze the snap. Maybe I'll try rebuilding the > snap and pushing it again later when I get some time. > > I'm running version 2.16 of snapcraft, which I believe is the latest > version. Oh wait...no...version 2.17 is the latest. > > I did... > > $ snapcraft release [myproj] 2 stable > > ...and that appeared to work, or so it said. > > Oh, I got it to work now! I see what I did wrong. I had revision 1 > released to the masses, which had errors. But revision 2 didn't, so I > guess now the "snap" command could find it in the store. If you try to > release a snap that had errors, it won't get found or be installable. > > Hurrah! That worked! Okay, thanks guys. My crappy app is now live. > ;) This is truly a momentous day for us all. > > On Thu, Sep 15, 2016 at 4:56 PM, Sergio Schvezov > > > wrote: > > El jueves, 15 de septiembre de 2016 18h'37:00 ART, Mark Shuttleworth > > escribió: > > On 15/09/16 14:28, Spencer Parkin wrote: > > I'm embarrassed to say that while, supposedly, I have > published a > candidate and a stable release, I have no idea how to "go > get them" > the same way that, say, a user would. I run "snap find > ", but it doesn't find it, and I can't find > it on any > store front page. Perhaps there's still some vetting to be > done. That's fine. If I've pushed up a beta or candidate, > though, shouldn't > I be able to do something with it? Is all this documented > somewhere > in a completely obvious place that I have completely overlooked? > > > Try 'sudo snap install --edge ' > > Have you done "snapcraft release" or just "snapcraft push"? Or > snapcraft > push --release=? > > > Just to expand a bit, if you did `snapcraft release` or `snapcraft > push --release` you should see a channel mapping for the current > state of your snap. If you did see this, and the output is correct > then you did indeed do the right thing. > > As a side note, we are working on some new APIs to get to query the > status of your snap on the store as well. > > Evan might be able to figure out what is going on store side if needed. > > Cheers > Sergio > > > > -- > Enviado con Dekko desde mi dispositivo Ubuntu > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > From spencertparkin at gmail.com Fri Sep 16 03:22:26 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Thu, 15 Sep 2016 21:22:26 -0600 Subject: candidate/release In-Reply-To: <1cf64e82-f068-a0e7-64c5-b6d808f354c4@ubuntu.com> References: <850eef05-d314-4306-88e6-b5844f7fe6f1@canonical.com> <1cf64e82-f068-a0e7-64c5-b6d808f354c4@ubuntu.com> Message-ID: Thanks, my guess is this: The snapcraft command (and the "Publish" button on the website) let you do... $ snapcraft release [myproj] X stable ...where X is any revision of your project, even one that did _not_ pass automated scrutiny. The command (and the website) tell you that publishing is successful, but it really wasn't. It would be nice if the command (and the website) gave feedback that it rejected publication due to the desired revision not having passed inspection. --Sp On Thu, Sep 15, 2016 at 8:32 PM, Michael Hall wrote: > If an app fails the automated review, it won't be published at all, > which is what happened here. > > In the overview page for your app > (https://myapps.developer.ubuntu.com/dev/click-apps/5902/) you will see > a red exclamation mark next to revision #1, which indicates that it had > some issues and was not published. > > If you click on that > (https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/1/) and > scroll down to the review and comment history, it will tell you what the > problem was (and Jamie gave advice on how to correct it). > > Since this was manual feedback from a human reviewer, I'm not sure how > we can get this across to the snapcraft commandline interface. Perhaps > the new APIs sergio mentioned for querying status can include them. > > Michael Hall > mhall119 at ubuntu.com > > On 09/15/2016 10:16 PM, Spencer Parkin wrote: > > Thanks all for your help. I followed the following instructions... > > > > https://developer.ubuntu.com/en/snappy/build-apps/upload-your-snap/ > > > > ...then, seeing the snap passed with no warnings or errors, pushed the > > "Publish" button at... > > > > https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/2/ > > > > It's been published now for 2 days. Maybe the system is still churning > > on it. I suppose, too, there's a way to do the publishing from the > > command line instead as some of you have pointed out. > > > > Reading through "snapcraft -h", I see that the upload option is > > deprecated. Perhaps the first URL I cited above should be revised to > > reflect this. > > > > Anyhow, it should already be published, but all variants of "sudo snap > > install [myprojname]" have failed with "snap not found." > > > > I've gone ahead and tried... > > > > $ snapcraft push [mysnapfile] > > > > This chugs for a while during the upload, then reports that an error > > occurred when trying to analyze the snap. Maybe I'll try rebuilding the > > snap and pushing it again later when I get some time. > > > > I'm running version 2.16 of snapcraft, which I believe is the latest > > version. Oh wait...no...version 2.17 is the latest. > > > > I did... > > > > $ snapcraft release [myproj] 2 stable > > > > ...and that appeared to work, or so it said. > > > > Oh, I got it to work now! I see what I did wrong. I had revision 1 > > released to the masses, which had errors. But revision 2 didn't, so I > > guess now the "snap" command could find it in the store. If you try to > > release a snap that had errors, it won't get found or be installable. > > > > Hurrah! That worked! Okay, thanks guys. My crappy app is now live. > > ;) This is truly a momentous day for us all. > > > > On Thu, Sep 15, 2016 at 4:56 PM, Sergio Schvezov > > > > > wrote: > > > > El jueves, 15 de septiembre de 2016 18h'37:00 ART, Mark Shuttleworth > > > escribió: > > > > On 15/09/16 14:28, Spencer Parkin wrote: > > > > I'm embarrassed to say that while, supposedly, I have > > published a > > candidate and a stable release, I have no idea how to "go > > get them" > > the same way that, say, a user would. I run "snap find > > ", but it doesn't find it, and I can't find > > it on any > > store front page. Perhaps there's still some vetting to be > > done. That's fine. If I've pushed up a beta or candidate, > > though, shouldn't > > I be able to do something with it? Is all this documented > > somewhere > > in a completely obvious place that I have completely > overlooked? > > > > > > Try 'sudo snap install --edge ' > > > > Have you done "snapcraft release" or just "snapcraft push"? Or > > snapcraft > > push --release=? > > > > > > Just to expand a bit, if you did `snapcraft release` or `snapcraft > > push --release` you should see a channel mapping for the current > > state of your snap. If you did see this, and the output is correct > > then you did indeed do the right thing. > > > > As a side note, we are working on some new APIs to get to query the > > status of your snap on the store as well. > > > > Evan might be able to figure out what is going on store side if > needed. > > > > Cheers > > Sergio > > > > > > > > -- > > Enviado con Dekko desde mi dispositivo Ubuntu > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > > > > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From spencertparkin at gmail.com Fri Sep 16 03:27:20 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Thu, 15 Sep 2016 21:27:20 -0600 Subject: candidate/release In-Reply-To: References: <850eef05-d314-4306-88e6-b5844f7fe6f1@canonical.com> <1cf64e82-f068-a0e7-64c5-b6d808f354c4@ubuntu.com> Message-ID: Err...sorry...I really think this all just user error on my part. Never mind. No more spam. Good night. On Thu, Sep 15, 2016 at 9:22 PM, Spencer Parkin wrote: > Thanks, my guess is this: The snapcraft command (and the "Publish" button > on the website) let you do... > > $ snapcraft release [myproj] X stable > > ...where X is any revision of your project, even one that did _not_ pass > automated scrutiny. The command (and the website) tell you that publishing > is successful, but it really wasn't. It would be nice if the command (and > the website) gave feedback that it rejected publication due to the desired > revision not having passed inspection. > > --Sp > > > > On Thu, Sep 15, 2016 at 8:32 PM, Michael Hall wrote: > >> If an app fails the automated review, it won't be published at all, >> which is what happened here. >> >> In the overview page for your app >> (https://myapps.developer.ubuntu.com/dev/click-apps/5902/) you will see >> a red exclamation mark next to revision #1, which indicates that it had >> some issues and was not published. >> >> If you click on that >> (https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/1/) and >> scroll down to the review and comment history, it will tell you what the >> problem was (and Jamie gave advice on how to correct it). >> >> Since this was manual feedback from a human reviewer, I'm not sure how >> we can get this across to the snapcraft commandline interface. Perhaps >> the new APIs sergio mentioned for querying status can include them. >> >> Michael Hall >> mhall119 at ubuntu.com >> >> On 09/15/2016 10:16 PM, Spencer Parkin wrote: >> > Thanks all for your help. I followed the following instructions... >> > >> > https://developer.ubuntu.com/en/snappy/build-apps/upload-your-snap/ >> > >> > ...then, seeing the snap passed with no warnings or errors, pushed the >> > "Publish" button at... >> > >> > https://myapps.developer.ubuntu.com/dev/click-apps/5902/rev/2/ >> > >> > It's been published now for 2 days. Maybe the system is still churning >> > on it. I suppose, too, there's a way to do the publishing from the >> > command line instead as some of you have pointed out. >> > >> > Reading through "snapcraft -h", I see that the upload option is >> > deprecated. Perhaps the first URL I cited above should be revised to >> > reflect this. >> > >> > Anyhow, it should already be published, but all variants of "sudo snap >> > install [myprojname]" have failed with "snap not found." >> > >> > I've gone ahead and tried... >> > >> > $ snapcraft push [mysnapfile] >> > >> > This chugs for a while during the upload, then reports that an error >> > occurred when trying to analyze the snap. Maybe I'll try rebuilding the >> > snap and pushing it again later when I get some time. >> > >> > I'm running version 2.16 of snapcraft, which I believe is the latest >> > version. Oh wait...no...version 2.17 is the latest. >> > >> > I did... >> > >> > $ snapcraft release [myproj] 2 stable >> > >> > ...and that appeared to work, or so it said. >> > >> > Oh, I got it to work now! I see what I did wrong. I had revision 1 >> > released to the masses, which had errors. But revision 2 didn't, so I >> > guess now the "snap" command could find it in the store. If you try to >> > release a snap that had errors, it won't get found or be installable. >> > >> > Hurrah! That worked! Okay, thanks guys. My crappy app is now live. >> > ;) This is truly a momentous day for us all. >> > >> > On Thu, Sep 15, 2016 at 4:56 PM, Sergio Schvezov >> > > >> > wrote: >> > >> > El jueves, 15 de septiembre de 2016 18h'37:00 ART, Mark Shuttleworth >> > > escribió: >> > >> > On 15/09/16 14:28, Spencer Parkin wrote: >> > >> > I'm embarrassed to say that while, supposedly, I have >> > published a >> > candidate and a stable release, I have no idea how to "go >> > get them" >> > the same way that, say, a user would. I run "snap find >> > ", but it doesn't find it, and I can't find >> > it on any >> > store front page. Perhaps there's still some vetting to be >> > done. That's fine. If I've pushed up a beta or candidate, >> > though, shouldn't >> > I be able to do something with it? Is all this documented >> > somewhere >> > in a completely obvious place that I have completely >> overlooked? >> > >> > >> > Try 'sudo snap install --edge ' >> > >> > Have you done "snapcraft release" or just "snapcraft push"? Or >> > snapcraft >> > push --release=? >> > >> > >> > Just to expand a bit, if you did `snapcraft release` or `snapcraft >> > push --release` you should see a channel mapping for the current >> > state of your snap. If you did see this, and the output is correct >> > then you did indeed do the right thing. >> > >> > As a side note, we are working on some new APIs to get to query the >> > status of your snap on the store as well. >> > >> > Evan might be able to figure out what is going on store side if >> needed. >> > >> > Cheers >> > Sergio >> > >> > >> > >> > -- >> > Enviado con Dekko desde mi dispositivo Ubuntu >> > >> > -- >> > Snapcraft mailing list >> > Snapcraft at lists.snapcraft.io >> > Modify settings or unsubscribe at: >> > https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > >> > >> > >> > >> > >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.chiluk at canonical.com Fri Sep 16 03:35:29 2016 From: dave.chiluk at canonical.com (Dave Chiluk) Date: Thu, 15 Sep 2016 22:35:29 -0500 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> Message-ID: So I clearly what's going on here is that file has /etc/magic hardcoded in it's c-code. Unfortunately it's not being installed there. Certainly hard-coded path names in C code, is not new. How are people snapping up similar C programs with hard-coded pathnames? Thanks, Dave. Some people have mentioned that they've had to mod On 09/15/2016 04:10 PM, Dave Chiluk wrote: > Hey guys, I'm trying to snap up file. I thought it would be simple. > This is what I have for my snapcraft.yaml. > > > vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv > name: file # the name of the snap > version: 0 # the version of the snap > summary: Unix file utility # 79 char long summary > description: Unix file utility # a longer description for the snap > confinement: devmode # use "strict" to enforce system access only via > declared interfaces > > apps: > file: > command: usr/bin/file > > parts: > file: # Replace with a part name of your liking > # Get more information about plugins by running > # snapcraft help plugins > # and more information about the available plugins > # by running > # snapcraft list-plugins > plugin: nil > stage-packages: > - file > - libmagic1 > filesets: > libmagic1: > - etc/magic* > - usr/* > file: > - usr/bin/file > snap: > - $libmagic1 > - $file > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > However when I go to run file out of my snap I get > " > chiluk at localhost:~$ file /home/chiluk/file_0_amd64.snap > /etc/magic, 0: Warning: using regular magic file `/usr/share/misc/magic' > file: could not find any valid magic files! > " > > Both of these files are included with my snap, but not being found when > running "file /home/chiluk/ Feel free to post links to > documentation/examples as searches for ubuntu snapcraft add config file > have been useless. I'm mostly trying to understand / learn so simply > fixing this yaml is not what I'm looking for. > > Also once file is capable of finding it's configs, how do you make them > customizable such that root could modify them? > > Thanks, > Dave. > > From manik at canonical.com Fri Sep 16 04:52:05 2016 From: manik at canonical.com (Manik Taneja) Date: Thu, 15 Sep 2016 21:52:05 -0700 Subject: ubuntu-core image hygiene In-Reply-To: <1473936752.6549.7.camel@ubuntu.com> References: <1473856159.17795.120.camel@ubuntu.com> <1473936752.6549.7.camel@ubuntu.com> Message-ID: On Thu, Sep 15, 2016 at 3:52 AM, Oliver Grawert wrote: > hi, > On Mi, 2016-09-14 at 16:59 -0700, Manik Taneja wrote: > > > > woot woot ;) a big thank you for cleaning this up. what's causing the > > difference > > still between the 2 architectures? > > this is a matter of how the compilers assemble the binaries... > > 64bit binaries are usually a bit bigger than 32bit ones ... > > along with that the dependencies per-arch can indeed slightly differ, a > dependency on arch X might not be available or even needed on arch Y > > (the x86 arches are both above 60MB while the arm arches both are below > ... ppc64el and s390x are somewhere in between ... ) > > thanks for the background ogra. very useful information indeed. cheers, manik -------------- next part -------------- An HTML attachment was scrubbed... URL: From manik at canonical.com Fri Sep 16 06:06:45 2016 From: manik at canonical.com (Manik Taneja) Date: Thu, 15 Sep 2016 23:06:45 -0700 Subject: Ubuntu Core 16 beta images available In-Reply-To: References: Message-ID: On Thu, Sep 15, 2016 at 12:09 PM, Michael wrote: > Hello, > > The RPi3 image has been deleted. Was this intentional? I have the file, but if this was intentionally taken down, I probably don't want to flash it to my card. I did get the "unxz: ubuntu-core-16-pi3.img.xz: Unexpected end of input" error noted by others when I extracted it. > > > Yes, our apologies for the deletion. We discovered an issue in the previous Rpi3 image and are working on a fix. We will be posting the updated image shortly. Please stay tuned! Regards, Manik Taneja Product Manager, Ubuntu Core -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Fri Sep 16 08:55:23 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 16 Sep 2016 10:55:23 +0200 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> Message-ID: <1474016123.17795.131.camel@ubuntu.com> hi, Am Donnerstag, den 15.09.2016, 16:10 -0500 schrieb Dave Chiluk: > Hey guys, I'm trying to snap up file.  I thought it would be simple.  > This is what I have for my snapcraft.yaml. > ... > > However when I go to run file out of my snap I get > " > chiluk at localhost:~$ file /home/chiluk/file_0_amd64.snap > /etc/magic, 0: Warning: using regular magic file > `/usr/share/misc/magic' > file: could not find any valid magic files! you will need a wrapper ... make it call file -m and point to the $SNAP location of the magic file: (from the manpage) ... -m, --magic-file magicfiles              Specify an alternate list of files and directories containing magic.  This can be a single item, or a colon-separated list.  If a compiled              magic file is found alongside a file or directory, it will be used instead. ... ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From dave.chiluk at canonical.com Fri Sep 16 13:11:05 2016 From: dave.chiluk at canonical.com (Dave Chiluk) Date: Fri, 16 Sep 2016 08:11:05 -0500 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <1474016123.17795.131.camel@ubuntu.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> <1474016123.17795.131.camel@ubuntu.com> Message-ID: <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> Can't we do something more elegant than that? Having to implement wrappers and or code changes to each and every program with hardcoded values seems overly painful. Couldn't we do something fancy/smart with system calls such that fopen prepends $SNAP to most requests? I can't imagine there being more than a few system calls that would need to be modified. I also realize that it would take more work than this simplistic suggestion, but has anyone researched something like that? Plus it would give each application even better isolation since the isolation would be enforced by snappy. Dave. On 09/16/2016 03:55 AM, Oliver Grawert wrote: > hi, > Am Donnerstag, den 15.09.2016, 16:10 -0500 schrieb Dave Chiluk: >> Hey guys, I'm trying to snap up file. I thought it would be simple. >> This is what I have for my snapcraft.yaml. >> > ... >> >> However when I go to run file out of my snap I get >> " >> chiluk at localhost:~$ file /home/chiluk/file_0_amd64.snap >> /etc/magic, 0: Warning: using regular magic file >> `/usr/share/misc/magic' >> file: could not find any valid magic files! > > you will need a wrapper ... make it call file -m and point to the $SNAP > location of the magic file: > > (from the manpage) > ... > -m, --magic-file magicfiles > Specify an alternate list of files and directories > containing magic. This can be a single item, or a colon-separated > list. If a compiled > magic file is found alongside a file or directory, it will > be used instead. > ... > > ciao > oli > > > From mark at ubuntu.com Fri Sep 16 13:15:08 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 16 Sep 2016 06:15:08 -0700 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> <1474016123.17795.131.camel@ubuntu.com> <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> Message-ID: <49bf089f-1f1d-e1d8-f070-d850ad6e089b@ubuntu.com> On 16/09/16 06:11, Dave Chiluk wrote: > I also realize that it would take more work than this simplistic > suggestion, but has anyone researched something like that? Plus it > would give each application even better isolation since the isolation > would be enforced by snappy. Except that the NICE think about snaps is that they CAN be integrated elsewhere on the filesystem, and a sledgehammer such as the one you propose would make that more awkward by the looks of it. I do think we need a quilt or other mechanism for carrying patches. I nack'd that previously, but having dug in, I've come around. Mark From jamie at canonical.com Fri Sep 16 13:45:22 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Fri, 16 Sep 2016 08:45:22 -0500 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> <1474016123.17795.131.camel@ubuntu.com> <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> Message-ID: <1474033522.12872.26.camel@canonical.com> On Fri, 2016-09-16 at 08:11 -0500, Dave Chiluk wrote: > Can't we do something more elegant than that?  Having to implement > wrappers and or code changes to each and every program with hardcoded > values seems overly painful. > > Couldn't we do something fancy/smart with system calls such that fopen > prepends $SNAP to most requests?  I can't imagine there being more than > a few system calls that would need to be modified. > There is a bug related to this[1] and the idea Gustavo had is that snappy would provide a way for you to declare in your yaml a way to map /etc/magic to $SNAP/etc/magic. [1]https://bugs.launchpad.net/snapcraft/+bug/1577514/comments/6 -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From ogra at ubuntu.com Fri Sep 16 14:02:04 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 16 Sep 2016 16:02:04 +0200 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <49bf089f-1f1d-e1d8-f070-d850ad6e089b@ubuntu.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> <1474016123.17795.131.camel@ubuntu.com> <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> <49bf089f-1f1d-e1d8-f070-d850ad6e089b@ubuntu.com> Message-ID: <1474034524.15944.3.camel@ubuntu.com> hi, On Fr, 2016-09-16 at 06:15 -0700, Mark Shuttleworth wrote: >  > I do think we need a quilt or other mechanism for carrying patches. I > nack'd that previously, but having dug in, I've come around. >  while i agree that we need a patch system, didn't we say we don't want to actually replace debs with snaps ? looking at the kind of questions that start hitting this list in the last weeks it feels like people start doing exactly that ...  snapping something like the file command as a standalone snap seems to be actually moving in this direction ... ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From spencertparkin at gmail.com Fri Sep 16 14:05:03 2016 From: spencertparkin at gmail.com (Spencer) Date: Fri, 16 Sep 2016 08:05:03 -0600 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: <1474033522.12872.26.camel@canonical.com> References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> <1474016123.17795.131.camel@ubuntu.com> <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> <1474033522.12872.26.camel@canonical.com> Message-ID: I never like hard coded paths. Can YAML files somehow configure what C pre processor definitions we get during compilation? An if-def could be used in places where the program is to locate resources and so use either $SNAP prepended or something else when not run as a snap. > On Sep 16, 2016, at 7:45 AM, Jamie Strandboge wrote: > >> On Fri, 2016-09-16 at 08:11 -0500, Dave Chiluk wrote: >> Can't we do something more elegant than that? Having to implement >> wrappers and or code changes to each and every program with hardcoded >> values seems overly painful. >> >> Couldn't we do something fancy/smart with system calls such that fopen >> prepends $SNAP to most requests? I can't imagine there being more than >> a few system calls that would need to be modified. > > There is a bug related to this[1] and the idea Gustavo had is that snappy would > provide a way for you to declare in your yaml a way to map /etc/magic to > $SNAP/etc/magic. > > [1]https://bugs.launchpad.net/snapcraft/+bug/1577514/comments/6 > > -- > Jamie Strandboge | http://www.canonical.com > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From mark at ubuntu.com Sat Sep 17 00:58:24 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 16 Sep 2016 17:58:24 -0700 Subject: How do you add config files - "file" can't find it's configuration files? In-Reply-To: References: <9c586a08-a858-2e7d-1947-0dc2bccdcc34@canonical.com> <1474016123.17795.131.camel@ubuntu.com> <1e9ce7c6-be75-39ee-e3b3-838e21c3e3fc@canonical.com> <1474033522.12872.26.camel@canonical.com> Message-ID: On 16/09/16 07:05, Spencer wrote: > I never like hard coded paths. Can YAML files somehow configure what C pre processor definitions we get during compilation? An if-def could be used in places where the program is to locate resources and so use either $SNAP prepended or something else when not run as a snap. Yes, but this does mean patching the upstream code to be aware of those environment settings. Mark From spencertparkin at gmail.com Sat Sep 17 06:23:10 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Sat, 17 Sep 2016 00:23:10 -0600 Subject: non-free Message-ID: At the risk of e-mailing the list again too soon (low frequency of spamage is a good policy, especially if, like me, you have a tenancy to alienate yourself from people with excessive naivity and stupidity), I wanted to ask: how can I re-assure the Ubuntu Store that my program is indeed free? It says it's "non-free," because "[it] may contain non-free components." I'm using wxWidgets, OpenGL, and boost. I'm fairly certain those are free. I'm not a legal expert, though. And I can't think of anything else I might be using that would make it non-free. Thoughts? Thanks, --Sp -------------- next part -------------- An HTML attachment was scrubbed... URL: From ericoporto2008 at gmail.com Sat Sep 17 14:10:50 2016 From: ericoporto2008 at gmail.com (=?UTF-8?B?w4lyaWNvIFA=?=) Date: Sat, 17 Sep 2016 11:10:50 -0300 Subject: non-free In-Reply-To: References: Message-ID: Hello Spencer, I may be wrong, but right now all snaps are flagged as non-free. I don't know if there's a way to tell right now if the snap is a result of open or proprietary code - and binary blobs. The way I can think of doing this would be remotely building every snap in the same environment, away from the developer/packager, to ensure, based on the listed sources and licenses, that what this environment is grabbing to build the snap, is all indeed free open source code. There must be other forms, I don't know what's planned. Regards, Érico Vieira Porto Em 17 de set de 2016 3:24 AM, "Spencer Parkin" escreveu: > At the risk of e-mailing the list again too soon (low frequency of spamage > is a good policy, especially if, like me, you have a tenancy to alienate > yourself from people with excessive naivity and stupidity), I wanted to > ask: how can I re-assure the Ubuntu Store that my program is indeed free? > It says it's "non-free," because "[it] may contain non-free components." > > I'm using wxWidgets, OpenGL, and boost. I'm fairly certain those are > free. I'm not a legal expert, though. And I can't think of anything else > I might be using that would make it non-free. > > Thoughts? > > Thanks, > --Sp > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Sat Sep 17 14:44:56 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Sat, 17 Sep 2016 16:44:56 +0200 Subject: unofficial/experimental daily edge builds Message-ID: <1474123496.15944.9.camel@ubuntu.com> hi, since the implementation of official daily ubuntu-core image builds for the various channels will still take a bit but people are asking about them all the time, i threw together a little script that runs every day at 6:30 UTC and publishes img files at http://people.canonical.com/~ogra/snappy/all-snaps/daily/current/ the images use the edge channel and are completely untested (for official and tested beta or stable builds please use [1] instead) and build for amd64, pi2, pi3 and the dragonboard. have fun :) ciao oli [1] http://cdimage.ubuntu.com/ubuntu-snappy/xenial/current/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From spencertparkin at gmail.com Sat Sep 17 15:56:22 2016 From: spencertparkin at gmail.com (Spencer) Date: Sat, 17 Sep 2016 09:56:22 -0600 Subject: non-free In-Reply-To: References: Message-ID: I see. Well, I'm not sure what the answer should be, but I can tell you that it being "non-free" is a significant deterrent against anyone downloading and installing this or anyone's snap, or so I would think. > On Sep 17, 2016, at 8:10 AM, Érico P wrote: > > Hello Spencer, > > I may be wrong, but right now all snaps are flagged as non-free. I don't know if there's a way to tell right now if the snap is a result of open or proprietary code - and binary blobs. > > The way I can think of doing this would be remotely building every snap in the same environment, away from the developer/packager, to ensure, based on the listed sources and licenses, that what this environment is grabbing to build the snap, is all indeed free open source code. There must be other forms, I don't know what's planned. > > Regards, > > Érico Vieira Porto > > > Em 17 de set de 2016 3:24 AM, "Spencer Parkin" escreveu: >> At the risk of e-mailing the list again too soon (low frequency of spamage is a good policy, especially if, like me, you have a tenancy to alienate yourself from people with excessive naivity and stupidity), I wanted to ask: how can I re-assure the Ubuntu Store that my program is indeed free? It says it's "non-free," because "[it] may contain non-free components." >> >> I'm using wxWidgets, OpenGL, and boost. I'm fairly certain those are free. I'm not a legal expert, though. And I can't think of anything else I might be using that would make it non-free. >> >> Thoughts? >> >> Thanks, >> --Sp >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From eloy.garcia.pca at gmail.com Sun Sep 18 19:19:43 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Sun, 18 Sep 2016 21:19:43 +0200 Subject: No such file or directory when try to execute gsettings In-Reply-To: <7472fdcf-abdd-5faa-448f-17ba05a73482@ubuntu.com> References: <07556db2-d35e-3aa6-e703-0a98ece000aa@ubuntu.com> <7472fdcf-abdd-5faa-448f-17ba05a73482@ubuntu.com> Message-ID: Hi Sebastien, hi all :) Sorry for not answering before, but I don't have too much spare time and I spent the whole last week trying different things with my snap package. Sebastien, you pointed to me in the right direction, THANK YOU VERY MUCH! :) I have tryied snapcraft-desktop-helpers with gsettings interface and they did the trick! (with some problems that I think I'll described in another thread in order to not mess this one which i think is solved). So again, thank you very much for the help :) Best, Eloy 2016-09-08 15:38 GMT+02:00 Sebastien Bacher : > Le 07/09/2016 à 12:04, Eloy García (PC Actual) a écrit : > > > > > > I get your point and I think this is the real problem. I guess > > gsettings command line tool from the snap is storing the correct > > key/value inside the snap, but the "native" dconf database is not > > getting the new value, because the change wasn't made using "native" > > gsettings command line tool. > No, it's the other way around. Writting to dconf goes through a proxy > service and the client talks to the server using dbus, so with the > gsettings interface your snap can contact the desktop service and the > write is done in the real-session's dconf database. > > Reading is done by opening a file in the user directory though and that > doesn't work "out of the box" from inside the snaps because your code is > going to see the private-snap-dir and not the real system one. > If you use the desktop common part it has a hack for that though > https://github.com/ubuntu/snapcraft-desktop-helpers/ > blob/master/gtk/launcher-specific#L79 > > > I have tried to search dconf database within snap confinement, but I > > couldn't find it. Where should I search for it to see if the new value > > for the wallpaper was set there? > You should rather look on your desktop using e.g dconf-editor > > > Cheers, > Sebastien Bacher > > -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Sat Sep 17 20:58:29 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sat, 17 Sep 2016 13:58:29 -0700 Subject: non-free In-Reply-To: References: Message-ID: On 17/09/16 08:56, Spencer wrote: > I see. Well, I'm not sure what the answer should be, but I can tell > you that it being "non-free" is a significant deterrent against anyone > downloading and installing this or anyone's snap, or so I would think. Agreed - I think this is just a flag in the database, and I'm pretty sure there is a way to change that, just not sure where it is. But this is the right list to ask on :) Mark From otfried at ipe.airpost.net Sun Sep 18 21:03:08 2016 From: otfried at ipe.airpost.net (Otfried Cheong) Date: Sun, 18 Sep 2016 23:03:08 +0200 Subject: Writable directories in /var/snap? Message-ID: <1474232588.3140532.729481113.10F59714@webmail.messagingengine.com> Hello, I'm trying to create my first snap, and have a first question: When a snap is installed (say hello-world), the system creates two data directories, /var/snap/hello-world/ and /var/snap/hello-world/common. Apparently these are meant for the snap when it needs writable storage, but both directories are owned by root, and I haven't found any way to write to them, either from inside the snap or outside, whether installing in --devmode or not. Cheers, Otfried From ogra at ubuntu.com Sun Sep 18 21:15:56 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Sun, 18 Sep 2016 23:15:56 +0200 Subject: Writable directories in /var/snap? In-Reply-To: <1474232588.3140532.729481113.10F59714@webmail.messagingengine.com> References: <1474232588.3140532.729481113.10F59714@webmail.messagingengine.com> Message-ID: <1474233356.15944.70.camel@ubuntu.com> hi, On So, 2016-09-18 at 23:03 +0200, Otfried Cheong wrote: > Hello, > > I'm trying to create my first snap, and have a first question: > > When a snap is installed (say hello-world), the system creates two > data > directories, /var/snap/hello-world/ and > /var/snap/hello-world/common.    > > Apparently these are meant for the snap when it needs writable > storage, > but both directories are owned by root, and I haven't found any way > to > write to them, either from inside the snap or outside, whether > installing in --devmode or not. >  they are for services (which always run as root by default) ... user writable dirs are in ~/snap/// in the running snap environment this is represented by the SNAP_USER_DATA environment variable. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From david.calle at canonical.com Sun Sep 18 21:18:09 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Sun, 18 Sep 2016 23:18:09 +0200 Subject: Writable directories in /var/snap? In-Reply-To: <1474233356.15944.70.camel@ubuntu.com> References: <1474232588.3140532.729481113.10F59714@webmail.messagingengine.com> <1474233356.15944.70.camel@ubuntu.com> Message-ID: <7b6e76be-c7eb-cbd6-cab3-686dd91a7562@canonical.com> On 18/09/2016 23:15, Oliver Grawert wrote: > hi, > On So, 2016-09-18 at 23:03 +0200, Otfried Cheong wrote: >> Hello, >> >> I'm trying to create my first snap, and have a first question: >> >> When a snap is installed (say hello-world), the system creates two >> data >> directories, /var/snap/hello-world/ and >> /var/snap/hello-world/common. >> >> Apparently these are meant for the snap when it needs writable >> storage, >> but both directories are owned by root, and I haven't found any way >> to >> write to them, either from inside the snap or outside, whether >> installing in --devmode or not. >> > they are for services (which always run as root by default) ... > > user writable dirs are in ~/snap/// > in the running snap environment this is represented by the > SNAP_USER_DATA environment variable. > > ciao > oli > > You also have $SNAP_USER_COMMON: ~/snap//common/ , that persists between upgrades of the snap. David -------------- next part -------------- An HTML attachment was scrubbed... URL: From otfried at ipe.airpost.net Sun Sep 18 21:55:57 2016 From: otfried at ipe.airpost.net (Otfried Cheong) Date: Sun, 18 Sep 2016 23:55:57 +0200 Subject: Writable directories in /var/snap? In-Reply-To: <7b6e76be-c7eb-cbd6-cab3-686dd91a7562@canonical.com> References: <1474232588.3140532.729481113.10F59714@webmail.messagingengine.com> <1474233356.15944.70.camel@ubuntu.com> <7b6e76be-c7eb-cbd6-cab3-686dd91a7562@canonical.com> Message-ID: <1474235757.3149343.729519393.4A156073@webmail.messagingengine.com> On Sun, Sep 18, 2016, at 23:18, David Callé wrote: You also have $SNAP_USER_COMMON: ~/snap//common/ , that persists between upgrades of the snap. $SNAP_USER_COMMON is not created automatically when the snap is installed, and trying to create it from the snap fails (permission denied). If I create it from the host I can then access it from the snap, it seems, but not being able to create the directory makes it somewhat hard to use... Cheers, Otfried From sergio.schvezov at canonical.com Sun Sep 18 23:05:25 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Sun, 18 Sep 2016 20:05:25 -0300 Subject: non-free In-Reply-To: References: Message-ID: El sábado, 17 de septiembre de 2016 17h'58:29 ART, Mark Shuttleworth escribió: > On 17/09/16 08:56, Spencer wrote: >> I see. Well, I'm not sure what the answer should be, but I can tell >> you that it being "non-free" is a significant deterrent against anyone >> downloading and installing this or anyone's snap, or so I would think. > > Agreed - I think this is just a flag in the database, and I'm pretty > sure there is a way to change that, just not sure where it is. But this > is the right list to ask on :) The license feature needs to be brought back into snapd. If there is no license The Software Center defaults to proprietary. I know seb128 or rancell were on top of this at one point and logged a bug against snapd. -- Enviado con Dekko desde mi dispositivo Ubuntu From robert.ancell at canonical.com Sun Sep 18 23:36:48 2016 From: robert.ancell at canonical.com (Robert Ancell) Date: Sun, 18 Sep 2016 23:36:48 +0000 Subject: non-free In-Reply-To: References: Message-ID: Correct, snapd doesn't provide any license information currently and GNOME Software defaults to interpreting this as propietary. The bug tracking this is: https://bugs.launchpad.net/bugs/1555567 On Mon, Sep 19, 2016 at 11:06 AM Sergio Schvezov < sergio.schvezov at canonical.com> wrote: > El sábado, 17 de septiembre de 2016 17h'58:29 ART, Mark Shuttleworth > escribió: > > On 17/09/16 08:56, Spencer wrote: > >> I see. Well, I'm not sure what the answer should be, but I can tell > >> you that it being "non-free" is a significant deterrent against anyone > >> downloading and installing this or anyone's snap, or so I would think. > > > > Agreed - I think this is just a flag in the database, and I'm pretty > > sure there is a way to change that, just not sure where it is. But this > > is the right list to ask on :) > > The license feature needs to be brought back into snapd. If there is no > license The Software Center defaults to proprietary. > > I know seb128 or rancell were on top of this at one point and logged a bug > against snapd. > > > > -- > Enviado con Dekko desde mi dispositivo Ubuntu > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kyle.fazzari at canonical.com Mon Sep 19 01:22:19 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Sun, 18 Sep 2016 18:22:19 -0700 Subject: Writable directories in /var/snap? In-Reply-To: <1474235757.3149343.729519393.4A156073@webmail.messagingengine.com> References: <1474232588.3140532.729481113.10F59714@webmail.messagingengine.com> <1474233356.15944.70.camel@ubuntu.com> <7b6e76be-c7eb-cbd6-cab3-686dd91a7562@canonical.com> <1474235757.3149343.729519393.4A156073@webmail.messagingengine.com> Message-ID: <52a9349b-60c3-192f-6340-cf6f9d64539c@canonical.com> On 09/18/2016 02:55 PM, Otfried Cheong wrote: > On Sun, Sep 18, 2016, at 23:18, David Callé wrote: > > You also have $SNAP_USER_COMMON: ~/snap//common/ , that > persists between upgrades of the snap. > > > $SNAP_USER_COMMON is not created automatically when the snap is > installed, and trying to create it from the snap fails (permission > denied). > > If I create it from the host I can then access it from the snap, it > seems, but not being able to create the directory makes it somewhat hard > to use... Indeed, a known bug[1] that should be fixed soon. [1]: https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1611063 -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From otfried at ipe.airpost.net Mon Sep 19 08:37:38 2016 From: otfried at ipe.airpost.net (Otfried Cheong) Date: Mon, 19 Sep 2016 10:37:38 +0200 Subject: Using xdg-open from snap Message-ID: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Hello, my app has a manual in html. I normally show this using "xdg-open ", but from the snap this results in "xdg-open: Permission denied", leaving this log: [21249.231634] audit: type=1400 audit(1474273861.873:383): apparmor="DENIED" operation="exec" profile="snap.ipe.sh" name="/usr/local/bin/xdg-open" pid=9551 comm="sh" requested_mask="x" denied_mask="x" fsuid=1000 ouid=0 According to https://lists.ubuntu.com/archives/snapcraft/2016-September/001048.html this should work. I did refresh ubuntu-core from the beta channel and currently have revision 636 of ubuntu-core. Slightly related: If I understand https://lists.ubuntu.com/archives/snapcraft/2016-September/001118.html correctly, the host filesystem should be exposed to the snap as /var/lib/snapd/hostfs in devmode? It isn't on my system. Cheers, Otfried From ogra at ubuntu.com Mon Sep 19 13:10:57 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Mon, 19 Sep 2016 15:10:57 +0200 Subject: unofficial/experimental daily edge builds In-Reply-To: <1474123496.15944.9.camel@ubuntu.com> References: <1474123496.15944.9.camel@ubuntu.com> Message-ID: <1474290657.7138.5.camel@ubuntu.com> hi, On Sa, 2016-09-17 at 16:44 +0200, Oliver Grawert wrote: ... > http://people.canonical.com/~ogra/snappy/all-snaps/daily/current/ ... > the images use the edge channel and are completely untested (for > official and tested beta or stable builds please use [1] instead) and > build for amd64, pi2, pi3 and the dragonboard. > i worked on creating a beaglebone black image over the weekend ... while this is still very buggy you can find it at the above URL to tinker with it. note that there is a known OOPS of the network card driver with the current kernel [1] ( you can use a USB NIC to work around this) and there are issues with the firstboot service, so "snap list" wont find any snaps ... you can work around the latter by logging in and issuing: sudo rm /var/lib/snapd/state.json sudo systemctl restart snapd.firstboot.service  # ^^ (this can take very long (i.e. around 10min)) sudo reboot on the new boot "snap list" should then work. caio oli [1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1625177 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From sergio.schvezov at canonical.com Mon Sep 19 13:17:25 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Mon, 19 Sep 2016 10:17:25 -0300 Subject: Setting release version in snapcraft build In-Reply-To: References: Message-ID: Sorry for the late reply El 08/09/16 a las 11:28, Olivier Tilloy escribió: > That would work for me too. Is there a bug report I can subscribe to > to follow progress/implementation? LP: #1594794 From eloy.garcia.pca at gmail.com Mon Sep 19 13:34:29 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Mon, 19 Sep 2016 15:34:29 +0200 Subject: Problem with snapcraft-desktop-helpers and MATE desktop Message-ID: Hi all. I'm using snapcraft-desktop-helpers ( https://github.com/ubuntu/snapcraft-desktop-helpers) part for providing a "bridge" for gsettings between my snapped application and the host system (I'm using desktop/gtk3 part although I tried desktop/gtk2 part and even desktop/glib-only). The application sets a custom selected wallpaper and it is working great under GNOME 3 and Unity. However, it is not working for MATE desktop. This is the command I execute inside the snap for Unity and GNOME 3 desktops: gsettings set org.gnome.desktop.background picture-uri file://blablabla.jpg And this is the command I execute for MATE desktop: gsettings set org.mate.background picture-filename blablabla.jpg Checking the application's log I find: [15:17:40] AWT-EventQueue-0 ERROR LinuxWallpaperChanger:102 - No such schema 'org.mate.background' Those commands are working fine using the native application (I mean, the application directly running on the host and not installed via snap package) Do I need to do something else to support MATE desktop or maybe it's a bug for snapcraft-desktop-helpers wiki part? Thank you very much for your help. Best, Eloy -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Mon Sep 19 14:20:51 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Mon, 19 Sep 2016 16:20:51 +0200 Subject: Problem with snapcraft-desktop-helpers and MATE desktop In-Reply-To: References: Message-ID: <86cbe5a8-1bb2-1715-15f8-6b57d436fa4e@ubuntu.com> Le 19/09/2016 à 15:34, Eloy García (PC Actual) a écrit : > Hi all. > > I'm using snapcraft-desktop-helpers > (https://github.com/ubuntu/snapcraft-desktop-helpers) part for > providing a "bridge" for gsettings between my snapped application and > the host system (I'm using desktop/gtk3 part although I tried > desktop/gtk2 part and even desktop/glib-only). The application sets a > custom selected wallpaper and it is working great under GNOME 3 and > Unity. However, it is not working for MATE desktop. > > This is the command I execute inside the snap for Unity and GNOME 3 > desktops: > gsettings set org.gnome.desktop.background picture-uri > file://blablabla.jpg > > And this is the command I execute for MATE desktop: > gsettings set org.mate.background picture-filename blablabla.jpg > > Checking the application's log I find: > > [15:17:40] AWT-EventQueue-0 ERROR LinuxWallpaperChanger:102 - No such > schema 'org.mate.background' > > Those commands are working fine using the native application (I mean, > the application directly running on the host and not installed via > snap package) > > Do I need to do something else to support MATE desktop or maybe it's a > bug for snapcraft-desktop-helpers wiki part? > > Thank you very much for your help. Hey Eloy! Actually, it's not a bug, but the way gsettings is working. You need to have the schema files you are writing to and/or reading from available to your application for gsettings validating its input/output. By default, you application isn't providing those I guess. You can either ship directly the xml files as part of your snap or stage the ubuntu package containing the schema. This is the reason why you can see this playpen example shipping gsettings-desktop-schemas: https://github.com/ubuntu/snappy-playpen/blob/master/ubuntukylin-icon-theme/snapcraft.yaml#L35 In your ubuntu-mate case, after a quick search, it seems that you want to stage mate-desktop-common. Then, the desktop wrapper will do the right thing compiling them the first time you install your snap or upgrade it. I hope this helps! Cheers, Didier From michael.vogt at canonical.com Mon Sep 19 15:21:28 2016 From: michael.vogt at canonical.com (Michael Vogt) Date: Mon, 19 Sep 2016 17:21:28 +0200 Subject: Updated Ubuntu Core 16 beta images available Message-ID: <20160919152128.GB27970@bod> Ubuntu Core 16 Images ===================== The Ubuntu snappy team is happy to announce new beta images for Ubuntu Core 16. The images use the snapd package manager to install and update all components of the system including kernel, core, gadget and applications. Some hightlights for this release: - a preview of snapweb is installed by default, snapweb is a web-ui to control an all-snap system - pi3 image available - smaller image size to fit better on sd-cards - bugfixes The images are available for PC (amd64, i386) and Raspberry Pi2/3 (armhf) and the Dragonboard 410. You can download them at: http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ The images are bootable, the pc image can be booted directly in qemu-kvm or virtualenv. When running the images in qemu-kvm it is helpful to use the "-redir" feature of qemu-kvm. E.g.: $ kvm -m 1500 -redir tcp:10022::22 -redir tcp:14200::14200 ubuntu-core-16-amd64.img The message from console-conf is a bit misleading in this setup. It will say "ssh USER at 10.0.2.15". However due to the way that qemu-kvm user networking works, you will actually have to run: $ ssh -p 10022 USER at localhost to ssh into the images. Snapweb will be available at http://localhost:14200 in this setup. The pi2/pi3/dragonboard image can be written to a sdcard via dd. An alternative way to write the image is to use "go-dd": $ sudo snap install --devmode --beta godd $ sudo /snap/bin/godd ubuntu-core-16-pi2.img [this will print a message showing what devices are removable] $ sudo /snap/bin/godd ubuntu-core-16-pi2.img /dev/sdXX The advantage of godd is that it will not write to devices that are still mounted and that it can help in detecting what removable devices are available (and it shows a progressbar and syncs at the end). After booting the image you can enter your Ubuntu SSO email and it will automatically create a matching user with the right ssh keys. If you do not have an Ubuntu SSO account yet you can create one at: https://login.ubuntu.com/ These images follow the "beta" channel. Enjoy the fresh images! If you find any bugs or issues, please let us know via: https://bugs.launchpad.net/snappy/ Known issues: - On the pi3 the wireless interface does not work for the initial setup, you need to use the wired interface http://pad.lv/1624322 and http://pad.lv/1623120 - snapweb on i386 works only after the first `snap refresh` (will happen automatically) Cheers, Michael (on behalf of the snappy team) From leo.arias at canonical.com Mon Sep 19 16:59:18 2016 From: leo.arias at canonical.com (Leo Arias) Date: Mon, 19 Sep 2016 10:59:18 -0600 Subject: Snap layer for Juju reactive charms In-Reply-To: References: Message-ID: Thanks Stuart! This is huge. Snaps are a great way to deliver the software that composes a charm. And charms are a great way to put those snaps to good use. With both teams working together we can make a consistent and smooth story to grow the adoption. I'm eager to start playing with your layer. Now we need to experiment so we can identify the main missing points on snapd that affect charms, and start defining good practices for charming. Some suggestions: * We have a vault charm [1] and a vault snap [2]. Both seem simple and it should be straightforward to combine them; but at the same time vault is awesome, it's becoming popular and it will benefit a lot from the security and isolation of snaps. Here's the chance to get some good noise and get more people interested. * We have a nextcloud snap [3] which is great for a home installation because it bundles all the requirements, it's just a one click install. But it doesn't have high availability, it hardcodes mysql and apache, and you can't easily scale or deploy to various clouds. Upstream was really nice and helpful, so maybe we can get them engaged to support bigger deployments. If anybody wants to help, please reply. pura vida [1] https://jujucharms.com/u/jamesbeedy/vault/13 [2] https://github.com/elopio/vault/tree/snapcraft [3] https://github.com/nextcloud/nextcloud-snap -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Mon Sep 19 18:38:13 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Mon, 19 Sep 2016 15:38:13 -0300 Subject: New GH review capabilities Message-ID: Hello all, GitHub has finally implemented the last few pieces missing to have sane reviews directly on PRs, which is great for us. Most importantly, we can now draft out a full review, including the final comments, before we ship the comments all at once, including a verdict. The verdict may be one of "Comment", "Approve", and "Request changes". An important note about "Request changes": we have configured the master branch so that this one will BLOCK the branch from being merged before YOU get back to it, so please use it mindfully. If you have some notes on the branch that SHOULD be addressed, but that you think somebody else could do a good job ensuring they got done, please use the "Comment" veredict instead of requesting changes formally. This way you don't "Approve" the change to go in as-is, but also don't block someone else from analyzing and potentially merging the branch when it's ready. In summary, "Request changes" means "It's not ready and I want to be in the loop on this one.". Let's be careful to use it as such, and observe how things go in the next few days. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From eloy.garcia.pca at gmail.com Mon Sep 19 21:26:07 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Mon, 19 Sep 2016 23:26:07 +0200 Subject: Problem with snapcraft-desktop-helpers and MATE desktop In-Reply-To: <86cbe5a8-1bb2-1715-15f8-6b57d436fa4e@ubuntu.com> References: <86cbe5a8-1bb2-1715-15f8-6b57d436fa4e@ubuntu.com> Message-ID: Hi Didier! Thank you very much for your help. It worked like a charm and now my snap package containing my application works flawlesly even in MATE desktops :) Again, THANK YOU! :D Best, Eloy 2016-09-19 16:20 GMT+02:00 Didier Roche : > Le 19/09/2016 à 15:34, Eloy García (PC Actual) a écrit : > > Hi all. > > > > I'm using snapcraft-desktop-helpers > > (https://github.com/ubuntu/snapcraft-desktop-helpers) part for > > providing a "bridge" for gsettings between my snapped application and > > the host system (I'm using desktop/gtk3 part although I tried > > desktop/gtk2 part and even desktop/glib-only). The application sets a > > custom selected wallpaper and it is working great under GNOME 3 and > > Unity. However, it is not working for MATE desktop. > > > > This is the command I execute inside the snap for Unity and GNOME 3 > > desktops: > > gsettings set org.gnome.desktop.background picture-uri > > file://blablabla.jpg > > > > And this is the command I execute for MATE desktop: > > gsettings set org.mate.background picture-filename blablabla.jpg > > > > Checking the application's log I find: > > > > [15:17:40] AWT-EventQueue-0 ERROR LinuxWallpaperChanger:102 - No such > > schema 'org.mate.background' > > > > Those commands are working fine using the native application (I mean, > > the application directly running on the host and not installed via > > snap package) > > > > Do I need to do something else to support MATE desktop or maybe it's a > > bug for snapcraft-desktop-helpers wiki part? > > > > Thank you very much for your help. > > Hey Eloy! > > Actually, it's not a bug, but the way gsettings is working. You need to > have the schema files you are writing to and/or reading from available > to your application for gsettings validating its input/output. By > default, you application isn't providing those I guess. > > You can either ship directly the xml files as part of your snap or stage > the ubuntu package containing the schema. This is the reason why you can > see this playpen example shipping gsettings-desktop-schemas: > https://github.com/ubuntu/snappy-playpen/blob/master/ > ubuntukylin-icon-theme/snapcraft.yaml#L35 > > In your ubuntu-mate case, after a quick search, it seems that you want > to stage mate-desktop-common. > > Then, the desktop wrapper will do the right thing compiling them the > first time you install your snap or upgrade it. > > I hope this helps! > Cheers, > Didier > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From stuart.bishop at canonical.com Tue Sep 20 07:52:00 2016 From: stuart.bishop at canonical.com (Stuart Bishop) Date: Tue, 20 Sep 2016 14:52:00 +0700 Subject: CLI UX for plugs with parameters? Message-ID: Hi. I believe that slots will have support for arguments, for example allowing a snap to connect to the 'home' slot and provide an argument requesting access to hidden files. Does anyone know what the command line interface will look like? In the snap layer for Juju, I need to define the data structure to declare what additional connections need to be made after installation of the snap. I'd like to get this right the first time since I'm going to be stuck with it. -- Stuart Bishop -------------- next part -------------- An HTML attachment was scrubbed... URL: From spencertparkin at gmail.com Tue Sep 20 13:46:11 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Tue, 20 Sep 2016 07:46:11 -0600 Subject: Using xdg-open from snap In-Reply-To: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> References: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Message-ID: This is related to a question I had as well. I have a program that uses wxLaunchDefaultBrowser which, looking at its implementation, tries to make the system call "exec()" to launch the default browser with a URL. If snap programs are not allowed to start other processes, that's fine; but if enough people need to launch the default browser with a URL, then I'm sure a secure solution just for this could somehow be implemented for snaps. I gather that one design goal of snaps, however, is the ability for people to write programs for any environment, but also have them work as snaps so that the programmer doesn't have to write snap-specific code, or make snap-specific considerations in their code. In other words, your code should be "none-the-wiser" that it is running in the confined area. So with that in mind, I'm not sure how to solve the problem. Any secure API exposed to snap applications already breaks the above design goal. Of course, it's not unreasonable for my program to have "#ifdef WIN32" or "#ifdef UNIX", and in the latter case, I may be looking to utilize something in a standard unix environment which, I believe, is synthesized in Unbuntu Core. That's where I believe the snap environment can intercept what an application is doing and provide a secure solution, and this may be the "xdg-open" thing Otfried was talking about. On Mon, Sep 19, 2016 at 2:37 AM, Otfried Cheong wrote: > Hello, > > my app has a manual in html. I normally show this using "xdg-open > ", but from the snap this results in "xdg-open: Permission denied", > leaving this log: > > [21249.231634] audit: type=1400 audit(1474273861.873:383): > apparmor="DENIED" operation="exec" profile="snap.ipe.sh" > name="/usr/local/bin/xdg-open" pid=9551 comm="sh" requested_mask="x" > denied_mask="x" fsuid=1000 ouid=0 > > According to > https://lists.ubuntu.com/archives/snapcraft/2016-September/001048.html > this should work. > I did refresh ubuntu-core from the beta channel and currently have > revision 636 of ubuntu-core. > > > Slightly related: If I understand > https://lists.ubuntu.com/archives/snapcraft/2016-September/001118.html > correctly, the host filesystem should be exposed to the snap as > /var/lib/snapd/hostfs in devmode? It isn't on my system. > > Cheers, > Otfried > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From eloy.garcia.pca at gmail.com Wed Sep 21 07:18:21 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Wed, 21 Sep 2016 09:18:21 +0200 Subject: Using xdg-open from snap In-Reply-To: References: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Message-ID: Hi all. I have the same problem in my snap java-based application. I use xdg-open command to launch the default browser so, it would be great a solution :) Best, Eloy 2016-09-20 15:46 GMT+02:00 Spencer Parkin : > This is related to a question I had as well. I have a program that uses > wxLaunchDefaultBrowser which, looking at its implementation, tries to make > the system call "exec()" to launch the default browser with a URL. > > If snap programs are not allowed to start other processes, that's fine; > but if enough people need to launch the default browser with a URL, then > I'm sure a secure solution just for this could somehow be implemented for > snaps. > > I gather that one design goal of snaps, however, is the ability for people > to write programs for any environment, but also have them work as snaps so > that the programmer doesn't have to write snap-specific code, or make > snap-specific considerations in their code. In other words, your code > should be "none-the-wiser" that it is running in the confined area. > > So with that in mind, I'm not sure how to solve the problem. Any secure > API exposed to snap applications already breaks the above design goal. > > Of course, it's not unreasonable for my program to have "#ifdef WIN32" or > "#ifdef UNIX", and in the latter case, I may be looking to utilize > something in a standard unix environment which, I believe, is synthesized > in Unbuntu Core. That's where I believe the snap environment can intercept > what an application is doing and provide a secure solution, and this may be > the "xdg-open" thing Otfried was talking about. > > > On Mon, Sep 19, 2016 at 2:37 AM, Otfried Cheong > wrote: > >> Hello, >> >> my app has a manual in html. I normally show this using "xdg-open >> ", but from the snap this results in "xdg-open: Permission denied", >> leaving this log: >> >> [21249.231634] audit: type=1400 audit(1474273861.873:383): >> apparmor="DENIED" operation="exec" profile="snap.ipe.sh" >> name="/usr/local/bin/xdg-open" pid=9551 comm="sh" requested_mask="x" >> denied_mask="x" fsuid=1000 ouid=0 >> >> According to >> https://lists.ubuntu.com/archives/snapcraft/2016-September/001048.html >> this should work. >> I did refresh ubuntu-core from the beta channel and currently have >> revision 636 of ubuntu-core. >> >> >> Slightly related: If I understand >> https://lists.ubuntu.com/archives/snapcraft/2016-September/001118.html >> correctly, the host filesystem should be exposed to the snap as >> /var/lib/snapd/hostfs in devmode? It isn't on my system. >> >> Cheers, >> Otfried >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm >> an/listinfo/snapcraft >> > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.lenton at canonical.com Wed Sep 21 08:39:58 2016 From: john.lenton at canonical.com (John Lenton) Date: Wed, 21 Sep 2016 09:39:58 +0100 Subject: Using xdg-open from snap In-Reply-To: References: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Message-ID: Eloy, Spencer, Otfried, The xdg-open we ship in /usr/local in the snap-core snap failing like that is a bug; it seems we weren't covering this use case in our tests. jdstrand has now addressed this, and although with his fix right now you'll need to ask for the unity7 interface it is expected to grow into a more fine-grained interface at some point, it was put there to unblock people (i.e. you). We expect this fix to be part of the 2.15 release, but it might slip to 2.16. This is not the whole story, however. You'll also need the snapd-xdg-open package (or a dbus service providing OpenURL on the com.canonical.SafeLauncher interfacee) in your classic system. You can install that in yakkety, or get it from -proposed for xenial (https://launchpad.net/ubuntu/+source/snapd-xdg-open), or get the source from https://github.com/snapcore/snapd-xdg-open. As soon as it gets out of -proposed and into -updates we'll have snapd recommend it, but this might not be ready for 2.15. On 21 September 2016 at 08:18, Eloy García (PC Actual) wrote: > Hi all. > > I have the same problem in my snap java-based application. I use xdg-open > command to launch the default browser so, it would be great a solution :) > > Best, > > Eloy > > 2016-09-20 15:46 GMT+02:00 Spencer Parkin : >> >> This is related to a question I had as well. I have a program that uses >> wxLaunchDefaultBrowser which, looking at its implementation, tries to make >> the system call "exec()" to launch the default browser with a URL. >> >> If snap programs are not allowed to start other processes, that's fine; >> but if enough people need to launch the default browser with a URL, then I'm >> sure a secure solution just for this could somehow be implemented for snaps. >> >> I gather that one design goal of snaps, however, is the ability for people >> to write programs for any environment, but also have them work as snaps so >> that the programmer doesn't have to write snap-specific code, or make >> snap-specific considerations in their code. In other words, your code >> should be "none-the-wiser" that it is running in the confined area. >> >> So with that in mind, I'm not sure how to solve the problem. Any secure >> API exposed to snap applications already breaks the above design goal. >> >> Of course, it's not unreasonable for my program to have "#ifdef WIN32" or >> "#ifdef UNIX", and in the latter case, I may be looking to utilize something >> in a standard unix environment which, I believe, is synthesized in Unbuntu >> Core. That's where I believe the snap environment can intercept what an >> application is doing and provide a secure solution, and this may be the >> "xdg-open" thing Otfried was talking about. >> >> >> On Mon, Sep 19, 2016 at 2:37 AM, Otfried Cheong >> wrote: >>> >>> Hello, >>> >>> my app has a manual in html. I normally show this using "xdg-open >>> ", but from the snap this results in "xdg-open: Permission denied", >>> leaving this log: >>> >>> [21249.231634] audit: type=1400 audit(1474273861.873:383): >>> apparmor="DENIED" operation="exec" profile="snap.ipe.sh" >>> name="/usr/local/bin/xdg-open" pid=9551 comm="sh" requested_mask="x" >>> denied_mask="x" fsuid=1000 ouid=0 >>> >>> According to >>> https://lists.ubuntu.com/archives/snapcraft/2016-September/001048.html >>> this should work. >>> I did refresh ubuntu-core from the beta channel and currently have >>> revision 636 of ubuntu-core. >>> >>> >>> Slightly related: If I understand >>> https://lists.ubuntu.com/archives/snapcraft/2016-September/001118.html >>> correctly, the host filesystem should be exposed to the snap as >>> /var/lib/snapd/hostfs in devmode? It isn't on my system. >>> >>> Cheers, >>> Otfried >>> >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.snapcraft.io >>> Modify settings or unsubscribe at: >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.snapcraft.io >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > > > > -- > Eloy García Almadén > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > From eloy.garcia.pca at gmail.com Wed Sep 21 10:32:34 2016 From: eloy.garcia.pca at gmail.com (=?UTF-8?B?RWxveSBHYXJjw61hIChQQyBBY3R1YWwp?=) Date: Wed, 21 Sep 2016 12:32:34 +0200 Subject: Using xdg-open from snap In-Reply-To: References: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Message-ID: Hi John, Thank you very much for your response and for the good work all of you are doing :) I'll try to test it when it is possible. Best, Eloy 2016-09-21 10:39 GMT+02:00 John Lenton : > Eloy, Spencer, Otfried, > > The xdg-open we ship in /usr/local in the snap-core snap failing like > that is a bug; it seems we weren't covering this use case in our > tests. > > jdstrand has now addressed this, and although with his fix right now > you'll need to ask for the unity7 interface it is expected to grow > into a more fine-grained interface at some point, it was put there to > unblock people (i.e. you). We expect this fix to be part of the 2.15 > release, but it might slip to 2.16. > > This is not the whole story, however. You'll also need the > snapd-xdg-open package (or a dbus service providing OpenURL on the > com.canonical.SafeLauncher interfacee) in your classic system. You can > install that in yakkety, or get it from -proposed for xenial > (https://launchpad.net/ubuntu/+source/snapd-xdg-open), or get the > source from https://github.com/snapcore/snapd-xdg-open. As soon as it > gets out of -proposed and into -updates we'll have snapd recommend it, > but this might not be ready for 2.15. > > On 21 September 2016 at 08:18, Eloy García (PC Actual) > wrote: > > Hi all. > > > > I have the same problem in my snap java-based application. I use xdg-open > > command to launch the default browser so, it would be great a solution :) > > > > Best, > > > > Eloy > > > > 2016-09-20 15:46 GMT+02:00 Spencer Parkin : > >> > >> This is related to a question I had as well. I have a program that uses > >> wxLaunchDefaultBrowser which, looking at its implementation, tries to > make > >> the system call "exec()" to launch the default browser with a URL. > >> > >> If snap programs are not allowed to start other processes, that's fine; > >> but if enough people need to launch the default browser with a URL, > then I'm > >> sure a secure solution just for this could somehow be implemented for > snaps. > >> > >> I gather that one design goal of snaps, however, is the ability for > people > >> to write programs for any environment, but also have them work as snaps > so > >> that the programmer doesn't have to write snap-specific code, or make > >> snap-specific considerations in their code. In other words, your code > >> should be "none-the-wiser" that it is running in the confined area. > >> > >> So with that in mind, I'm not sure how to solve the problem. Any secure > >> API exposed to snap applications already breaks the above design goal. > >> > >> Of course, it's not unreasonable for my program to have "#ifdef WIN32" > or > >> "#ifdef UNIX", and in the latter case, I may be looking to utilize > something > >> in a standard unix environment which, I believe, is synthesized in > Unbuntu > >> Core. That's where I believe the snap environment can intercept what an > >> application is doing and provide a secure solution, and this may be the > >> "xdg-open" thing Otfried was talking about. > >> > >> > >> On Mon, Sep 19, 2016 at 2:37 AM, Otfried Cheong < > otfried at ipe.airpost.net> > >> wrote: > >>> > >>> Hello, > >>> > >>> my app has a manual in html. I normally show this using "xdg-open > >>> ", but from the snap this results in "xdg-open: Permission > denied", > >>> leaving this log: > >>> > >>> [21249.231634] audit: type=1400 audit(1474273861.873:383): > >>> apparmor="DENIED" operation="exec" profile="snap.ipe.sh" > >>> name="/usr/local/bin/xdg-open" pid=9551 comm="sh" requested_mask="x" > >>> denied_mask="x" fsuid=1000 ouid=0 > >>> > >>> According to > >>> https://lists.ubuntu.com/archives/snapcraft/2016-September/001048.html > >>> this should work. > >>> I did refresh ubuntu-core from the beta channel and currently have > >>> revision 636 of ubuntu-core. > >>> > >>> > >>> Slightly related: If I understand > >>> https://lists.ubuntu.com/archives/snapcraft/2016-September/001118.html > >>> correctly, the host filesystem should be exposed to the snap as > >>> /var/lib/snapd/hostfs in devmode? It isn't on my system. > >>> > >>> Cheers, > >>> Otfried > >>> > >>> > >>> -- > >>> Snapcraft mailing list > >>> Snapcraft at lists.snapcraft.io > >>> Modify settings or unsubscribe at: > >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft > >> > >> > >> > >> -- > >> Snapcraft mailing list > >> Snapcraft at lists.snapcraft.io > >> Modify settings or unsubscribe at: > >> https://lists.ubuntu.com/mailman/listinfo/snapcraft > >> > > > > > > > > -- > > Eloy García Almadén > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > -- Eloy García Almadén -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Wed Sep 21 11:18:41 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 21 Sep 2016 08:18:41 -0300 Subject: Using xdg-open from snap In-Reply-To: References: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Message-ID: That was a good explanation indeed, thanks John. Can we do something better than just recommend it on classic? The feature is common enough that this should be a requirement, I think. The problem then is how to drop the package when building the Ubuntu Core image. On Wed, Sep 21, 2016 at 5:39 AM, John Lenton wrote: > Eloy, Spencer, Otfried, > > The xdg-open we ship in /usr/local in the snap-core snap failing like > that is a bug; it seems we weren't covering this use case in our > tests. > > jdstrand has now addressed this, and although with his fix right now > you'll need to ask for the unity7 interface it is expected to grow > into a more fine-grained interface at some point, it was put there to > unblock people (i.e. you). We expect this fix to be part of the 2.15 > release, but it might slip to 2.16. > > This is not the whole story, however. You'll also need the > snapd-xdg-open package (or a dbus service providing OpenURL on the > com.canonical.SafeLauncher interfacee) in your classic system. You can > install that in yakkety, or get it from -proposed for xenial > (https://launchpad.net/ubuntu/+source/snapd-xdg-open), or get the > source from https://github.com/snapcore/snapd-xdg-open. As soon as it > gets out of -proposed and into -updates we'll have snapd recommend it, > but this might not be ready for 2.15. > > On 21 September 2016 at 08:18, Eloy García (PC Actual) > wrote: > > Hi all. > > > > I have the same problem in my snap java-based application. I use xdg-open > > command to launch the default browser so, it would be great a solution :) > > > > Best, > > > > Eloy > > > > 2016-09-20 15:46 GMT+02:00 Spencer Parkin : > >> > >> This is related to a question I had as well. I have a program that uses > >> wxLaunchDefaultBrowser which, looking at its implementation, tries to > make > >> the system call "exec()" to launch the default browser with a URL. > >> > >> If snap programs are not allowed to start other processes, that's fine; > >> but if enough people need to launch the default browser with a URL, > then I'm > >> sure a secure solution just for this could somehow be implemented for > snaps. > >> > >> I gather that one design goal of snaps, however, is the ability for > people > >> to write programs for any environment, but also have them work as snaps > so > >> that the programmer doesn't have to write snap-specific code, or make > >> snap-specific considerations in their code. In other words, your code > >> should be "none-the-wiser" that it is running in the confined area. > >> > >> So with that in mind, I'm not sure how to solve the problem. Any secure > >> API exposed to snap applications already breaks the above design goal. > >> > >> Of course, it's not unreasonable for my program to have "#ifdef WIN32" > or > >> "#ifdef UNIX", and in the latter case, I may be looking to utilize > something > >> in a standard unix environment which, I believe, is synthesized in > Unbuntu > >> Core. That's where I believe the snap environment can intercept what an > >> application is doing and provide a secure solution, and this may be the > >> "xdg-open" thing Otfried was talking about. > >> > >> > >> On Mon, Sep 19, 2016 at 2:37 AM, Otfried Cheong < > otfried at ipe.airpost.net> > >> wrote: > >>> > >>> Hello, > >>> > >>> my app has a manual in html. I normally show this using "xdg-open > >>> ", but from the snap this results in "xdg-open: Permission > denied", > >>> leaving this log: > >>> > >>> [21249.231634] audit: type=1400 audit(1474273861.873:383): > >>> apparmor="DENIED" operation="exec" profile="snap.ipe.sh" > >>> name="/usr/local/bin/xdg-open" pid=9551 comm="sh" requested_mask="x" > >>> denied_mask="x" fsuid=1000 ouid=0 > >>> > >>> According to > >>> https://lists.ubuntu.com/archives/snapcraft/2016-September/001048.html > >>> this should work. > >>> I did refresh ubuntu-core from the beta channel and currently have > >>> revision 636 of ubuntu-core. > >>> > >>> > >>> Slightly related: If I understand > >>> https://lists.ubuntu.com/archives/snapcraft/2016-September/001118.html > >>> correctly, the host filesystem should be exposed to the snap as > >>> /var/lib/snapd/hostfs in devmode? It isn't on my system. > >>> > >>> Cheers, > >>> Otfried > >>> > >>> > >>> -- > >>> Snapcraft mailing list > >>> Snapcraft at lists.snapcraft.io > >>> Modify settings or unsubscribe at: > >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft > >> > >> > >> > >> -- > >> Snapcraft mailing list > >> Snapcraft at lists.snapcraft.io > >> Modify settings or unsubscribe at: > >> https://lists.ubuntu.com/mailman/listinfo/snapcraft > >> > > > > > > > > -- > > Eloy García Almadén > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.snapcraft.io > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm > an/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Wed Sep 21 11:47:27 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 21 Sep 2016 08:47:27 -0300 Subject: Using xdg-open from snap In-Reply-To: References: <1474274258.3027854.729871177.629C4236@webmail.messagingengine.com> Message-ID: <80ddd52d-b9f8-68c3-19e0-932a1026d877@canonical.com> El 21/09/16 a las 08:18, Gustavo Niemeyer escribió: > That was a good explanation indeed, thanks John. > > Can we do something better than just recommend it on classic? The > feature is common enough that this should be a requirement, I think. Even though Recommends doesn't sound like a fit, this is its use: This declares a strong, but not absolute, dependency. The Recommends field should list packages that would be found together with this one in all but unusual installations.[1] Suggests would be the wrong dependency declaration though: This is used to declare that one package may be more useful with one or more others. Using this field tells the packaging system and the user that the listed packages are related to this one and can perhaps enhance its usefulness, but that installing this one without them is perfectly reasonable.[1] > > The problem then is how to drop the package when building the Ubuntu > Core image. If things are still the same, images are built with --no-recommends. Cheers Sergio [1] https://www.debian.org/doc/debian-policy/ch-relationships.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.ancell at canonical.com Thu Sep 22 03:48:18 2016 From: robert.ancell at canonical.com (Robert Ancell) Date: Thu, 22 Sep 2016 03:48:18 +0000 Subject: Removing a snap without using sudo or using /v2/login Message-ID: Hi all, There's an open bug for gnome-software about how you can't remove a snap without first logging in with your Ubuntu One credentials [1]. However, if you use the command line you can easily do this using 'sudo snap remove foo'. There are a number of cases where it makes sense to be able to remove a snap without logging in: - You may be offline and so can't do an Ubuntu One login at this time. - You don't have an Ubuntu One account but you want to remove something. - The snap came pre-installed, and you want to remove it. You have no need to log in and install more snaps (at that time). - You are a different user to the one who installed it. The reason gnome-software triggers the login is the following happens: 1. g-s finds the snap using GET /v2/snaps on snapd. 2. The user clicks "Remove" 3. g-s requests removal using POST /v2/snaps/[name] 4. snapd returns the error "login-required" 5. g-s prompts the user for email address / password 6. g-s asks the system D-Bus service snapd-login-service to login with the email address / password 7. snapd-login-service checks with Polkit that the user is allowed to do this 8. Polkit triggers a password prompt to check the user is there 9. snapd-login-service calls POST /v2/login on snapd 10. snapd returns a macaroon 11. snapd-login-service returns the macaroon to g-s 12. g-s repeats POST /v2/snaps/[name] using the macaroon The question is - does it make sense to make a new request to snapd that gets an authorization macaroon without logging into Ubuntu One? i.e. something like this: 1. g-s finds the snap using GET /v2/snaps on snapd. 2. The user clicks "Remove" 3. g-s requests removal using POST /v2/snaps/[name] 4. snapd returns the error "login-required" 5. g-s asks the system D-Bus service snapd-login-service to authorize 6. snapd-login-service checks with Polkit that the user is allowed to do this 7. Polkit triggers a password prompt to check the user is there 8. snapd-login-service calls POST /v2/authorize on snapd 9. snapd returns a macaroon 10. snapd-login-service returns the macaroon to g-s 11. g-s repeats POST /v2/snaps/[name] using the macaroon If we do this at what point do we prompt a user to do an Ubuntu One login? Do we have two error messages "auth-required" and "login-required"? Or is login an optional thing that we let the user decide if they want to do it? --Robert [1] https://bugs.launchpad.net/bugs/1581713 -------------- next part -------------- An HTML attachment was scrubbed... URL: From rex.tsai at canonical.com Thu Sep 22 13:24:32 2016 From: rex.tsai at canonical.com (Rex Tsai) Date: Thu, 22 Sep 2016 21:24:32 +0800 Subject: Docker image for building Ubuntu (amd64/i386/armhf/arm64) snaps Message-ID: This image includes toolchains and snapcraft to help app developers (and hardware enablement engineers) who don't have a Ubuntu installation to build a snap software package. It's also support building snap on amd64, i386 (native and amd64 architecture), armhf (native and amd64 architecture), arm64 (native and amd64 architecture). That help an app developer to cross build a arm-based snap from amd64. To use the docker container to build a # ​ ​ Please put this line in your ${HOME}/.$(basename ${SHELL})rc ​ alias snapcraft-docker='docker run -u $(id -u) -t -i --rm -v $(pwd):/build \ -v ${HOME}/.snap:/build/.snap \ -v ${HOME}/.local/share/snapcraft:/build/.local/share/snapcraft \ snapcraft/xenial-amd64 snapcraft' snapcraft-docker tour cd snapcraft-tour/00-SNAPCRAFT/01-easy-start && snapcraft-docker Please find the README on https://hub.docker.com/r/snapcraft/xenial-amd64/ Cheers -Rex -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Thu Sep 22 16:19:02 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Thu, 22 Sep 2016 18:19:02 +0200 Subject: Docker image for building Ubuntu (amd64/i386/armhf/arm64) snaps In-Reply-To: References: Message-ID: Hi! Is there a big difference with the https://hub.docker.com/r/snapcore/snapcraft/ image? Perhaps we can merge the two to avoid confusion from end-users? Thanks, - Loïc On Thu, Sep 22, 2016 at 3:24 PM, Rex Tsai wrote: > > This image includes toolchains and snapcraft to help app developers (and > hardware enablement engineers) who don't have a Ubuntu installation to > build a snap software package. > > It's also support building snap on amd64, i386 (native and amd64 > architecture), armhf (native and amd64 architecture), arm64 (native and > amd64 architecture). That help an app developer to cross build a arm-based > snap from amd64. > > To use the docker container to build a > # > ​ ​ > Please put this line in your > ${HOME}/.$(basename ${SHELL})rc > ​ > alias snapcraft-docker='docker run -u $(id -u) -t -i --rm -v $(pwd):/build > \ > -v ${HOME}/.snap:/build/.snap \ > -v ${HOME}/.local/share/snapcraft:/build/.local/share/snapcraft \ > snapcraft/xenial-amd64 snapcraft' > > snapcraft-docker tour > cd snapcraft-tour/00-SNAPCRAFT/01-easy-start && snapcraft-docker > > Please find the README on https://hub.docker.com/r/snapcraft/xenial-amd64/ > > Cheers > -Rex > -- - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From rex.tsai at canonical.com Thu Sep 22 16:28:03 2016 From: rex.tsai at canonical.com (Rex Tsai) Date: Fri, 23 Sep 2016 00:28:03 +0800 Subject: Docker image for building Ubuntu (amd64/i386/armhf/arm64) snaps In-Reply-To: References: Message-ID: Since the image[1] does not use the auto-build hook with github nor share the Dockerfile, I can only check the binary image. Form the binary image it looks like it only supports x86_64 and using an old 2.15.1 snapcraft. Not sure who is the maintainer. I'm aiming to support system engineers to build Ubuntu Core image, maybe we are not sharing the same goal here. If it's useful please find the source code on github[2]. [1] https://hub.docker.com/r/snapcore/snapcraft/ [2] https://github.com/chihchun/snapcraft-docker Cheers -Rex On Fri, Sep 23, 2016 at 12:19 AM, Loïc Minier wrote: > Hi! > > Is there a big difference with the https://hub.docker.com/r/ > snapcore/snapcraft/ image? Perhaps we can merge the two to avoid > confusion from end-users? > > Thanks, > - Loïc > > On Thu, Sep 22, 2016 at 3:24 PM, Rex Tsai wrote: > >> >> This image includes toolchains and snapcraft to help app developers (and >> hardware enablement engineers) who don't have a Ubuntu installation to >> build a snap software package. >> >> It's also support building snap on amd64, i386 (native and amd64 >> architecture), armhf (native and amd64 architecture), arm64 (native and >> amd64 architecture). That help an app developer to cross build a arm-based >> snap from amd64. >> >> To use the docker container to build a >> # >> ​ ​ >> Please put this line in your >> ${HOME}/.$(basename ${SHELL})rc >> ​ >> alias snapcraft-docker='docker run -u $(id -u) -t -i --rm -v >> $(pwd):/build \ >> -v ${HOME}/.snap:/build/.snap \ >> -v ${HOME}/.local/share/snapcraft:/build/.local/share/snapcraft \ >> snapcraft/xenial-amd64 snapcraft' >> >> snapcraft-docker tour >> cd snapcraft-tour/00-SNAPCRAFT/01-easy-start && snapcraft-docker >> >> Please find the README on https://hub.docker.com/r/snapc >> raft/xenial-amd64/ >> >> Cheers >> -Rex >> > > > > -- > - Loïc > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.wakeling at webdrake.net Thu Sep 22 23:28:20 2016 From: joseph.wakeling at webdrake.net (Joseph Rushton Wakeling) Date: Fri, 23 Sep 2016 01:28:20 +0200 Subject: Access to other commands In-Reply-To: <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> Message-ID: <0cd92c8f-291d-e0de-d387-9c5b4619c4d0@webdrake.net> On 15/09/16 14:12, Mark Shuttleworth wrote: > On 15/09/16 05:00, Zygmunt Krynicki wrote: >> As discussed a few times this is technically challenging to do. >> >> All of “classic” is visible from /var/lib/snapd/hostfs/ but there is no guarantee that you can run them in any way. They may require the classic dynamic linker, the classic runtime libraries and the classic filesystem layout that are all lost when snap-confine sets up the execution environment. If there’s desire to run executables from the outside we could look for solutions but this is not as simple as “just use devmode” > > I think this is a topic for the next snapfest community event, in > October/November. Call it "snapping CLI utilities". Is this going to be an online or face-to-face event? This seems like something I ought to not miss ... :-) From leo.arias at canonical.com Fri Sep 23 02:13:05 2016 From: leo.arias at canonical.com (Leo Arias) Date: Thu, 22 Sep 2016 20:13:05 -0600 Subject: Access to other commands In-Reply-To: <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> Message-ID: On Thu, Sep 15, 2016 at 6:12 AM, Mark Shuttleworth wrote: > > I think this is a topic for the next snapfest community event, in > October/November. Call it "snapping CLI utilities". > > Here's another interesting one to take into account in that discussion: https://github.com/elopio/htop/tree/snapcraft (currently in the sandpit) In devmode, snappy-debug doesn't stop printing warnings. -- ¡paz y baile! http://www.ubuntu.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From spencertparkin at gmail.com Fri Sep 23 05:49:39 2016 From: spencertparkin at gmail.com (Spencer) Date: Thu, 22 Sep 2016 23:49:39 -0600 Subject: Ratings Message-ID: <3E38B41A-A98B-4407-90E7-2360B19C3B07@gmail.com> I'm about to add audio to my snap, yo, the ratings are about to go through the roof, dawg!... Say, how do you rate a snap, anyway? From simon.fels at canonical.com Fri Sep 23 06:36:36 2016 From: simon.fels at canonical.com (Simon Fels) Date: Fri, 23 Sep 2016 08:36:36 +0200 Subject: CLI UX for plugs with parameters? In-Reply-To: References: Message-ID: You wont set the attribute of the slot/plug in the command line you use to connect slot and plug. Every attribute as to be set in the definition of the plug/slot. Like plugs: my-plug: interface: really-awesome-interface attribute1: value1 ... Then you simply connect via $ snap connect my-snap:my-plug .. without specifying any attributes. regards, Simon On Tue, Sep 20, 2016 at 9:52 AM, Stuart Bishop wrote: > Hi. > > I believe that slots will have support for arguments, for example allowing > a snap to connect to the 'home' slot and provide an argument requesting > access to hidden files. > > Does anyone know what the command line interface will look like? > > In the snap layer for Juju, I need to define the data structure to declare > what additional connections need to be made after installation of the snap. > I'd like to get this right the first time since I'm going to be stuck with > it. > > > -- > Stuart Bishop > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Fri Sep 23 12:40:15 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Fri, 23 Sep 2016 07:40:15 -0500 Subject: Access to other commands In-Reply-To: References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> Message-ID: <1474634415.31629.25.camel@canonical.com> On Thu, 2016-09-22 at 20:13 -0600, Leo Arias wrote: > On Thu, Sep 15, 2016 at 6:12 AM, Mark Shuttleworth wrote: > > > > > > > I think this is a topic for the next snapfest community event, in > > October/November. Call it "snapping CLI utilities". > > > > > Here's another interesting one to take into account in that discussion: > https://github.com/elopio/htop/tree/snapcraft > (currently in the sandpit) > > In devmode, snappy-debug doesn't stop printing warnings. I don't see a snapcraft.yaml in that tree. There is an htop snap in the store and you need to connect the process-control and system-observe interfaces: $ sudo snap install htop $ sudo snap connect htop:process-control ubuntu-core:process-control $ sudo snap connect htop:system-observe ubuntu-core:system-observe $ htop (no denials) That is in strict mode. You can also install in devmode but you need to connect the interfaces for the log messages to go away. This is because devmode reports (but allows) violations against policy. If you don't connect the interfaces then the accesses aren't part of the allowed policy and you will see a lot of policy violations. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From mabnhdev at gmail.com Fri Sep 23 20:11:22 2016 From: mabnhdev at gmail.com (MikeB) Date: Fri, 23 Sep 2016 16:11:22 -0400 Subject: Updated Ubuntu Core 16 beta images available Message-ID: Do you know of any Vagrant base images for Ubuntu Core 16? I'd like to avoid having to build/maintain myself. Mike On 19-Sep-2016 at 08:22:25 -0700, Michael Vogt wrote: > > The images are available for PC (amd64, i386) and Raspberry Pi2/3 > (armhf) and the Dragonboard 410. You can download them at: > http://cdimage.ubuntu.com/ubuntu-snappy/16.04/current/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From tomekchacuk at gmail.com Sun Sep 25 11:31:17 2016 From: tomekchacuk at gmail.com (Tomasz Chacuk) Date: Sun, 25 Sep 2016 13:31:17 +0200 Subject: Ubuntu core 16 from 19.09.2016 soes not work on RPI3 Message-ID: Hello, I wanted to try new beta from 19.09 on my RPI3, but whenever i try tu burn the image (Win32Imager) loading hangs just after kernel strts loading - i get the image of 4 rapbberies on the top of the screen but no further messages - this hangs forever. I have only HDMI+Ethernet plugged (tried plogging and unpluggin USB keyboard - no effect) Will be grateful for info. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Sun Sep 25 12:33:57 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Sun, 25 Sep 2016 14:33:57 +0200 Subject: Ubuntu core 16 from 19.09.2016 soes not work on RPI3 In-Reply-To: References: Message-ID: <1474806837.14956.9.camel@ubuntu.com> hi, Am Sonntag, den 25.09.2016, 13:31 +0200 schrieb Tomasz Chacuk: > Hello, > > I wanted to try new beta from 19.09 on my RPI3, but whenever i try tu > burn the image (Win32Imager) loading hangs just after kernel strts > loading - i get the image of 4 rapbberies on the top of the screen > but no further messages - this hangs forever. I have only > HDMI+Ethernet plugged (tried plogging and unpluggin USB keyboard - no > effect) > > Will be grateful for info. please try a daily image from  http://people.canonical.com/~ogra/snappy/all-snaps/daily/current/ this has a second (non-serial) console enabled  ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From spencertparkin at gmail.com Mon Sep 26 00:28:40 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Sun, 25 Sep 2016 18:28:40 -0600 Subject: pulseaudio plug/slot Message-ID: Hi, I'm adding sound-fx to my snap-app and am running into the common problem of "it works in my real system, but not in the confined system." I'm not sure how to trouble-shoot this problem. I've added the "pulseaudio" plug to my snapcraft.yaml file, "snap interface" reports that my snap-app is using the "pulseaudio" plug, I've tried exporting the "SDL_AUDIODRIVER=pulseaudio" environment variable before running my snap to make sure that I'm using the appropriate audio back-end. Hmmm...maybe snap-apps can't read environment variables? Maybe I need to configure some env-vars in my snapcraft.yaml file? I'll try to see how to setup the environment (env-vars) for snaps. Anyhow, I can hear my sound-fx when run on my real system, but not under unbuntu core. Thanks for any ideas you may have, --Sp -------------- next part -------------- An HTML attachment was scrubbed... URL: From spencertparkin at gmail.com Mon Sep 26 04:27:08 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Sun, 25 Sep 2016 22:27:08 -0600 Subject: pulseaudio plug/slot In-Reply-To: References: Message-ID: So I did two things to resolve this, I'm not sure if one of them was necessary, but... 1) I changed my app command to "env SDL_AUDIODRIVER=pulseaudio desktop-launch $SNAP/bin/[myprog]" 2) I changed the confinement from strict to devmode. It's probably the latter that solved the problem. If the snapcraft developers are interested in testing out why audio doesn't work in strict mode, they can use my program as an test case, if they cared to. On Sun, Sep 25, 2016 at 6:28 PM, Spencer Parkin wrote: > Hi, > > I'm adding sound-fx to my snap-app and am running into the common problem > of "it works in my real system, but not in the confined system." I'm not > sure how to trouble-shoot this problem. I've added the "pulseaudio" plug > to my snapcraft.yaml file, "snap interface" reports that my snap-app is > using the "pulseaudio" plug, I've tried exporting the > "SDL_AUDIODRIVER=pulseaudio" environment variable before running my snap to > make sure that I'm using the appropriate audio back-end. Hmmm...maybe > snap-apps can't read environment variables? Maybe I need to configure some > env-vars in my snapcraft.yaml file? I'll try to see how to setup the > environment (env-vars) for snaps. > > Anyhow, I can hear my sound-fx when run on my real system, but not under > unbuntu core. > > Thanks for any ideas you may have, > --Sp > -------------- next part -------------- An HTML attachment was scrubbed... URL: From spencertparkin at gmail.com Mon Sep 26 05:16:36 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Sun, 25 Sep 2016 23:16:36 -0600 Subject: pulseaudio plug/slot In-Reply-To: References: Message-ID: Ugh...devmode snaps can't be released on the stable channel. And the devmode snap is probably accessing the outside environment to solve the problem anyway, which defeats the purpose. So I don't know how to solve the problem. If anyone cares to see my crappy snap-app, it is at... https://github.com/spencerparkin/ChineseCheckers.git Enable sound, then hold ALT+SHIFT while doing a left-click in the window to test sound. On Sun, Sep 25, 2016 at 10:27 PM, Spencer Parkin wrote: > So I did two things to resolve this, I'm not sure if one of them was > necessary, but... > > 1) I changed my app command to "env SDL_AUDIODRIVER=pulseaudio > desktop-launch $SNAP/bin/[myprog]" > 2) I changed the confinement from strict to devmode. > > It's probably the latter that solved the problem. If the snapcraft > developers are interested in testing out why audio doesn't work in strict > mode, they can use my program as an test case, if they cared to. > > > On Sun, Sep 25, 2016 at 6:28 PM, Spencer Parkin > wrote: > >> Hi, >> >> I'm adding sound-fx to my snap-app and am running into the common problem >> of "it works in my real system, but not in the confined system." I'm not >> sure how to trouble-shoot this problem. I've added the "pulseaudio" plug >> to my snapcraft.yaml file, "snap interface" reports that my snap-app is >> using the "pulseaudio" plug, I've tried exporting the >> "SDL_AUDIODRIVER=pulseaudio" environment variable before running my snap to >> make sure that I'm using the appropriate audio back-end. Hmmm...maybe >> snap-apps can't read environment variables? Maybe I need to configure some >> env-vars in my snapcraft.yaml file? I'll try to see how to setup the >> environment (env-vars) for snaps. >> >> Anyhow, I can hear my sound-fx when run on my real system, but not under >> unbuntu core. >> >> Thanks for any ideas you may have, >> --Sp >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Mon Sep 26 05:18:13 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Mon, 26 Sep 2016 07:18:13 +0200 Subject: pulseaudio plug/slot In-Reply-To: References: Message-ID: <26ff6c3f-8820-1d03-05ec-e48c028c1cd1@ubuntu.com> Le 26/09/2016 à 06:27, Spencer Parkin a écrit : > So I did two things to resolve this, I'm not sure if one of them was > necessary, but... > > 1) I changed my app command to "env SDL_AUDIODRIVER=pulseaudio > desktop-launch $SNAP/bin/[myprog]" > 2) I changed the confinement from strict to devmode. > > It's probably the latter that solved the problem. If the snapcraft > developers are interested in testing out why audio doesn't work in > strict mode, they can use my program as an test case, if they cared to. Hey Spencer! Some things you can do to test in strict mode is to tail /var/log/syslog, running your program, and check the apparmor denial you probably get when acessing to some files/do some calls. That would be a great hint for you on debugging this issue, and report a bug if a new interface tweak is needed! Remember that you won't be able to publish on the stable channel if your snap is in devmode. Cheers, Didier > > On Sun, Sep 25, 2016 at 6:28 PM, Spencer Parkin > > wrote: > > Hi, > > I'm adding sound-fx to my snap-app and am running into the common > problem of "it works in my real system, but not in the confined > system." I'm not sure how to trouble-shoot this problem. I've > added the "pulseaudio" plug to my snapcraft.yaml file, "snap > interface" reports that my snap-app is using the "pulseaudio" > plug, I've tried exporting the "SDL_AUDIODRIVER=pulseaudio" > environment variable before running my snap to make sure that I'm > using the appropriate audio back-end. Hmmm...maybe snap-apps > can't read environment variables? Maybe I need to configure some > env-vars in my snapcraft.yaml file? I'll try to see how to setup > the environment (env-vars) for snaps. > > Anyhow, I can hear my sound-fx when run on my real system, but not > under unbuntu core. > > Thanks for any ideas you may have, > --Sp > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.mcaleely at canonical.com Mon Sep 26 17:12:37 2016 From: john.mcaleely at canonical.com (John McAleely) Date: Mon, 26 Sep 2016 18:12:37 +0100 Subject: Updated Ubuntu Core 16 beta images available In-Reply-To: <20160919152128.GB27970@bod> References: <20160919152128.GB27970@bod> Message-ID: On 19 September 2016 at 16:21, Michael Vogt wrote: > Ubuntu Core 16 Images > ===================== > > The Ubuntu snappy team is happy to announce new beta images for Ubuntu > Core 16. The images use the snapd package manager to install and > update all components of the system including kernel, core, gadget and > applications. > > Some hightlights for this release: > - a preview of snapweb is installed by default, snapweb is a web-ui to > control an all-snap system > - pi3 image available > - smaller image size to fit better on sd-cards > - bugfixes > > The images are available for PC (amd64, i386) and Raspberry Pi2/3 > (armhf) and the Dragonboard 410. You can download them at: > > On the Pi2 image (the only one I've examined so far), should I expect to find libGLESv2.so, or some other evidence of OpenGL ES API support? I had heard that was expected. from the login shell: $ ls /lib/arm-linux-gnueabihf/libGLES* Am I missing something? -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Mon Sep 26 17:27:23 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Mon, 26 Sep 2016 19:27:23 +0200 Subject: Updated Ubuntu Core 16 beta images available In-Reply-To: References: <20160919152128.GB27970@bod> Message-ID: <1474910843.9676.13.camel@ubuntu.com> hi, On Mo, 2016-09-26 at 18:12 +0100, John McAleely wrote: > > On 19 September 2016 at 16:21, Michael Vogt om> wrote: > > Ubuntu Core 16 Images > > ===================== > > > > The Ubuntu snappy team is happy to announce new beta images for > > Ubuntu > > Core 16.  The images use the snapd package manager to install and > > update all components of the system including kernel, core, gadget > > and > > applications. > > > > Some hightlights for this release: > > - a preview of snapweb is installed by default, snapweb is a web-ui > > to > >   control an all-snap system > > - pi3 image available > > - smaller image size to fit better on sd-cards > > - bugfixes > > > > The images are available for PC (amd64, i386) and Raspberry Pi2/3 > > (armhf) and the Dragonboard 410. You can download them at: > > > > On the Pi2 image (the only one I've examined so far), should I expect > to find libGLESv2.so, or some other evidence of OpenGL ES API > support? > > I had heard that was expected. > > from the login shell: > > $ ls /lib/arm-linux-gnueabihf/libGLES* > > > Am I missing something? nope ... you dont ... the libs need to be shipped by the kernel snap but the new kernel snap definition does not allow anything beyond the default kernel bits (modules, firmware, dtbs) yet... so currently we can not ship any GLES on armhf til this is implemented. i think kevin found a way to ship GLES in his mir snap though, perhaps that could be a fallback for the moment ?  ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From john.mcaleely at canonical.com Mon Sep 26 17:37:52 2016 From: john.mcaleely at canonical.com (John McAleely) Date: Mon, 26 Sep 2016 18:37:52 +0100 Subject: OpenGL ES API support (was Re: Updated Ubuntu Core 16 beta images available) Message-ID: On 26 September 2016 at 18:27, Oliver Grawert wrote: > On Mo, 2016-09-26 at 18:12 +0100, John McAleely wrote: > > > > On the Pi2 image (the only one I've examined so far), should I expect > > to find libGLESv2.so, or some other evidence of OpenGL ES API > > support? > > > > > Am I missing something? > > nope ... you dont ... the libs need to be shipped by the kernel snap > but the new kernel snap definition does not allow anything beyond the > default kernel bits (modules, firmware, dtbs) yet... so currently we > can not ship any GLES on armhf til this is implemented. > > i think kevin found a way to ship GLES in his mir snap though, perhaps > that could be a fallback for the moment ? > Yes & no. The snap I'm working on would be a neat demo on two ARM boards that happen to use different GPUs (the RPi is one of them). Maybe we can add complexity to the workaround and ship two different GLES libraries, and select the right one somehow, but perhaps it would be better to wait (kgunn - if you already have all that tech, a pointer would be appreciated!) When should we expect support for these API that need a user-side lib (if I've captured the requirement correctly) from the kernel snap in Snappy? (perhaps not a question you can answer?) J -------------- next part -------------- An HTML attachment was scrubbed... URL: From kevin.gunn at canonical.com Mon Sep 26 21:26:18 2016 From: kevin.gunn at canonical.com (Kevin Gunn) Date: Mon, 26 Sep 2016 16:26:18 -0500 Subject: OpenGL ES API support (was Re: Updated Ubuntu Core 16 beta images available) In-Reply-To: References: Message-ID: @including the gpu drivers, yes, this is pretty simple - at least with dragonboard, you simply pull in mesa as a part and you get the freedreno driver and you would honestly get the same with rpi2/3 since afaik I think the vc5 drivers under gallium will get you what you want. Now with rpi2/3, the hitch isn't actually the gpu drivers per se, it's that our rpi kernel doesn't have some patches required that are available from the raspbian kernel - but that's the extent of my knowledge. tvoss was able to patch and rebuild the kernel about a week ago. So what's needed is having someone from the kernel team take custody of those patches and incorporate them into our kernel. br,kg On Mon, Sep 26, 2016 at 12:37 PM, John McAleely wrote: > > > On 26 September 2016 at 18:27, Oliver Grawert wrote: > >> On Mo, 2016-09-26 at 18:12 +0100, John McAleely wrote: >> > >> > On the Pi2 image (the only one I've examined so far), should I expect >> > to find libGLESv2.so, or some other evidence of OpenGL ES API >> > support? >> > > >> > >> > Am I missing something? >> >> nope ... you dont ... the libs need to be shipped by the kernel snap >> but the new kernel snap definition does not allow anything beyond the >> default kernel bits (modules, firmware, dtbs) yet... so currently we >> can not ship any GLES on armhf til this is implemented. >> >> i think kevin found a way to ship GLES in his mir snap though, perhaps >> that could be a fallback for the moment ? >> > > Yes & no. The snap I'm working on would be a neat demo on two ARM boards > that happen to use different GPUs (the RPi is one of them). Maybe we can > add complexity to the workaround and ship two different GLES libraries, and > select the right one somehow, but perhaps it would be better to wait (kgunn > - if you already have all that tech, a pointer would be appreciated!) > > When should we expect support for these API that need a user-side lib (if > I've captured the requirement correctly) from the kernel snap in Snappy? > (perhaps not a question you can answer?) > > J > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Mon Sep 26 21:38:55 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Mon, 26 Sep 2016 23:38:55 +0200 Subject: OpenGL ES API support (was Re: Updated Ubuntu Core 16 beta images available) In-Reply-To: References: Message-ID: <1474925935.9676.28.camel@ubuntu.com> hi, On Mo, 2016-09-26 at 16:26 -0500, Kevin Gunn wrote: > @including the gpu drivers, yes, this is pretty simple - at least > with dragonboard, you simply pull in mesa as a part and you get the > freedreno driver and you would honestly get the same with rpi2/3 > since afaik I think the vc5 drivers under gallium will get you what > you want. > Now with rpi2/3, the hitch isn't actually the gpu drivers per se, > it's that our rpi kernel doesn't have some patches required that are > available from the raspbian kernel - but that's the extent of my > knowledge. tvoss was able to patch and rebuild the kernel about a > week ago. So what's needed is having someone from the kernel team > take custody of those patches and incorporate them into our kernel. >  can you make sure paolo (CCed) gets these patches in his hand so we can get them added to our deb and snap packages ?  ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From thomas.voss at canonical.com Mon Sep 26 21:40:46 2016 From: thomas.voss at canonical.com (=?UTF-8?Q?Thomas_Vo=C3=9F?=) Date: Mon, 26 Sep 2016 23:40:46 +0200 Subject: OpenGL ES API support (was Re: Updated Ubuntu Core 16 beta images available) In-Reply-To: <1474925935.9676.28.camel@ubuntu.com> References: <1474925935.9676.28.camel@ubuntu.com> Message-ID: On Mon, Sep 26, 2016 at 11:38 PM, Oliver Grawert wrote: > hi, > On Mo, 2016-09-26 at 16:26 -0500, Kevin Gunn wrote: >> @including the gpu drivers, yes, this is pretty simple - at least >> with dragonboard, you simply pull in mesa as a part and you get the >> freedreno driver and you would honestly get the same with rpi2/3 >> since afaik I think the vc5 drivers under gallium will get you what >> you want. >> Now with rpi2/3, the hitch isn't actually the gpu drivers per se, >> it's that our rpi kernel doesn't have some patches required that are >> available from the raspbian kernel - but that's the extent of my >> knowledge. tvoss was able to patch and rebuild the kernel about a >> week ago. So what's needed is having someone from the kernel team >> take custody of those patches and incorporate them into our kernel. >> > can you make sure paolo (CCed) gets these patches in his hand so we can > get them added to our deb and snap packages ? > Paolo is working with me already, no worries :) Cheers, Thomas > ciao > oli > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > From spencertparkin at gmail.com Mon Sep 26 23:39:47 2016 From: spencertparkin at gmail.com (Spencer Parkin) Date: Mon, 26 Sep 2016 17:39:47 -0600 Subject: pulseaudio plug/slot In-Reply-To: <26ff6c3f-8820-1d03-05ec-e48c028c1cd1@ubuntu.com> References: <26ff6c3f-8820-1d03-05ec-e48c028c1cd1@ubuntu.com> Message-ID: Thanks to Didier, I got my snap app's audio working in strict mode. I needed to add the "process-control" plug, probably because my application needs to spawn an audio-thread for feeding the audio device. I have a few other app-armor denials that happen, but I'll look at those later when I get the chance. Thanks again, --Sp On Sun, Sep 25, 2016 at 11:18 PM, Didier Roche wrote: > Le 26/09/2016 à 06:27, Spencer Parkin a écrit : > > So I did two things to resolve this, I'm not sure if one of them was > necessary, but... > > 1) I changed my app command to "env SDL_AUDIODRIVER=pulseaudio > desktop-launch $SNAP/bin/[myprog]" > 2) I changed the confinement from strict to devmode. > > It's probably the latter that solved the problem. If the snapcraft > developers are interested in testing out why audio doesn't work in strict > mode, they can use my program as an test case, if they cared to. > > Hey Spencer! > > Some things you can do to test in strict mode is to tail /var/log/syslog, > running your program, and check the apparmor denial you probably get when > acessing to some files/do some calls. > That would be a great hint for you on debugging this issue, and report a > bug if a new interface tweak is needed! > > Remember that you won't be able to publish on the stable channel if your > snap is in devmode. > > Cheers, > Didier > > > On Sun, Sep 25, 2016 at 6:28 PM, Spencer Parkin > wrote: > >> Hi, >> >> I'm adding sound-fx to my snap-app and am running into the common problem >> of "it works in my real system, but not in the confined system." I'm not >> sure how to trouble-shoot this problem. I've added the "pulseaudio" plug >> to my snapcraft.yaml file, "snap interface" reports that my snap-app is >> using the "pulseaudio" plug, I've tried exporting the >> "SDL_AUDIODRIVER=pulseaudio" environment variable before running my snap to >> make sure that I'm using the appropriate audio back-end. Hmmm...maybe >> snap-apps can't read environment variables? Maybe I need to configure some >> env-vars in my snapcraft.yaml file? I'll try to see how to setup the >> environment (env-vars) for snaps. >> >> Anyhow, I can hear my sound-fx when run on my real system, but not under >> unbuntu core. >> >> Thanks for any ideas you may have, >> --Sp >> > > > > > > -- > Snapcraft mailing list > Snapcraft at lists.snapcraft.io > Modify settings or unsubscribe at: https://lists.ubuntu.com/ > mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From leo.arias at canonical.com Tue Sep 27 00:15:05 2016 From: leo.arias at canonical.com (Leo Arias) Date: Mon, 26 Sep 2016 18:15:05 -0600 Subject: sqlite trying to access /var/tmp Message-ID: Hello, I'm trying to snap mapbox studio: https://gist.github.com/elopio/d4e84e0d921151a68445a15d8dbc6ced It's failing when I try to create a map because sqlite tries to write to /var/tmp. According to [1], sqlite uses $TMPDIR before trying to use /var/tmp, so this is weird. Is anybody around with more sqlite knowledge who can help me? pura vida. [1] https://www.sqlite.org/tempfiles.html -- ¡paz y baile! http://www.ubuntu.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From mhall119 at ubuntu.com Tue Sep 27 13:25:23 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Tue, 27 Sep 2016 09:25:23 -0400 Subject: sqlite trying to access /var/tmp In-Reply-To: References: Message-ID: Set SQLITE_TMPDIR=/tmp and it should be happy. TMPDIR isn't explicitly set in the snap's runtime environment, which is why sqlite falls back to /var/tmp /tmp is mounted into your snap's runtime, and sqlite will try SQLITE_TMPDIR before TMPDIR, so setting that ensures that it won't impact anything other than sqlite. Michael Hall mhall119 at ubuntu.com On 09/26/2016 08:15 PM, Leo Arias wrote: > Hello, > > I'm trying to snap mapbox > studio: https://gist.github.com/elopio/d4e84e0d921151a68445a15d8dbc6ced > > It's failing when I try to create a map because sqlite tries to write to > /var/tmp. According to [1], sqlite uses $TMPDIR before trying to use > /var/tmp, so this is weird. > > Is anybody around with more sqlite knowledge who can help me? > > pura vida. > > [1] https://www.sqlite.org/tempfiles.html > > -- > ¡paz y baile! > http://www.ubuntu.com > > From daniel.holbach at canonical.com Tue Sep 27 14:34:43 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Tue, 27 Sep 2016 16:34:43 +0200 Subject: Learning to snap with codelabs Message-ID: <95dccf7b-04c7-1bf3-7d77-885dce96fa4b@canonical.com> Hello everybody, learning something new, especially new concepts and workflows usually works best if you see it first-hand and get to do things yourself. Didier and I talked a while about how to introduce the processes and ideas behind snapd and snapcraft to a new audience, particularly at a workshop or meetup and we found we were of the same opinion. Didier put quite a bit of work into solving the infrastructure question. We re-used the work which was put into Codelabs already, so adding a new codelab merely became a question of creating a Google Doc and adding it using a management command. It works nicely, the UI is simple and easy to understand and lets you focus on the content at hand. It was a lot of fun to work on the content and refine the individual steps in a self-teaching workshop style. Thanks a lot everyone for the reviews! After some discussion it became clear that a very fitting way for the codelabs to go out would be to ship them as a snap themselves. It's beautifully simple to get started: $ sudo snap install snap-codelabs All you need to do afterwards is point your browser to http://localhost:8123/ - that's all. >From thereon you can quickly start your snap adventure and get up and running in no time. It's a step-by-step workshop and you always know how much more time you need to complete it. Expect more codelabs to be added soon. If you have feedback, please let us know at https://github.com/ubuntu/codelabs Enjoy and let us know how things go. Have a great day, Daniel From jamie at canonical.com Tue Sep 27 15:29:34 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 27 Sep 2016 10:29:34 -0500 Subject: sqlite trying to access /var/tmp In-Reply-To: References: Message-ID: <1474990174.6567.34.camel@canonical.com> On Tue, 2016-09-27 at 09:25 -0400, Michael Hall wrote: > Set SQLITE_TMPDIR=/tmp and it should be happy. > > TMPDIR isn't explicitly set in the snap's runtime environment, which is > why sqlite falls back to /var/tmp > > /tmp is mounted into your snap's runtime, and sqlite will try > SQLITE_TMPDIR before TMPDIR, so setting that ensures that it won't > impact anything other than sqlite. > Hmmm, we used to export TMPDIR but it seems that it was dropped when moving to snap reexec. Can you file a bug so that sqlite users don't have to worry about SQLITE_TMPDIR unless they want to? I think the bug is simply: 'please set TMPDIR=/tmp when launching snaps' Thanks! > > Michael Hall > mhall119 at ubuntu.com > > On 09/26/2016 08:15 PM, Leo Arias wrote: > > > > Hello, > > > > I'm trying to snap mapbox > > studio: https://gist.github.com/elopio/d4e84e0d921151a68445a15d8dbc6ced > > > > It's failing when I try to create a map because sqlite tries to write to > > /var/tmp. According to [1], sqlite uses $TMPDIR before trying to use > > /var/tmp, so this is weird. > > > > Is anybody around with more sqlite knowledge who can help me? > > > > pura vida. > > > > [1] https://www.sqlite.org/tempfiles.html > > > > --  > > ¡paz y baile! > > http://www.ubuntu.com > > > > -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From leo.arias at canonical.com Tue Sep 27 15:46:09 2016 From: leo.arias at canonical.com (Leo Arias) Date: Tue, 27 Sep 2016 09:46:09 -0600 Subject: Access to other commands In-Reply-To: <1474634415.31629.25.camel@canonical.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> <1474634415.31629.25.camel@canonical.com> Message-ID: On Fri, Sep 23, 2016 at 6:40 AM, Jamie Strandboge wrote: > On Thu, 2016-09-22 at 20:13 -0600, Leo Arias wrote: > I don't see a snapcraft.yaml in that tree. It's here: https://github.com/elopio/htop/blob/snapcraft/snapcraft.yaml > There is an htop snap in the store > and you need to connect the process-control and system-observe interfaces: > > $ sudo snap install htop > $ sudo snap connect htop:process-control ubuntu-core:process-control > $ sudo snap connect htop:system-observe ubuntu-core:system-observe > $ htop > > (no denials) > > That is in strict mode. You can also install in devmode but you need to > connect > the interfaces for the log messages to go away. This is because devmode > reports > (but allows) violations against policy. If you don't connect the > interfaces then > the accesses aren't part of the allowed policy and you will see a lot of > policy > violations. > That's awesome. Have you tried to land it in the repo upstream? -- ¡paz y baile! http://www.ubuntu.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Tue Sep 27 16:01:11 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 27 Sep 2016 11:01:11 -0500 Subject: Access to other commands In-Reply-To: References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> <1474634415.31629.25.camel@canonical.com> Message-ID: <1474992071.6567.48.camel@canonical.com> On Tue, 2016-09-27 at 09:46 -0600, Leo Arias wrote: > On Fri, Sep 23, 2016 at 6:40 AM, Jamie Strandboge > wrote: > > > > > On Thu, 2016-09-22 at 20:13 -0600, Leo Arias wrote: > > I don't see a snapcraft.yaml in that tree. > > It's here: https://github.com/elopio/htop/blob/snapcraft/snapcraft.yaml > > > > > > There is an htop snap in the store > > and you need to connect the process-control and system-observe interfaces: > > > > $ sudo snap install htop > > $ sudo snap connect htop:process-control ubuntu-core:process-control > > $ sudo snap connect htop:system-observe ubuntu-core:system-observe > > $ htop > > > > (no denials) > > > > That is in strict mode. You can also install in devmode but you need to > > connect > > the interfaces for the log messages to go away. This is because devmode > > reports > > (but allows) violations against policy. If you don't connect the > > interfaces then > > the accesses aren't part of the allowed policy and you will see a lot of > > policy > > violations. > > > That's awesome. Have you tried to land it in the repo upstream? No, this isn't my snap. The snap developer for the snap in the store is listed as 'maxiberta'. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From leo.arias at canonical.com Tue Sep 27 16:19:36 2016 From: leo.arias at canonical.com (Leo Arias) Date: Tue, 27 Sep 2016 10:19:36 -0600 Subject: sqlite trying to access /var/tmp In-Reply-To: <1474990174.6567.34.camel@canonical.com> References: <1474990174.6567.34.camel@canonical.com> Message-ID: Thanks Michael. That worked. Now it gets stuck somewhere and runs out of memory, but I doubt that's the fault of the snap. Jamie, here is the bug: https://bugs.launchpad.net/snappy/+bug/1628193 I also found this related bug from a year ago: https://bugs.launchpad.net/snappy/+bug/1504187 pura vida -------------- next part -------------- An HTML attachment was scrubbed... URL: From leo.arias at canonical.com Tue Sep 27 16:30:08 2016 From: leo.arias at canonical.com (Leo Arias) Date: Tue, 27 Sep 2016 10:30:08 -0600 Subject: Access to other commands In-Reply-To: <1474992071.6567.48.camel@canonical.com> References: <99608bf4-7673-325c-5005-a3bd6a9f94ea@ubuntu.com> <57D9E08E.40807@canonical.com> <8AB44A88-4051-4F7A-8740-A50DC3AD3819@canonical.com> <57ae50cc-eef9-55b2-2948-bee94227db4b@ubuntu.com> <1474634415.31629.25.camel@canonical.com> <1474992071.6567.48.camel@canonical.com> Message-ID: On Tue, Sep 27, 2016 at 10:01 AM, Jamie Strandboge wrote: > No, this isn't my snap. The snap developer for the snap in the store is > listed > as 'maxiberta'. > > I've contacted him. Thanks Jamie. -- ¡paz y baile! http://www.ubuntu.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Wed Sep 28 07:57:00 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Wed, 28 Sep 2016 09:57:00 +0200 Subject: Writing snaps together Message-ID: Hello everybody, as part of the Snappy Playpen initiative we set up https://github.com/ubuntu/snappy-playpen/wiki/SandPit which is meant to be a list of current work-in-progress snaps. We wanted to create a space where we can easily collaborate on snaps and get them working together. Editing is easy, all you need is a Github account. If you have snaps on your hard-disk which you didn't quite get to finish yet, please add them there. If you're planning to create a new snap, make sure you have a look first and base your work on somebody else's. Happy snapping everyone! Have a great day, Daniel From julia.palandri at canonical.com Wed Sep 28 08:36:51 2016 From: julia.palandri at canonical.com (Julia Palandri) Date: Wed, 28 Sep 2016 10:36:51 +0200 Subject: Writing snaps together In-Reply-To: References: Message-ID: Hey, On Wed, Sep 28, 2016 at 9:57 AM, Daniel Holbach < daniel.holbach at canonical.com> wrote: > Hello everybody, > > as part of the Snappy Playpen initiative we set up > > https://github.com/ubuntu/snappy-playpen/wiki/SandPit > > which is meant to be a list of current work-in-progress snaps. We wanted > to create a space where we can easily collaborate on snaps and get them > working together. Editing is easy, all you need is a Github account. > > If you have snaps on your hard-disk which you didn't quite get to finish > yet, please add them there. If you're planning to create a new snap, > make sure you have a look first and base your work on somebody else's. > > Awesome idea! So the work done doesn't necessarily get wasted because of lack of time of will to continue! I myself are stuck with some snaps, will add them there. Thanks! -- Julia -------------- next part -------------- An HTML attachment was scrubbed... URL: From alberto.mardegan at canonical.com Wed Sep 28 08:54:40 2016 From: alberto.mardegan at canonical.com (Alberto Mardegan) Date: Wed, 28 Sep 2016 11:54:40 +0300 Subject: sqlite trying to access /var/tmp In-Reply-To: References: <1474990174.6567.34.camel@canonical.com> Message-ID: <0cadadd8-4bd4-4090-ef45-5d3331f2d513@canonical.com> On 27/09/2016 19:19, Leo Arias wrote: > Thanks Michael. That worked. Now it gets stuck somewhere and runs out of > memory, but I doubt that's the fault of the snap. It might be a totally different issue, but IIRC I also happened to see "out of memory" issues in my projects using SQLite, while in fact the error was that I was asking it to create the SQLite database in a directory which I didn't create yet. Just to save you some time, please double check that the path to the SQLite database exists and is accessible to the snap. Ciao, Alberto From michael.foley at canonical.com Wed Sep 28 23:42:20 2016 From: michael.foley at canonical.com (Michael Foley) Date: Thu, 29 Sep 2016 12:42:20 +1300 Subject: snapping a python2 application that uses a makefile Message-ID: Hello, I've started working on a few snaps last month and worked out how to build an initial snap of the gpodder podcast client. However in order to get it to build I had to do a few manual steps during the build step. This is a python2 application, but actually uses a makefile. I tried a few different tactics including trying to use the make plugin, but that does not create the command wrapper properly. And actually when I tried to rebuild my snap today the command wrapper is failing to find and exec gpodder. I think the command wrapper is no longer setting the PYTHONPATH. Has this changed in the python2 plugin. I see that is deprecated so tried the python plugin with "python-version: python2", but got the same result. A copy of my snapcraft.yaml file is here: http://pastebin.ubuntu.com/23248926/ Any suggestions on ways to work around these issues? Thanks, Michael From leo.arias at canonical.com Fri Sep 30 02:27:43 2016 From: leo.arias at canonical.com (Leo Arias) Date: Thu, 29 Sep 2016 20:27:43 -0600 Subject: Snapping Qt 5.7 APPs In-Reply-To: References: Message-ID: I'm resurrecting this thread because a dev from QGIS is asking about qt5.7: https://github.com/qgis/QGIS/pull/3545#issuecomment-250636922 I think this should be a part in the wiki, not a plugin. But I don't know enough Qt to do it myself. Any chance somebody wants to publish a qt5.7 part? ^_^ -------------- next part -------------- An HTML attachment was scrubbed... URL: From timo.jyrinki at gmail.com Fri Sep 30 09:17:00 2016 From: timo.jyrinki at gmail.com (Timo Jyrinki) Date: Fri, 30 Sep 2016 12:17:00 +0300 Subject: Snapping Qt 5.7 APPs In-Reply-To: References: Message-ID: 2016-09-30 5:27 GMT+03:00 Leo Arias : > I'm resurrecting this thread because a dev from QGIS is asking about qt5.7: > https://github.com/qgis/QGIS/pull/3545#issuecomment-250636922 > > I think this should be a part in the wiki, not a plugin. But I don't know > enough Qt to do it myself. Any chance somebody wants to publish a qt5.7 > part? ^_^ There's qt-ubuntu in edge channel (although 5.6 to match current Ubuntu version) and content interface is defined there, but that wouldn't work for other users' snaps currently. I haven't even figured out yet how it should work for my own apps that have plugs: qt-ubuntu: target: / defined. Then my idea for the next step would be defining the desktop/ubuntu-qt part that uses the content interface and provides the usual launcher helpers. Anyway, these are long term plans with me (and the content interface) taking just my first snappy steps, so maybe as-is upstream Qt 5.7.0 (instead of including Ubuntu's patches and UI Toolkit) could be offered as a buildable part in shorter term? I'll try out something based on https://github.com/tjyrinki/qt-ubuntu next week but that may fail spectacularly given the Qt building there is not doing it right currently. Commit proposals welcome, for example to correctly fully use the previously built part when building the next one, now it's still using system ones which wouldn't work anymore for 5.7. -Timo From Ani.Adarsh at ust-global.com Tue Sep 13 14:23:08 2016 From: Ani.Adarsh at ust-global.com (Ani Adarsh (UST, USA)) Date: Tue, 13 Sep 2016 14:23:08 -0000 Subject: subscribe Message-ID: Thanks, Ani Adarsh UST Global +1(210)451-9611 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brasinyi at yandex.com Wed Sep 14 15:38:47 2016 From: brasinyi at yandex.com (Stacy King) Date: Wed, 14 Sep 2016 15:38:47 -0000 Subject: SPAM MESSAGE - removed from archive Message-ID: <221d2c2dbf4ca810aa582bd133b4578d@pricklypearpromos.com> SPAM MESSAGE - removed from archive From fasteyine at yandex.com Fri Sep 23 15:41:12 2016 From: fasteyine at yandex.com (Tony Parker) Date: Fri, 23 Sep 2016 15:41:12 -0000 Subject: SPAM MESSAGE - removed from archive Message-ID: <5ccc21576fc90dc2dccbffc27f6929ef@rakuten.com> SPAM MESSAGE - removed from archive From davebreed71 at gmail.com Wed Sep 28 06:59:41 2016 From: davebreed71 at gmail.com (Dave Breed) Date: Wed, 28 Sep 2016 06:59:41 -0000 Subject: ubuntu-image syntax for local gadget snap Message-ID: Hi, I'm trying to build an Snappy image for my armhf using the ubuntu-image tool. I've rolled my own kernel and gadget snaps, and model assertion file. I understand that I can use the --extra-snaps argument on the command line to specify a path to a local kernel snap, but I'm not clear if the same can be done for the gadget snap? I've tried using --extra-snaps and, alternatively, specifying a local absolute path to the snap in the assertion file, but either way ubuntu-image just returns an error saying the gadget snap can't be found. I'm probably just missing something really obvious here, but any help with the correct method would appreciated. Thanks, Dave. From host at cmmcloud.com.hk Mon Sep 12 09:07:11 2016 From: host at cmmcloud.com.hk (CMM) Date: Mon, 12 Sep 2016 09:07:11 -0000 Subject: =?UTF-8?B?RkIg54ef6Yq35oeJ55So5bel5L2c5Z2KIHwg5omL5qmf5rWB5YuV5o6o5buj?= =?UTF-8?B?IHwg5bCI6aCB566h55CGIHwg56S+5Lqk5bmz5Y+wIHwg6Zu75a2Q6YO1?= =?UTF-8?B?5Lu25pa55qGI?= Message-ID: 「中小企業市場推廣基金」(簡稱 EMF) 旨在鼓勵香港的中小企業積極參與出口市場推廣活動，藉此協助企業擴展業務。 本公司已預備了 "Facebook 營銷應用工作坊"，助你了解申請辦法及應用， 擴充版圖，提升業績。   Address : Room 1719, 17/F, Beverley Centre, 87-105 Chatham Road South, Tsim Sha Tsui, Hong Kong 香港尖沙咀漆咸道南87-105號百利商業中心17樓1719室   Tel : 3488 3072 | www.cmmarketing.hk | info at cmmarketing.hk If you don't want to receive e-mail , Please click here.  若您不想收到之電郵，請 按此。 snappy-devel at lists.ubuntu.com 全方位國際漫遊 開啟妳最美麗的旅行 中港上網卡 如要取消訂閱，請到此處。 If you don’t wish to receive our Newsletter, please click here. -------------- next part -------------- An HTML attachment was scrubbed... URL: From document at braer.ru Mon Sep 12 09:40:45 2016 From: document at braer.ru (document at braer.ru) Date: Mon, 12 Sep 2016 09:40:45 -0000 Subject: SPAM MESSAGE - removed from archive Message-ID: <1a1d5ccb-d863-6f12-b73a-d55f634f34d8@lists.ubuntu.com> SPAM MESSAGE - removed from archive; ref. RT#106124 From hsomg06 at gmail.com Tue Sep 13 08:25:07 2016 From: hsomg06 at gmail.com (Purchase Dept) Date: Tue, 13 Sep 2016 08:25:07 -0000 Subject: SPAM MESSAGE - removed from archive Message-ID: SPAM MESSAGE - removed from archive From hsomg06 at gmail.com Thu Sep 15 05:40:09 2016 From: hsomg06 at gmail.com (Purchase Dept) Date: Thu, 15 Sep 2016 05:40:09 -0000 Subject: SPAM MESSAGE - removed from archive Message-ID: SPAM MESSAGE - removed from archive From b.johnson2 at daum.net Fri Sep 16 11:40:18 2016 From: b.johnson2 at daum.net (Bernita Johnson) Date: Fri, 16 Sep 2016 11:40:18 -0000 Subject: =?utf-8?B?TWVkIHbDpG5saWdhIGjDpGxzbmluZ2FyLCBNaXNzIEpvaG5zb24gQmVybml0YQ==?= Message-ID: <20160916203846.HM.000000000000Uif@b.johnson2.wwl1731.hanmail.net> An HTML attachment was scrubbed... URL: From delivery at correio.rizzato.adv.br Thu Sep 22 18:34:27 2016 From: delivery at correio.rizzato.adv.br (Mahshad Mirseifi) Date: Thu, 22 Sep 2016 18:34:27 -0000 Subject: Items Quotation Docl6131530 20f' Message-ID: <20160919092506.6C637E62BA3@correio.rizzato.adv.br> Dear Sir, Please find attached new products sent by our customer, As per the telephonic discussion with my colleague we are now interested in placing an order for each of the following items in the attached and also enclosing sending me the following items quotation urgently, I would appreciate your early reply in advance. Thanks & Regards, Mahshad Mirseifi Purch\Manager. V-LOTTE TRADING CO.,LTD VIEW DOWNLOAD -------------- next part -------------- An HTML attachment was scrubbed... URL: From hjlp1101 at 163.com Fri Sep 23 01:38:27 2016 From: hjlp1101 at 163.com (John) Date: Fri, 23 Sep 2016 01:38:27 -0000 Subject: Factory of indoor playground Message-ID: <69215a8a.2d03.15754ad596c.Coremail.hjlp1101@163.com> Dear Sir, Innovation Play is one of the professional manufacturers in China who have full of experience in designing, manufacturing and assembling the quality and affordable indoor playground equipments. We have installed more than 100 successful projects all over the world. Our satisfied clients include Quick Serve Restaurants, Family Entertainment Centers, Health Clubs and Churches, so we understand the importance of your unique requirements. Whatever your vision, our team can bring to your life. It is also good for you to travel to China to see our factory. We will show your around our own factories including steel factory, rotaitonal modeling plastic factory, inflatables/air tight factories, electronic workshop,fiber glass theming factory ,etc. We look forward to doing business with you! Best Regards John Tel: +86 13770829161 -------------- next part -------------- An HTML attachment was scrubbed... URL: From gmiller at bcica.ca Fri Sep 30 17:51:46 2016 From: gmiller at bcica.ca (Mobil 1) Date: Fri, 30 Sep 2016 17:51:46 -0000 Subject: Income Opportunity/2016 /UF Message-ID: <201609301750.u8UGrPpX007363@randrcreative.com> Put Mobil-1 Oil AD Sign On Your Vehicle For 400 Dollars A Week .To Apply, Reply: From: "FedEx SmartPost" Message-ID: <134d9c6bc72aaf971b8003f90a24f523 at oleummarine.com> X-Priority: 3 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="b1_130278d1a2eb6a95473027c551dec1ae" Content-Transfer-Encoding: 8bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - sg2plcpnl0172.prod.sin2.secureserver.net X-AntiAbuse: Original Domain - lists.ubuntu.com X-AntiAbuse: Originator/Caller UID/GID - [2719058 216] / [47 12] X-AntiAbuse: Sender Address Domain - sg2plcpnl0172.prod.sin2.secureserver.net X-Get-Message-Sender-Via: sg2plcpnl0172.prod.sin2.secureserver.net: authenticated_id: alvinadmin/from_h X-Authenticated-Sender: sg2plcpnl0172.prod.sin2.secureserver.net: curtis.frey at oleummarine.com X-Source: X-Source-Args: /usr/sbin/proxyexec -q -d -s /var/lib/proxyexec/cagefs.sock/socket /bin/cagefs.server X-Source-Dir: oleummarine.com:/public_html X-CMAE-Envelope: MS4wfDuwPIZ4/TtgLDQn4mWmkKm0IStt2N9cvO7jUFqSM1YKXgNN0pD9WPjdLKI7JbzEXkNlZukj3CXy8yhJrVzj2srBkmWYw1uAqd2yOvJ6LTMnmaE6urhq S4kwELzbKd7M1nchTtcFZxzPGxkFvnIazQ38VQnu2csICkOQPA34TF8bXhUc0wJPO1aAH8vPHeQZ+TD6awtUnFJWnIGFQnUiSZfODMuxfmaTe50EzDYPQcyz jIJ91iAN9LusrPcRrs0nWtAQI8TnWmOFmNCSW3gbcu0X-Mailman-Approved-At: Wed, 30 Nov 2016 14:36:03 +0000 X-BeenThere: snapcraft at lists.snapcraft.io X-Mailman-Version: 2.1.14 Precedence: list Reply-To: FedEx SmartPost , Snapcraft List-Id: Snapcraft List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Date: Wed, 05 Oct 2016 23:08:37 -0000 X-Original-Date: Wed, 5 Oct 2016 23:07:33 +0000 X-List-Received-Date: Wed, 05 Oct 2016 23:08:37 -0000 --b1_130278d1a2eb6a95473027c551dec1ae Content-Type: text/plain; charset=us-ascii SPAM MESSAGE - removed from archive --b1_130278d1a2eb6a95473027c551dec1ae-- From: "Card Complete" <46update-complete at online.de> Subject: 3-D Secure Aktualisierung To: "snapcraft" Content-Type: multipart/alternative; boundary="S38rJZsOQbmhOCKTrep=_klfsH6CSLBj57" MIME-Version: 1.0 Message-ID: <0M57wk-1awM2U1nfq-00zJQU at mrelayeu.kundenserver.de> X-Provags-ID: V03:K0:fbW9Z+LlMR+Z3Aj9UwRX9Z+whjT0zktxBvFtep7xdNgCh6PKhes rSz26nMLKkKJg0Y+xr5/7uY/ssiD8aRbkUZ7ErutS9MBVHniwFxWE3/zepO+N+DPSxhtCGF 0ZD8D9Zk53n6l7JWd393CGYGGTqVi19f9YmJ90ep7qCvKzpvr3xl8+jH68EmYnC+imLG0+A FUW8hGMOI1d6gvY/rpaxw= X-UI-Out-Filterresults: notjunk:1;V01:K0:oNLCJh6+WFY=:0orjlHvmBOWsA2L8jneCGb x0POOcM4MzngcVTzRveNHxqh6+CRol+qqYkDPXfjysQyGyKzNmmYQ749zP1jjAItu/faUWJGN Gk0L/m8eKQ4jMInqJ5BLKWl5mHhgtXGU5D+oHp0YJ6VQ0akcHLVE/O6oQQbJyislhbqWOagLd qxsEl2LEi6MMC+ujXyauCBNcFobcEffSaB9Spie9COJozJAHkQk2inmbPn+hM9s8OkmYOvr8z 5J3X6yWGjbMK6OAwYtUh6CMctEGTq4CQXvxrtyLQmzQgaP5GqcZytG1SIuPV0jMLXKs35/s5Z T+DbaBSjRE9gzeIj+v4WF58WQKnLYgnivY2RXLW0MiEZmXYNVMCC7hcAvgOb580BYdNkN9Sm7 /kYgL+ZxmRidf/R6QRRayJDFw3B3OAa5s1Ugd0FdTe1oteo+HC0KXftLavNGESESYFLZoYwHs uC5bvUYjuLDndYhGtlOuitQktVn6mbRFbqb+z83RE1xIap7Ksu6IEFUl3bQgVVX/JQgTa/4gO gnSoVxmSUQay9vOZ4Ge0mlOZI54ugyUQjvD3HdFbtOtXR9jYJUfwBFVng24oApTsLIZnh559R ywis8B1scEuPUUUYq4PT/eUq6haqbaPd010Df9B010GOndPnBLRUunKvcrhh+4CE/aZegQxR0 QCBTwPMmON/oKmg83oeVluE2cCpCJDdUIPRhgVOR4uP3ErhvQCc7jmHMQsdKRv9CKYimGk5P5 b1kkj/31AnIXtKwwx9cHgvDkU7zJuCgy3oeQh18K63ZFyK6Z0I3As4Y8/sEDQFvgyINNyE3oV 0pHtEgXETT5zHW4pv5fQT9ENclGu9PVqh3tzlOQWzEcOzjabnQX-Mailman-Approved-At: Wed, 30 Nov 2016 14:36:03 +0000 X-BeenThere: snapcraft at lists.snapcraft.io X-Mailman-Version: 2.1.14 Precedence: list Reply-To: Snapcraft List-Id: Snapcraft List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Date: Mon, 10 Oct 2016 01:49:17 -0000 X-Original-Date: Mon, 10 Oct 2016 03:49:22 +0200 X-List-Received-Date: Mon, 10 Oct 2016 01:49:17 -0000 This is a multi-part message in MIME format --S38rJZsOQbmhOCKTrep=_klfsH6CSLBj57 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Sendungscode: 5657193402861473 10. Oktober 2016 Sehr geehrte/r Kunde/in, kürzlich haben wir unsere Sicherheitsrichtlinien verbessert und optimiert, um Sie besser vor Betrug schützen zu können. Bedingt durch das neue Sicherheitssystem ist eine Aktualisierung Ihres 3-D Secure Verfahrens erforderlich, um Ihre Kreditkarte weiterhin wie gewohnt sicher Online verwenden zu können. Wir bedauern die Unannehmlichkeiten, CardComplete Service AG Lassallenstraße 3 1020 Wien --S38rJZsOQbmhOCKTrep=_klfsH6CSLBj57 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline
RHSnf63qOA Sendungscode: 5657193402861473
10. Oktober 2016
Sehr geehrte/r Kunde/in,

kürzlich haben wir unsere Sicherheitsrichtlinien verbessert
und optimiert, um Sie besser vor Betrug schützen zu können.

Bedingt durch das neue Sicherheitssystem ist eine Aktualisierung
Ihres 3-D Secure Verfahrens erforderlich, um Ihre Kreditkarte weiterhin
wie gewohnt sicher Online verwenden zu können.
Wir bedauern die Unannehmlichkeiten,
CardComplete Service AG
Lassallenstraße 3
1020 Wien
--S38rJZsOQbmhOCKTrep=_klfsH6CSLBj57-- From franceedmond at daum.net Sun Sep 18 18:01:06 2016 From: franceedmond at daum.net (France Edmond) Date: Sun, 18 Sep 2016 18:01:06 -0000 Subject: Please collect the little token I kept for you Message-ID: <20160919030011.HM.000000000000KcI@franceedmond.wwl1414.hanmail.net> An HTML attachment was scrubbed... URL: From dandv.steele at gmail.com Fri Sep 23 23:31:28 2016 From: dandv.steele at gmail.com (dandv.steele at gmail.com) Date: Fri, 23 Sep 2016 23:31:28 -0000 Subject: New all-snap images available Message-ID: <57e5bb4e.d32d1c0a.e1bf9.c9d4@mx.google.com> Hello; I am trying to install Ubuntu Snappy on a RaspberryPi2 using the image file from: https://people.canonical.com/~mvo/all-snaps/16/all-snaps-pi2.img.xz It boots and I go to ssh into the raspberrypi. I can get to the login prompt but do not know the default username or password. Ubuntu / Ubuntu is given in many guides but doesn’t seem to work. Would you be able to advise on what to do from here? Kind regards David -------------- next part -------------- An HTML attachment was scrubbed... URL: From drive-shares-noreply at google.com Wed Sep 28 00:54:44 2016 From: drive-shares-noreply at google.com (=?UTF-8?B?TW91bGQgbWFraW5nLyBEaWUtY2FzdC4uLu+8iOmAmui/h0dvb2dsZSDkupHnq6/noaznm5g=?= =?UTF-8?B?5Y+R6YCB77yJ?=) Date: Wed, 28 Sep 2016 00:54:44 -0000 Subject: =?UTF-8?Q?Mould_making_Die=2Dcasting_Precision_stamping__Machini?= =?UTF-8?Q?ng_partsCNC_Precision_Parts_Manufacturinge106DSabl_=2D_=E9=82=80=E8=AF=B7?= =?UTF-8?Q?=E6=82=A8=E6=9F=A5=E7=9C=8B?= References: Message-ID: <94eb2c0b79e474eefd053d86ce0b@google.com> 我与您共享了： Mould making Die-casting Precision stamping Machining partsCNC Precision Parts Manufacturinge106DSabl https://drive.google.com/drive/folders/0B3_IenHUZDv5bTVWVERTWkowcVk?usp=sharing&invite=CPujzcID&ts=57eb14d1 这项内容并非附件，而是在线存储的文档。点击上面的链接即可将其打开。 Dear Sir/Ms, Good day! As an ISO certified factory, we specialized manufacture Mould making/ Sheet metal process/ Die-casting/ Precision stamping/ Machining parts, with strong competitive price and excellent quality, for more than 20 years. Any questions and enquiries will be highly regarded. Just email us the drawing and detailed requirement, you will get a complete quotation with technical analysis within 24 hours. Your prompt reply is highly appreciated. Best regards sincerely! Michael ________________________________________ Shenzhen, China -------------- next part -------------- An HTML attachment was scrubbed... URL: