jujubigdata 4.6.0 is released
Cory Johns
cory.johns at canonical.com
Mon Nov 23 17:56:49 UTC 2015
I meant to follow up on this to expand on the point about working with
layered charms. The previous big data charms were using a fork of
charmhelpers that included some refactors to the old "services framework"
that were never merged into upstream. This library used some of those
changes, but they're not really applicable to the new layered charm
approach.
So, with this update, you should be able to create layered big data
charms. I have an example of the Spark charm rewritten using layers that
can be used as an example: https://github.com/johnsca/layer-apache-spark
We also plan eventually to refactor this library to break out the utils
from the Hadoop bits and move it to live under the charms.X package
namespace.
This is still very much in progress, though, so feedback is welcome.
On Fri, Nov 20, 2015 at 2:42 PM, Kevin Monroe <kevin.monroe at canonical.com>
wrote:
> The jujubigdata 4.6.0 release is now generally available:
>
> https://pypi.python.org/pypi/jujubigdata/4.6.0
>
> All charms referencing "pypi: jujubigdata>=4.0.0,<5.0.0" in resources.yaml
> will pick up the most recent 4.x.x release during deployment -- this
> includes all currently promulgated apache-hadoop charms as well as those
> currently in the ~bigdata-dev development namespace.
>
> Notable fixes since our last minor release include: better proxy support,
> LZO configuration in core hadoop charms, and support for charms written
> using the layered charming approach. Full commit log since v4.5.0:
>
> - 07191bb (HEAD, origin/master, origin/HEAD, master) lint fix, bump
> VERSION for minor release
> - c5f8959 dont set secondary attrs if we dont have a ready secondary unit
> - 529e0d9 Merge pull request #16 from johnsca/master
> - 6da2a5f Fixed empty slaves
> - 9b363c0 Merge pull request #15 from andrewdmcleod/lzo_config
> - c92c34d Update handlers.py
> - 89b8301 Disable NodeManager vmem check
> - 46cf508 bump VERSION for bugfix release
> - 665da2f fix another environment split (missed in previous commit)
> - 45811ff bump VERSION for bugfix release
> - b127c74 Handle /etc/environment values that contain '='
> - ff5f155 Store java release version
> - 4c545a1 Merge pull request #11 from andrewdmcleod/master
> - 24e712a Added Travis CI badge to README
> - 3f1d095 Moved cpu_arch helper
> - 56bc87f Hack around reliance on forked charmhelpers until layers
> transition is complete
> - 93c6b86 Switched to tox and added support for Travis CI
> - 2a41cae Make relations import conditional for layers migration
> - 12c3c16 bump VERSION for bugfix release
> - de1eabb rework proxy passing to account for all envars that end with
> _proxy
> - e5dd26b Merge branch 'lzo_config' of
> https://github.com/andrewdmcleod/jujubigdata into lzo_config
> - 9d52f68 core-site.xml changes for lzo
> - 0768f83 more indent fixes
> - 61fa205 indent fix
> - 82fe62d add lzo config to core-site.xml if lzo is installed
> - ad1d560 include proxy config when returning /etc/environment dict
> - 7cd6a51 update README links
>
> Questions/concerns with this release? Please open an issue here:
>
> https://github.com/juju-solutions/jujubigdata/issues
>
> Thanks!
>
> --
> Bigdata mailing list
> Bigdata at lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/bigdata
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/bigdata/attachments/20151123/452dcff5/attachment.html>
More information about the Bigdata
mailing list