[Bug 1625201] [NEW] "lvconvert --repair" destroys thin pool LV

marek 1625201 at bugs.launchpad.net
Mon Sep 19 13:55:52 UTC 2016 

Public bug reported:

Hi
  If pool meta is 16GB , lvconvert  --repair will destroy logical volumes.
After rapair logical volumes inside the pool can not be activated again. Steps to reproduce:


truncate  --size 100GiB /tmp/test
losetup /dev/loop1 /tmp/test
pvcreate  /dev/loop1
vgcreate vg02 /dev/loop1
lvcreate -n pool0meta -L 16G vg02
lvcreate -n pool0 -L 16GiB vg02
lvcreate -n data -V 1GiB --thinpool vg02/pool0
vgchange  -a n  vg02
lvconvert  --repair vg02/pool0


# vgchange  -a y  vg02
  device-mapper: resume ioctl on (252:3) failed: Invalid argument
  Unable to resume vg02-pool0-tpool (252:3)
  device-mapper: resume ioctl on (252:3) failed: Invalid argument
  Unable to resume vg02-pool0-tpool (252:3)
  1 logical volume(s) in volume group "vg02" now active


Syslog :


[6039585.407800] device-mapper: thin: 252:2: metadata device (4145152 blocks) too small: expected 4161600
[6039585.407859] device-mapper: table: 252:2: thin-pool: preresume failed, error = -22
[6039586.535215] device-mapper: thin: 252:2: metadata device (4145152 blocks) too small: expected 4161600
[6039586.535258] device-mapper: table: 252:2: thin-pool: preresume failed, error = -22
[6039613.657170] device-mapper: thin: 252:3: metadata device (4145152 blocks) too small: expected 4161600
[6039613.657212] device-mapper: table: 252:3: thin-pool: preresume failed, error = -22
[6039614.716005] device-mapper: thin: 252:3: metadata device (4145152 blocks) too small: expected 4161600
[6039614.716047] device-mapper: table: 252:3: thin-pool: preresume failed, error = -22


Workaround for this is to shirk pool meta ( only when it is possible:
You have free space in VG, and when meta will fit ) :

thin_dump /dev/mapper/vg02-pool0_tmeta > lvm_meta_dum
lvcreate -n pool0meta2 -L 12G vg02
thin_restore -i lvm_meta_dump -o /dev/mapper/vg02-pool0meta2
lvconvert --thinpool vg02/pool0 --poolmetadata vg02/pool0meta2


-- 
   Kind Regards
     Marek Grzybowski

ProblemType: Bug
DistroRelease: Ubuntu 16.04
Package: lvm2 2.02.133-1ubuntu10
ProcVersionSignature: Ubuntu 4.4.0-28.47-generic 4.4.13
Uname: Linux 4.4.0-28-generic x86_64
ApportVersion: 2.20.1-0ubuntu2.1
Architecture: amd64
Date: Mon Sep 19 13:29:54 2016
ProcEnviron:
 TERM=xterm
 PATH=(custom, no user)
 LANG=C.UTF-8
 LC_MESSAGES=POSIX
 SHELL=/bin/bash
SourcePackage: lvm2
UpgradeStatus: No upgrade log present (probably fresh install)

** Affects: lvm2 (Ubuntu)
     Importance: Undecided
         Status: New


** Tags: amd64 apport-bug xenial

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to lvm2 in Ubuntu.
https://bugs.launchpad.net/bugs/1625201

Title:
  "lvconvert  --repair" destroys thin pool LV

Status in lvm2 package in Ubuntu:
  New

Bug description:
  Hi
    If pool meta is 16GB , lvconvert  --repair will destroy logical volumes.
  After rapair logical volumes inside the pool can not be activated again. Steps to reproduce:

  
  truncate  --size 100GiB /tmp/test
  losetup /dev/loop1 /tmp/test
  pvcreate  /dev/loop1
  vgcreate vg02 /dev/loop1
  lvcreate -n pool0meta -L 16G vg02
  lvcreate -n pool0 -L 16GiB vg02
  lvcreate -n data -V 1GiB --thinpool vg02/pool0
  vgchange  -a n  vg02
  lvconvert  --repair vg02/pool0


  
  # vgchange  -a y  vg02
    device-mapper: resume ioctl on (252:3) failed: Invalid argument
    Unable to resume vg02-pool0-tpool (252:3)
    device-mapper: resume ioctl on (252:3) failed: Invalid argument
    Unable to resume vg02-pool0-tpool (252:3)
    1 logical volume(s) in volume group "vg02" now active

  
  Syslog :

  
  [6039585.407800] device-mapper: thin: 252:2: metadata device (4145152 blocks) too small: expected 4161600
  [6039585.407859] device-mapper: table: 252:2: thin-pool: preresume failed, error = -22
  [6039586.535215] device-mapper: thin: 252:2: metadata device (4145152 blocks) too small: expected 4161600
  [6039586.535258] device-mapper: table: 252:2: thin-pool: preresume failed, error = -22
  [6039613.657170] device-mapper: thin: 252:3: metadata device (4145152 blocks) too small: expected 4161600
  [6039613.657212] device-mapper: table: 252:3: thin-pool: preresume failed, error = -22
  [6039614.716005] device-mapper: thin: 252:3: metadata device (4145152 blocks) too small: expected 4161600
  [6039614.716047] device-mapper: table: 252:3: thin-pool: preresume failed, error = -22


  Workaround for this is to shirk pool meta ( only when it is possible:
  You have free space in VG, and when meta will fit ) :

  thin_dump /dev/mapper/vg02-pool0_tmeta > lvm_meta_dum
  lvcreate -n pool0meta2 -L 12G vg02
  thin_restore -i lvm_meta_dump -o /dev/mapper/vg02-pool0meta2
  lvconvert --thinpool vg02/pool0 --poolmetadata vg02/pool0meta2


  -- 
     Kind Regards
       Marek Grzybowski

  ProblemType: Bug
  DistroRelease: Ubuntu 16.04
  Package: lvm2 2.02.133-1ubuntu10
  ProcVersionSignature: Ubuntu 4.4.0-28.47-generic 4.4.13
  Uname: Linux 4.4.0-28-generic x86_64
  ApportVersion: 2.20.1-0ubuntu2.1
  Architecture: amd64
  Date: Mon Sep 19 13:29:54 2016
  ProcEnviron:
   TERM=xterm
   PATH=(custom, no user)
   LANG=C.UTF-8
   LC_MESSAGES=POSIX
   SHELL=/bin/bash
  SourcePackage: lvm2
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/lvm2/+bug/1625201/+subscriptions

More information about the foundations-bugs mailing list