kernel-packages team mailing list archive

[Frederik Bosch <f.bosch@xxxxxxxxx>]

Public bug reported:

Because "ubuntu-bug linux" not works because I am on kernel 3.19.0-17, I
create this report manually.

Summary of problem: we create backups with HyperV of 4 linux virtual
machines. There is 1 CentOS machine (XFS) and 3 Ubuntu 14.04 machines
(EXT4). Problems occur on the Ubuntu machine only and have occurred on
multiple hypervisors (physical machines). During the start or at the end
of the creation of a snapshot, the filesystem switches to read-only
mode. This happens at completely random moments: at least we are not
able to detect a pattern. What we find weird is that the CentOS machine
with XFS is not having any problems.

What actions did we undertake to get rid of problem? We tried multiple
backup strategies with HyperV, including snapshots and exports. When
this was not successful, we were adviced by the #ubuntu-kernel chat
channel to upgrade our kernel to 3.19.0-17 (coming from 3.13.0-52). The
reasons that this could solve the problem is that hyperv backups require
freeze and thaw methods (https://msdn.microsoft.com/en-
us/library/aa384589%28v=vs.85%29.aspx). Unfortunately the problem did
not go away.

The message we get from dmesg when the filesystem goes read-only is as
follows on the 3.19.0-17 kernel:

[85630.239889] blk_update_request: I/O error, dev sda, sector 201697672
[85630.239931] Aborting journal on device sda1-8.
[85630.253386] EXT4-fs error (device sda1): ext4_journal_check_start:56: Detected aborted journal
[85630.253428] EXT4-fs (sda1): Remounting filesystem read-only

While it was as follows on the 3.13.0-52 kernel (different machine):

[66392.076569] end_request: I/O error, dev sda, sector 65127256
[66392.076610] Aborting journal on device sda5-8.
[66392.280052] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal
[66392.280090] EXT4-fs (sda5): Remounting filesystem read-only
[66392.482903] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal

System details:
- Version signature: Ubuntu 3.19.0-17.17~14.04.1-generic 3.19.6
- lspci.log: http://paste.ubuntu.com/11242312/
- boot.log: http://paste.ubuntu.com/11242303/

** Affects: linux (Ubuntu)
     Importance: Undecided
         Status: Incomplete

** Description changed:

  Filesystem goes read-only while creating VSS snapshot on HyperV
  
  Because "ubuntu-bug linux" not works because I am on kernel 3.19.0-17, I
  create this report manually.
  
  Summary of problem: we create VSS snapshots with HyperV of 4 linux
  virtual machines. There is 1 CentOS machine (XFS) and 3 Ubuntu 14.04
  machines (EXT4). Problems occur on the Ubuntu machine only and have
  occurred on multiple hypervisors (physical machines). During the start
  or at the end of the creation of a snapshot, the filesystem switches to
  read-only mode. This happens at completely random moments: at least we
  are not able to detect a pattern. What we find weird is that the CentOS
  machine with XFS is not having any problems.
  
  What actions did we undertake to get rid of problem? We tried multiple
  backup strategies with HyperV. When this was not successful, we were
  adviced by the #ubuntu-kernel chat channel to upgrade our kernel to
  3.19.0-17 (coming from 3.13.0-52). The reasons that this could solve the
  problem is that hyperv backups require freeze and thaw methods
  (https://msdn.microsoft.com/en-us/library/aa384589%28v=vs.85%29.aspx).
  Unfortunately the problem did not go away.
  
  The message we get from dmesg when the filesystem goes read-only is as
  follows on the 3.19.0-17 kernel:
  
  [85630.239889] blk_update_request: I/O error, dev sda, sector 201697672
  [85630.239931] Aborting journal on device sda1-8.
  [85630.253386] EXT4-fs error (device sda1): ext4_journal_check_start:56: Detected aborted journal
  [85630.253428] EXT4-fs (sda1): Remounting filesystem read-only
  
- While it was as follows on the 3.13.0-52 kernel:
+ While it was as follows on the 3.13.0-52 kernel (different machine):
  
  [66392.076569] end_request: I/O error, dev sda, sector 65127256
  [66392.076610] Aborting journal on device sda5-8.
  [66392.280052] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal
  [66392.280090] EXT4-fs (sda5): Remounting filesystem read-only
  [66392.482903] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal
  
  System details:
  - Version signature: Ubuntu 3.19.0-17.17~14.04.1-generic 3.19.6
  - lspci.log: http://paste.ubuntu.com/11242312/
  - boot.log: http://paste.ubuntu.com/11242303/

** Summary changed:

- Filesystem goes read-only while creating VSS snapshot on HyperV
+ Filesystem goes read-only while creating backups on HyperV

** Description changed:

- Filesystem goes read-only while creating VSS snapshot on HyperV
- 
  Because "ubuntu-bug linux" not works because I am on kernel 3.19.0-17, I
  create this report manually.
  
- Summary of problem: we create VSS snapshots with HyperV of 4 linux
- virtual machines. There is 1 CentOS machine (XFS) and 3 Ubuntu 14.04
- machines (EXT4). Problems occur on the Ubuntu machine only and have
- occurred on multiple hypervisors (physical machines). During the start
- or at the end of the creation of a snapshot, the filesystem switches to
- read-only mode. This happens at completely random moments: at least we
- are not able to detect a pattern. What we find weird is that the CentOS
- machine with XFS is not having any problems.
+ Summary of problem: we create backups with HyperV of 4 linux virtual
+ machines. There is 1 CentOS machine (XFS) and 3 Ubuntu 14.04 machines
+ (EXT4). Problems occur on the Ubuntu machine only and have occurred on
+ multiple hypervisors (physical machines). During the start or at the end
+ of the creation of a snapshot, the filesystem switches to read-only
+ mode. This happens at completely random moments: at least we are not
+ able to detect a pattern. What we find weird is that the CentOS machine
+ with XFS is not having any problems.
  
  What actions did we undertake to get rid of problem? We tried multiple
- backup strategies with HyperV. When this was not successful, we were
- adviced by the #ubuntu-kernel chat channel to upgrade our kernel to
- 3.19.0-17 (coming from 3.13.0-52). The reasons that this could solve the
- problem is that hyperv backups require freeze and thaw methods
- (https://msdn.microsoft.com/en-us/library/aa384589%28v=vs.85%29.aspx).
- Unfortunately the problem did not go away.
+ backup strategies with HyperV, including snapshots and exports. When
+ this was not successful, we were adviced by the #ubuntu-kernel chat
+ channel to upgrade our kernel to 3.19.0-17 (coming from 3.13.0-52). The
+ reasons that this could solve the problem is that hyperv backups require
+ freeze and thaw methods (https://msdn.microsoft.com/en-
+ us/library/aa384589%28v=vs.85%29.aspx). Unfortunately the problem did
+ not go away.
  
  The message we get from dmesg when the filesystem goes read-only is as
  follows on the 3.19.0-17 kernel:
  
  [85630.239889] blk_update_request: I/O error, dev sda, sector 201697672
  [85630.239931] Aborting journal on device sda1-8.
  [85630.253386] EXT4-fs error (device sda1): ext4_journal_check_start:56: Detected aborted journal
  [85630.253428] EXT4-fs (sda1): Remounting filesystem read-only
  
  While it was as follows on the 3.13.0-52 kernel (different machine):
  
  [66392.076569] end_request: I/O error, dev sda, sector 65127256
  [66392.076610] Aborting journal on device sda5-8.
  [66392.280052] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal
  [66392.280090] EXT4-fs (sda5): Remounting filesystem read-only
  [66392.482903] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal
  
  System details:
  - Version signature: Ubuntu 3.19.0-17.17~14.04.1-generic 3.19.6
  - lspci.log: http://paste.ubuntu.com/11242312/
  - boot.log: http://paste.ubuntu.com/11242303/

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1456985

Title:
  Filesystem goes read-only while creating backups on HyperV

Status in linux package in Ubuntu:
  Incomplete

Bug description:
  Because "ubuntu-bug linux" not works because I am on kernel 3.19.0-17,
  I create this report manually.

  Summary of problem: we create backups with HyperV of 4 linux virtual
  machines. There is 1 CentOS machine (XFS) and 3 Ubuntu 14.04 machines
  (EXT4). Problems occur on the Ubuntu machine only and have occurred on
  multiple hypervisors (physical machines). During the start or at the
  end of the creation of a snapshot, the filesystem switches to read-
  only mode. This happens at completely random moments: at least we are
  not able to detect a pattern. What we find weird is that the CentOS
  machine with XFS is not having any problems.

  What actions did we undertake to get rid of problem? We tried multiple
  backup strategies with HyperV, including snapshots and exports. When
  this was not successful, we were adviced by the #ubuntu-kernel chat
  channel to upgrade our kernel to 3.19.0-17 (coming from 3.13.0-52).
  The reasons that this could solve the problem is that hyperv backups
  require freeze and thaw methods (https://msdn.microsoft.com/en-
  us/library/aa384589%28v=vs.85%29.aspx). Unfortunately the problem did
  not go away.

  The message we get from dmesg when the filesystem goes read-only is as
  follows on the 3.19.0-17 kernel:

  [85630.239889] blk_update_request: I/O error, dev sda, sector 201697672
  [85630.239931] Aborting journal on device sda1-8.
  [85630.253386] EXT4-fs error (device sda1): ext4_journal_check_start:56: Detected aborted journal
  [85630.253428] EXT4-fs (sda1): Remounting filesystem read-only

  While it was as follows on the 3.13.0-52 kernel (different machine):

  [66392.076569] end_request: I/O error, dev sda, sector 65127256
  [66392.076610] Aborting journal on device sda5-8.
  [66392.280052] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal
  [66392.280090] EXT4-fs (sda5): Remounting filesystem read-only
  [66392.482903] EXT4-fs error (device sda5): ext4_journal_check_start:56: Detected aborted journal

  System details:
  - Version signature: Ubuntu 3.19.0-17.17~14.04.1-generic 3.19.6
  - lspci.log: http://paste.ubuntu.com/11242312/
  - boot.log: http://paste.ubuntu.com/11242303/

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1456985/+subscriptions