[LU-9865] Pacemaker fails to start target due to zpool I/O error Created: 11/Aug/17  Updated: 11/Aug/17

Status: Open
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Minor
Reporter: Tom Nabarro (Inactive) Assignee: WC Triage
Resolution: Unresolved Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

https://github.com/intel-hpdd/intel-manager-for-lustre/issues/205

IML attempts to start a Lustre target on a zfs-backed device with pacemaker but the action fails with the following description (in messages log):

Aug 10 08:51:14 lotus-56vm15.lotus.hpdd.lab.intel.com lrmd[7679]:  notice: testfs-OST0001_67da1f_start_0:1930:stderr [ Error importing pool: 'Error (1) running 'zpool import -f zfs_pool_scsi0QEMU_QEMU_HARDDISK_disk13': '#011Recovery is possible, but will result in some data loss. ]
Aug 10 08:51:14 lotus-56vm15.lotus.hpdd.lab.intel.com lrmd[7679]:  notice: testfs-OST0001_67da1f_start_0:1930:stderr [ #011Returning the pool to its state as of Thu Aug 10 08:51:08 2017 ]
Aug 10 08:51:14 lotus-56vm15.lotus.hpdd.lab.intel.com lrmd[7679]:  notice: testfs-OST0001_67da1f_start_0:1930:stderr [ #011should correct the problem.  Recovery can be attempted ]
Aug 10 08:51:14 lotus-56vm15.lotus.hpdd.lab.intel.com lrmd[7679]:  notice: testfs-OST0001_67da1f_start_0:1930:stderr [ #011by executing 'zpool import -F zfs_pool_scsi0QEMU_QEMU_HARDDISK_disk13'.  A scrub of the pool ]
Aug 10 08:51:14 lotus-56vm15.lotus.hpdd.lab.intel.com lrmd[7679]:  notice: testfs-OST0001_67da1f_start_0:1930:stderr [ #011is strongly recommended after recovery. ]
Aug 10 08:51:14 lotus-56vm15.lotus.hpdd.lab.intel.com lrmd[7679]:  notice: testfs-OST0001_67da1f_start_0:1930:stderr [ ' 'cannot import 'zfs_pool_scsi0QEMU_QEMU_HARDDISK_disk13': I/O error ]

What is the suggestion for handling this kind of scenario?



 Comments   
Comment by Brian Murrell (Inactive) [ 11/Aug/17 ]

Just to extract the pertinent messages from the decoration... The command executed was:

zpool import -f zfs_pool_scsi0QEMU_QEMU_HARDDISK_disk13

and the error it returned was:

Recovery is possible, but will result in some data loss.
Returning the pool to its state as of Thu Aug 10 08:51:08 2017
should correct the problem. Recovery can be attempted
by executing 'zpool import -F zfs_pool_scsi0QEMU_QEMU_HARDDISK_disk13'. A scrub of the pool
is strongly recommended after recovery.

Any idea what would cause this error?

Generated at Sat Feb 10 02:30:00 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.