Beau,

 

Looks like you have upgraded to CentOS 7.3. Now you have to update the vdsm package to 4.17.35.

 

 

From: <users-bounces@ovirt.org> on behalf of Beau Sapach <bsapach@ualberta.ca>
Date: Wednesday 18 January 2017 at 23:56
To: "users@ovirt.org" <users@ovirt.org>
Subject: [ovirt-users] Select As SPM Fails

 

Hello everyone,

 

I'm about to start digging through the mailing list archives in search of a solution but thought I would post to the list as well.  I'm running oVirt 3.6 on a 2 node CentOS7 cluster backed by fiber channel storage and with a separate engine VM running outside of the cluster (NOT  hosted-engine).

 

When I try to move the SPM role from one node to the other I get the following in the web interface:

 

 

When I look into /var/log/ovirt-engine/engine.log I see the following:

 

2017-01-18 13:35:09,332 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetAllTasksStatusesVDSCommand] (default task-26) [6990cfca] Failed in 'HSMGetAllTasksStatusesVDS' method

2017-01-18 13:35:09,340 ERROR [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (default task-26) [6990cfca] Correlation ID: null, Call Stack: null, Custom Event ID: -1, Message: VDSM v6 command failed: Logical Volume extend failed

 

When I look at the task list on the host currently holding the SPM role (in this case 'v6'), using: vdsClient -s 0 getAllTasks, I see a long list like this:

 

dc75d3e7-cea7-449b-9a04-76fd8ef0f82b :

         verb = downloadImageFromStream

         code = 554

         state = recovered

         tag = spm

         result =

         message = Logical Volume extend failed

         id = dc75d3e7-cea7-449b-9a04-76fd8ef0f82b

 

When I look at /var/log/vdsm/vdsm.log on the host in question (v6) I see messages like this:

 

'531dd533-22b1-47a0-aae8-76c1dd7d9a56': {'code': 554, 'tag': u'spm', 'state': 'recovered', 'verb': 'downloadImageFromStreaam', 'result': '', 'message': 'Logical Volume extend failed', 'id': '531dd533-22b1-47a0-aae8-76c1dd7d9a56'}

 

As well as the error from the attempted extend of the logical volume:

 

e980df5f-d068-4c84-8aa7-9ce792690562::ERROR::2017-01-18 13:24:50,710::task::866::Storage.TaskManager.Task::(_setError) Task=`e980df5f-d068-4c84-8aa7-9ce792690562`::Unexpected error

Traceback (most recent call last):

  File "/usr/share/vdsm/storage/task.py", line 873, in _run

    return fn(*args, **kargs)

  File "/usr/share/vdsm/storage/task.py", line 332, in run

    return self.cmd(*self.argslist, **self.argsdict)

  File "/usr/share/vdsm/storage/securable.py", line 77, in wrapper

    return method(self, *args, **kwargs)

  File "/usr/share/vdsm/storage/sp.py", line 1776, in downloadImageFromStream

    .copyToImage(methodArgs, sdUUID, imgUUID, volUUID)

  File "/usr/share/vdsm/storage/image.py", line 1373, in copyToImage

    / volume.BLOCK_SIZE)

  File "/usr/share/vdsm/storage/blockVolume.py", line 310, in extend

    lvm.extendLV(self.sdUUID, self.volUUID, sizemb)

  File "/usr/share/vdsm/storage/lvm.py", line 1179, in extendLV

    _resizeLV("lvextend", vgName, lvName, size)

  File "/usr/share/vdsm/storage/lvm.py", line 1175, in _resizeLV

    raise se.LogicalVolumeExtendError(vgName, lvName, "%sM" % (size, ))

LogicalVolumeExtendError:

Logical Volume extend failed: 'vgname=ae05947f-875c-4507-ad51-62b0d35ef567 lvname=caaef597-eddd-4c24-8df2-a61f35f744f8 newsize=1M'

e980df5f-d068-4c84-8aa7-9ce792690562::DEBUG::2017-01-18 13:24:50,711::task::885::Storage.TaskManager.Task::(_run) Task=`e980df5f-d068-4c84-8aa7-9ce792690562`::Task._run: e980df5f-d068-4c84-8aa7-9ce792690562 () {} failed - stopping task

 

The logical volume in question is an OVF_STORE disk that lives on one of the fiber channel backed LUNs.  If I run:

 

vdsClient -s 0 ClearTask TASK-UUID-HERE

 

for each task that appears in the:

 

vdsClient -s 0 getAllTasks 

 

output then they disappear and I'm able to move the SPM role to the other host.

 

This problem then crops up again on the new host once the SPM role is moved.  What's going on here?  Does anyone have any insight as to how to prevent this task from re-appearing?  Or why it's failing in the first place?

 

Beau