Hi All,

I was testing the gluster geo-replication on RHEV storage domain backed by gluster volume.
In this case, storage domain ( data domain ) was created with gluster replica 3 volume.

The VMs additional disks are carved out from this storage domain.

Now I have geo-replicated[1] the gluster volume to the remote volume.
When I try importing this storage domain in another RHEVM instance, it fails with error "internal engine error"
 I see the following error in engine.log

<snip>
2016-04-20 05:13:47,685 ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.DetachStorageDomainVDSCommand] (ajp-/127.0.0.1:8702-3) [20f6ea4c] Failed in 'DetachStorageDomainVDS' method
2016-04-20 05:13:47,708 ERROR [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (ajp-/127.0.0.1:8702-3) [20f6ea4c] Correlation ID: null, Call Stack: null, Custom Event ID: -1, Message: VDSM command failed: Cannot acquire host id: (u'89061d19-fb76-47c9-a4aa-22b0062b769e', SanlockException(-262, 'Sanlock lockspace add failure', 'Sanlock exception'))
2016-04-20 05:13:47,708 INFO [org.ovirt.engine.core.vdsbroker.irsbroker.DetachStorageDomainVDSCommand] (ajp-/127.0.0.1:8702-3) [20f6ea4c] Command 'org.ovirt.engine.core.vdsbroker.irsbroker.DetachStorageDomainVDSCommand' return value 'StatusOnlyReturnForXmlRpc [status=StatusForXmlRpc [code=661, message=Cannot acquire host id: (u'89061d19-fb76-47c9-a4aa-22b0062b769e', SanlockException(-262, 'Sanlock lockspace add failure', 'Sanlock exception'))]]'
</snip>

The complete logs are available in the fpaste[2]
Attaching the part of vdsm log to this mail

[1] - geo-replication is the feature in glusterfs where the contents of volume are asynchronously replicated in remote volume.
This is used for disaster-recovery workflow

[2] - https://paste.fedoraproject.org/357701/11448771/

Thanks,
Satheesaran S

_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users