Error Message

VCS WARNING V-16-10051-9553 VMDg::monitor:Query imported cluster diskgroup state: The Diskgroup is not present.

The issue is because SCSI resets are being fired from a node on which the FD VMDg resource is offline. Issuing SCSI reset is by design as the DiskGroup is absent from the node on which FireDrill SG is offline and so SFW will attempt to determine the status of the disk group.

These SCSI resets are causing a reservation loss from the online node.

• If the reservation thread is able to get back the reservation before VMDg resource monitor is called – then the VMDg resource will not fault.

But

• If the VMDg resource monitor is scheduled before the reservation thread reserves the disk again – then we see that VMDg resource faults unexpectedly.

Hence, this issue is intermittent.

This will happen only in case of VMDg resources have non-existent DiskGroups where DGGuid attribute is not set – like is the case with the FD VMDg resource.

This issue is addressed in SFW-HA 6.0.1 CP3 (Hotfix_6_0_10022_ 3386077) and is also included in SFW-HA 6.1 or higher.

Please subscribe to this technical article and/or via the Veritas Operations Readiness Tools site (SORT) for updates regarding the issue and additional fixes for additional versions.

The fix adds a new attribute for VMDg resource ‘ForFireDrill’.  For the fix to work ‘ForFireDrill’ attribute value will need to be set to ‘true’ for resources that are part of the FireDrill service group only.

Please contact Veritas Technical Support for additional questions or concerns regarding this update.

Applies To

Veritas Storage Foundation HA for Windows 6.0,6.0.1 (SFW-HA)

Microsoft Windows Server 2008 R2