Symptoms
Consider the following scenario:
-
You have Cluster Shared Volumes (CSV) resources configured on a Windows Server 2012 R2-based failover cluster.
-
You create backups of the CSV volumes.
Note Volume Shadow Copy Service (VSS) snapshots of the CSV volumes are accumulated by the Volsnap provider. -
The cluster service tries to bring a CSV volume online, and the service queries for active snapshots that are maintained by the Volsnap provider.
In this scenario, the cluster service takes longer than expected to bring the CSV volume online.
Note The time-out value for bringing a CSV volume online is three minutes. If the cluster service takes more than three minutes to bring the volume online, the operation fails.
Resolution
Update information
To resolve this issue, install update rollup 2903939. For more information about how to obtain this update rollup package, click the following article number to view the article in the Microsoft Knowledge Base:
2903939 Windows RT 8.1, Windows 8.1, and Windows Server 2012 R2 update rollup: December 2013
The update introduces the following two configurable new policies:
-
Snapshot diff area is limited to 20% of the volume size, up to a max of 1 Tera Byte (TB).
-
The maximum snapshot lifetime is 7 days. The old snapshots will be deleted automatically by the cluster service.
Status
Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section.
More Information
For more information about software update terminology, click the following article number to view the article in the Microsoft Knowledge Base:
824684 Description of the standard terminology that is used to describe Microsoft software updates