Sometimes DBAs get into a situation where they don’t know what the next steps is to fix the issue. Here is one of the situation where DBA contacted me with below error in installing a patch for SQL Server in clustered environment. If you don’t know then I can tell you that I do provide quick consultancy. You can read more here On Demand (55 minutes). In this blog post we will learn about how to fix an installation error.
I asked him to share the Summary.txt file by following this blog. SQL SERVER – Installation Log Summary File Location – 2012 – 2008 R2
Here is the information form the summary.
Instance MSSQLSERVER overall summary:
Final result: The patch installer has failed to update the shared features. To determine the reason for failure, review the log files.
Exit code (Decimal): -568706566
Exit facility code: 1562
Exit error code: 14842
Exit message: The cluster resource is not available. (Exception from HRESULT: 0x8007138E)
Start time: 2018-01-02 15:00:21
End time: 2018-01-02 15:06:14
Requested action: Patch
As we can see above, there is some issue with cluster resource. So as a next steps I asked for the detail.txt to know more about the exception.
After carefully reading the error, I realized that error stack is as following:
Interop.MSClusterLib.ISClusResource.get_Disk
ClusterPhysicalDisk.get_Partitions
ClusterDiskPublicConfigObject.IsPathOnSharedDisk
ValidateNotOnSharedDisk
Validate_InstallSharedDir
If we look at stack, it looks similar as my earlier blog SQL SERVER – Installation Error – The wrong diskette is in the drive. Insert (Volume Serial Number: ) into drive.
When I checked disk, resources found that one of the disk was in failed state. It looks like all the roles and the disks associated with it should be online apart from the role we are concerned with.
WORKAROUND/SOLUTION
There were some hardware issues which happened on this cluster. Due to which one of the disk was in failed state. Even if it was not used by our SQL instance, it looks like SQL Setup checks for all the disks. We removed the disk from the cluster and then issue was resolved.
Reference: Pinal Dave (https://blog.sqlauthority.com)
