So I had these 4 Hyper-V servers at a remote location and was having trouble with the cluster configuration validation report reporting the following error:
(Hashing the error out here for the search engines – Two disks have been found on node XXXX that cannot be distinguished from one another. The disks involved have disk signature XXXX, SCSI page xxx 83h VPD descriptor XXXX, SCSI page 80h VPD serial number XXXX. Please verify the storage configuration. You must either mask (unpresent, detach) one of these LUNs at this node, or, run validation and specify a disk list that includes only one of these disks, for example by using the Test-Cluster cmdlet in Windows PowerShell.). Phewww..!!
Now since clusters don’t like disk issues, this had to be fixed before I proceeded. The above error tells me that the hosts thinks it has 2 identical disks. Now, I know this isn’t likely especially with them being disks from an array. The hosts all had 2 HBA cards and used PowerPath to get to a EMX Symmetrix array. This took me to the PowerPath utility where I noticed that both HBAs had active paths to the array. I asked the storage team as to what kind of array the Symmetrix was, turns out it was a active-passive array. I then set one of the HBAs to ‘Standby’ and Voila!, the error’s now gone. Check the resultant PowerPath config below:
The next error was:
Hashing the error out here again for the search engines – Test Disk 0 does not support SCSI-3 Persistent Reservations commands needed to support clustered Storage Pools. Some storage devices require specific firmware versions or settings to function properly with failover clusters. Please contact your storage administrator or storage vendor to check the configuration of the storage to allow it to function properly with failover clusters.
Whipped out my Google-fu and discovered others had been having the same issues. Here’s a link to someone who had the same problem. Read the article for the details, in short the error/warning’s benign. I carried on with creating my cluster and there haven’t been any issues since.