NT Cluster - Add disk array

[coolb4all]

Hello All,

I've got a cluster running on 2 Proliants, connected to a Ra4000 disk array. I've now purchased a VA7400 (SAN) disk array from HP and I would like to migrate my data to the HP. Therefor I've installed a second HBA in my nodes, to connect to the HP. No problem over here. I can see the new LUNs and the new ntfs partitions I've created on both nodes(even in explorer I can use them). BUT... when I want to create a new "physical disk" device in cluster administrator the new disks are not detected. Is it possible that MSCS doesn't detect this new "shared disk device" automatically. Should something be installed manually or changed in the registry or is it not possible and should I re-install the cluster (at the beginning of the install I know that the setup asks which disks to include in the shared device). It seems to me that it would be very user-unfriendly if you should re-install your cluster when your old disk array gets full.

Regards.

 

[mrcook]

Sorry I don't have an answer for you but I am having the exact same problem. If you managed to find the answer elsewhere I'd appreicate it if you could post here for me.

Thanks,

mrcook

 

[wisandanu]

I found this in knowledge base, may be it will be usefull for you, just let us know, OK

Try this article:
How to Extend the Disk Space of an Existing Shared Disk with Windows Clustering (Q263590)
How to Install Additional Drives on the Shared SCSI Bus (Q175278)

 

[mrcook]

What we found that worked was :

1) create virtual disk and present to nodes of cluster.
2) rescan for disks in disk management. may require a reboot.
3) If you see new disk format and label with same drive letter in both nodes.
4) reboot, seems problems were sessional
5) we were then able to add "physical disk" resource in cluster admin

 

[SuperFunk]

This is whats called a "rolling" upgrade. It's a bit tedius, but understandably so.

To add a new drive to the shared space, you will need to...

1. Turn off the cluster service on both servers
2. Power down a server
3. Assign, and configure the new drives on the server that is still on. Configure the drive letters/format/fdisk etc.
4. Then you power this server down, and turn the other server on. Then you double check to see if the second server can see the drives and the drive letters are the same as the previous server
5. If everyhing checks out so far, start back up the cluster services, then reboot (you may not need to, but I can't rememner. It's safe to reboot though!)
6. Start up the cluster admin, then you should be able to add the new disks onto the cluster as resources.
7. Start up the other server and turn on the cluster service.
8. everything should be honky dory