Start a Conversation

Unsolved

This post is more than 5 years old

2795

July 30th, 2012 19:00

open migrator with sql cluster

I had  a issue last week , while trying to migrate the old SAN to the new one using open migrator . being the sql cluster the passive node was shut down before the server reboot,  after the migration so that the new luns get the old san drive letter . As per the process it new san did get the drive letter , but as soon as we removed the old san we lost the connectivity to the new one . Can anyone has come across any issue like this as before .As well I have listed the steps as follow , please assist if I am wrong.

1. Assign the new san to both the node on the cluster (active/passive)

2. reboot he active node which has the OM install after the filter divers are attched.

3, after the reboot start the migration

4, once the migration is done , complete the migration and shut down the passive node , so the active node is rebooted,

5. then rebuit the qorum disk , and as well recreate the cluster resource.

please advise.

180 Posts

July 31st, 2012 05:00

Hello,

The team is going to move this post to the Support Forums community where your question can be addressed.

Thanks,

Stephanie

666 Posts

July 31st, 2012 06:00

this thread has been moved to the Support Community to ensure it gets to the right audience.

29 Posts

July 31st, 2012 08:00

we did check all the parameters and we did all the required ugrades prior to the migration , it still dont make sense why the new devices will become not accesable when we remove the old drive from the system.

29 Posts

July 31st, 2012 08:00

OM version (3.12)  , OS -2003 R2 sql cluster.

89 Posts

July 31st, 2012 08:00

Can you please clarify what version of Open Migrator you are using as well as the OS version? 

Thanks

89 Posts

July 31st, 2012 08:00

Your summary appears sound, but you will need to review the specific steps outlined in the Product Guide and Release Notes for version 3.12. 

Ensuring proper connectivity to both SAN's also depends heavily on the connectivity requirements of the server.  You will be able to review if your FC or iSCSI HBAs are proper by using the support tools found here:

Home > Support > Product and Diagnostic Tools > Environment Analysis Tools

89 Posts

July 31st, 2012 13:00

dkakljkml wrote:

5. then rebuit the qorum disk , and as well recreate the cluster resource.

The only place I can see room for error might be here. You replaced the quorum disk from the old SAN with a disk from the new SAN, correct?  There shouldn't be a need to recreate any resources. 

89 Posts

July 31st, 2012 13:00

If the migration is finalized and the new data disk is no longer part of the cluster there should be no impact removing the old SAN disk.  Replacing the quorum disk with a new one should also not cause lost connection to the SAN especially if you are using Cluster Administrator to do it.

http://support.microsoft.com/kb/280353

29 Posts

July 31st, 2012 13:00

what  happens , if i remove the data and quorum disk from old san just on the active cluster , with passive completely shut down. Do this still cause the data not accessible on the active node ( with quorum not yet configured & passive node is down). then i see all the data on the new san i then recreate the quorum from the scratch, is this possible .

29 Posts

August 1st, 2012 11:00

Thank Maliboo

No Events found!

Top