March 10, 2011 at 4:57 pm
I saw some solutions and aswers in your forum and found they very interesting.
I was wondering if someone can help me providing another answer.
I recently changed my position and I got the responsibility of a Production SQL 2005 Cluster running on Windows 2003.
I have two Disk Drives configured on it where bigger DBs reside and they are getting full soon,
Those disks are disk resources in the cluster.
Since I am not DBA, I am planning to add 2 bigger disks and copy the DBs to these two new disks with the same drive letters and no move SQL settings.
I cannot convert those new disks to dynamics neither to GPTs.
Since this is production cluster, I got a short nightly maintenance window and I got nervous because I have only 6 hrs to do this and I don't want to loss something. I do not have an environment to test this so I would appreciate any input.
This is the plan that I have
1. Stop the App and SQL Services
2. Login to the primary node 1 of the cluster
3. Take offline the SQL Server, SQL Server Agent and SQL Server Fulltext resources in SQL Group clearing the affect group check box for them.
4. Remove the dependencies from the Resources Disks in SQL Group.
5. Delete Resources Disk
6. On Disk Management Tool change labels and letters for "Old" disks to any
7. Initialize new Disks and Create new Partitions
8. Perform Quick Format and assign letters to new disks as the old disks had
9. Login to the secondary node 2 of the cluster
10. On Disk Management Tool change labels and letters for "Old" disks to any.
11. Login to primary Node 1 of the cluster
12. Create Resources Disks and set the dependencies from the Resources in SQL Group
13. Stop antivirus services related
14. Copy all content data from old disks to new disks
15. Reenable check box “affect the group” in the SQL resources.
16. Start SQL Engine and services related
17. Take online the resources in SQL Group
18. Perform cluster failover and check if it is working properly.
Regarding step 10, I don't know if when I delete the resources the "old" disks appears as available in node 2 in order to change the letters and then when I perform the failover node 2 can see the new disks with the new letters as set in node 1.
Are those steps fine?
Please help me!
Thank you.
Alex Salazar
Viewing 0 posts
You must be logged in to reply to this topic. Login to reply