June 26, 2009 at 10:26 am
In cluster Administrator, we have 2 groups.
Sql group -> D, E, F, and T drives
Backup group -> Z drive.
How to make the Backup group drive Z as dependency on sql group. Because Backup group is not moving to other node when failover occurs.
If you have only one instance on the active node:
1.How many groups you create?
2.Does we need to create a backup group to keep the backups on the backup drive in that group
3.How to make the backup drive as dependency on sql group?
Please advice me?
June 26, 2009 at 12:03 pm
We create a minimum of 3 cluster groups:
Cluster Group - contains the cluster resources
MSDTC Group - contains the resources for MSDTC
SQL Group - contains the resources for a SQL Server instance
If we need to add a drive for backups, that drive is added to the SQL Group and the SQL Server resource will have a dependency on that drive, as well as a dependency on the data/log/temp/system drives.
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
June 26, 2009 at 12:26 pm
Thank You
We create a minimum of 3 cluster groups:
Cluster Group - contains the cluster resources
MSDTC Group - contains the resources for MSDTC
SQL Group - contains the resources for a SQL Server instance
We also have the above goups and an extra backup group for backups.
Now, I'm not able to add the Backup drive which is in backup group as dependency on sql group.
That's why its not failing over to other node when fail over occurs and the backups are failing.
So Our configuration is wrong and I need to delete the backup group and add the backup drive in the SQL group and add as a depency right?
Correct me if I understood wrong?
thanks
June 26, 2009 at 12:46 pm
Hi,
from this link http://www.mssqltips.com/tip.asp?tip=1651
Here we have a separate group DEVBACKUP to keep the bakups. In this scenario, how this group failover to other node when failover initiated due to some hardware problems
Please clarify me whether we can create a backup group and add it as dependency on sql group or DO NOT create any backup group just include the backup drive in the sql group.
thanks for yor help
June 26, 2009 at 12:47 pm
Not so much a question of right/wrong - but, more like which is better for you. I don't see a problem with having a resource group setup just for the backup drives, but there are things that you need to be aware of.
For example, if the drive is not in the same group - the SQL Server cannot see the drive. So, for backups you would have to reference the drive using the share name. If the drive is in the same group, SQL Server can see it and it can be referenced directly in your maintenance plans.
Other than that - no problems with failover. If the node is down - all groups will failover. If the node is not up, but SQL Server is failed over - won't matter.
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
June 26, 2009 at 3:18 pm
Thank You,
For example, if the drive is not in the same group - the SQL Server cannot see the drive. So, for backups you would have to reference the drive using the share name
But I did not create any share but my backups working well to the Z drive. I had given the path like below
Backup database db_name to disk ='Z:\Backups\db_name.bak'---and its writing the backup to Z drive..
can you explain me a bit more plz...
June 26, 2009 at 3:36 pm
klnsuddu (6/26/2009)
Thank You,For example, if the drive is not in the same group - the SQL Server cannot see the drive. So, for backups you would have to reference the drive using the share name
But I did not create any share but my backups working well to the Z drive. I had given the path like below
Backup database db_name to disk ='Z:\Backups\db_name.bak'---and its writing the backup to Z drive..
can you explain me a bit more plz...
Sorry, should have clarified that just a little. If the backup group is on node1 and the SQL group is on node2, then SQL Server will not be able to see the drive because it does not exist on that node.
That is why you would need to either move the groups at the same time (failover), or reference the drive using the UNC path (\\{name resource of backup group}\z$).
My recommendation would be to put the Z: drive in the same group as SQL Server. You don't need to make SQL Server dependent on that drive - but if it is in the same group it will always be on the same node and accessible.
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
June 27, 2009 at 8:59 pm
Thank you,
I understand the scenario as below.
If we create a Backup group, then we need to give the UNC path for the backups so that even if the node fail over to other node, the backups will not fail.
If we add the backup drive in SQL group itself, then no need to create a separate backup group and no need to mention the UNC path for the backups.
So what is the advantage/disadvantage of having or not having backup group?
June 27, 2009 at 11:29 pm
I don't use a separate backup group - so I really couldn't say what advantages there are.
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
June 28, 2009 at 10:37 pm
Even we follow same technique as of Jeffery of having backup disks cluster resources in sql server cluster group rather than creating seperate group. I also faced this problem of no failover of backup group when sql server group fails.
On my research i found only two possible solutions:
1. Move backup disk resource in same cluster group as of sql server
2. Execute a script which runs to check whether all cluster groups have failed over when one of them fails.
Found first one the best option of both. 🙂
Viewing 10 posts - 1 through 9 (of 9 total)
You must be logged in to reply to this topic. Login to reply