NMSQL and Cluster behaviour

Question

Hi All,

I am working on a project to role out the MSSQL module to a number of clustered servers.

The version of the module is 5.2 and we have hit an interesting observation. Backup server is 7.6.1.5

The cluster has 2 physical node and 4 virtual nodes it is running W2K3, 3 of the Instances backup a 2Mb test database in 20-30seconds where as 1 of the instances backs up the 2Meg database in 1min 20secs. When we run SQL Enterprise Manager scripted backups for all SQL Server instances and db's on the cluster the slow instance takes over 50 minutes on that instance, with all of the other instances managing to complete all of their backups in 6-10 mins.

Can anyone shed any light on what causes to look for, that would explain the significant increase in time required to peform the 2Mb database for just one of the instances in the cluster. The behaviour is the same no matter where the instance is running or the combination of instances that are running at the time the backup is run.

Any ideas?

Thanks,

Mark.

ble1 · Answer

Sounds as this is something on SQL side and some SQL debugging will be required to see where does the whole thing get stuck.  All my SQL clusters have only 1 virtual node (but up to 8 instances and then up to 20 databases per instance) and I have never seen any behaviour as yours.

MarkL2 · Answer

I suspect as much myself. I have logged a call with EMC Support as well just genereally trying to find out if there is anything in NetWorker that could cause this issue. Not being a DBA myself makes it a little more tricky to troubleshoot as I have to rely on what the DBA's are telling me.

That said my colleagues and I all think this is something specifically to do with that instance and possibly some setting somewhere.

Will be doing more troubleshooting today and over the next few days so will provide an update if we find the answer.

NetWorker

NMSQL and Cluster behaviour

Was this post helpful?