SQL Response blocks NetApp SnapManager

normajeanbnormajeanb Posts: 3
edited July 10, 2010 2:18PM in SQL Monitor Previous Versions
Our production system runs on Blade servers managed by NetApp filers.
Like many NetApp users, SnapManager is the most prominent part of our disaster recovery plan - this tool allows you to configure, schedule and manage SnapShot backups.
One option in the SnapManager backup step is the ability to mount a copy of the newly taken snapshot, and run DBCC checkdb to verify the snapshot backup of the database. After this step completes, SnapManager detaches the database and destroys the temporary LUN it was mounted on.
SnapManager backup jobs (with verification) fail when they run concurrent with SQLResponse because SQLResponse polls and accesses ALL the databases on the server, thereby blocking the detach action of the new database by SnapManager. Once this happens, all subsequent SnapManager backup jobs fail.
If only I could configure SQL Response to only poll certain databases - OR if we could schedule the polling periods for SQLResponse to exclude certain timespans we could work around this. But, I can't seem to locate these configuration options...
It's impractical for me to manually start and stop the service around SnapShot times - and I really lose the value of the product if it's not going most of the time to feed me alerts as they occur.
Do you have any suggestions?
Thanks, Normajean Bowen

Comments

  • Brian DonahueBrian Donahue Posts: 6,590 Bronze 1
    Hi,

    SQL Response polls the SQL Server every 15 seconds or so, and there isn't any way I know to change that. I would probably solve this problem by scheduling a maintenance interval for the time that you want to verify your snapshots, then create a batch file on the Response repository to stop and start the SQL Response Repository Service during those times by putting the batch file in a Windows Scheduled Task.

    Hopefully this would work for you.
  • Yes, I had thought of doing that too - so I went ahead and created the jobs, scheduling them around the snapshot database cloning time BUT once SQLResponse gets a force disconnect from the stopped service, it doesn't try to reconnect by itself...I have to remember to manually go back in to the app and re-establish the connection in order to get back online for my raised alerts.
    I don't think this is going to be a workable "workaround" for me...
    Normajean
  • Eddie DEddie D Posts: 1,807 Rose Gold 5
    Hi

    Following our telephone conversation earlier today, I have added your Support Call details to an existing Feature Request whose reference is SSR-1377 "Ability to configure the amount of e-mails that you get for certain alerts". Following our discussion regarding the E-Mail notifications.

    I created a new Feature Request whose reference is SSR-1602 "To be able to deselect databases on a SQL Instance from being monitored"

    Many Thanks
    Eddie
    Eddie Davis
    Senior Product Support Engineer
    Redgate Software Ltd
    Email: support@red-gate.com
  • I have seen many companies depending on NetApp SnapManager to backup their SQL using snapshot, though we have IBM DS8000 & would like to do what others are doing with NetApp SnapManager for SQL. The best I was able to get to was IBM Tivoli Storage FlashCopy Manager Tutorial though I am still curious how stable is this IBM Tivoli Storage FlashCopy Manager & how much of the NetApp SnapManager functionality it can offer. Is it worth it?

    Thanks for your help in advance,
Sign In or Register to comment.