My understanding is that windows scheduled jobs are not cluster aware therefore the job needs to enabled on both nodes in case 1 node dies. Unfortunately this leads to the annoying failure email message from SMSQL on the passive node.
The requested SnapManager operation cannot start at this time because there is another SnapManager operation currently in progress. Please retry this operation after the current operation has finished.
Has anyone found a way round this or are SQL jobs the best solution on clusters?
We too have recently started receiving these e-mails in our Notifications inbox. Nothing has changed recently, in particular no timings or frequency of backup jobs have changed so I am completely at a loss as to why we are receiving these messages all of a sudden.
If anybody has any solution or additional information that can be provided it would be most gratefully appreciated.
OK, so we have now resolved this small problem. I basically checked the Scheduled Tasks and found that there were two jobs that were running at exactly the same time. Basic really but it took a while to find it.
As a plan B we configure all our schedule jobs on both nodes to call a script on a cluster drive path. Only the active node can access the drive which has the scripts that do the work. So the passive node will just go back to sleep. Reduces maintenance too, as only a single copy of the 'active' scripts live on the shared drive and the local schedule batch files are just pointers.
Hope it helps
PS - Always worth closing the thread once you have your answer.