1

I'm experiencing a problem with the CQL on an exchange server which is acting as a passive node in a DAG cluster. To make the counters show the correct values in EMC and PS i've been adviced to restart the Cluster Service on the passive node. Is this safe to do in a production environment? The Passive Exchange-role is the only role for this server so no other systems should be affected as far as I know.

Thanks for your help!

simme
  • 113
  • 1
  • 6
  • Okay, I dont know if this may help you give us a solution. We tried by updating the catalog index with Update-MailBoxDatabaseCopy -CatalogOnly. The database have now been in the resychronizing stage for a good half an hour. Any ideas? – simme Jul 20 '12 at 12:03

2 Answers2

0

As long as the passive node in the cluster is actually passive (eg no cluster aware applications are currently running on it) you should be OK to reboot it or whatever you want to do to it. You can verify node ownership of resources in the GUI or with get-clusterresource

Jim B
  • 24,081
  • 4
  • 36
  • 60
0

Check that your database is mounted in the Active copy in the Exchange Management Console. Exchange uses some functions from Failover Clustering, but don't do the mistake of trying to administer your DAG from within Cluster Manager. Check that the database(s) are all mounted on your active server, and only replicating on your passive. If that's the case you can safely go ahead and reboot the passive one or restart the clustering service, whichever you prefer. When it's back online monitor the replication queue length on your databases (still in the Exchange Management Console).Should be zero or close to zero.

Trondh
  • 4,201
  • 24
  • 27