K00lguy
asked on
Restart Passive Member of Exchange 2007 CCR Cluster
While restarting the passive node in our CCR cluster all our terminal server users lost connection to Exchange. It appears that all the storage groups went offline even though the active member ofthe cluster remained up.
Here was my procedure:
Passive Node:
1) Paused the node in Failover cluster management
2) Suspended replication between the 2 members of the clusters for all the storage groups
3) Test to make sure mail was still working - OK
4) Restart the passive node
Did I do something wrong or is there some sort of problem in my environment. To my knowledge this should work.
Thanks
Here was my procedure:
Passive Node:
1) Paused the node in Failover cluster management
2) Suspended replication between the 2 members of the clusters for all the storage groups
3) Test to make sure mail was still working - OK
4) Restart the passive node
Did I do something wrong or is there some sort of problem in my environment. To my knowledge this should work.
Thanks
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
JJ, does this means that we need to issue Suspend Replication command as well before failing it over ?
if CCR is robust or smart enough then why do we need to use the Move Storage Group cmd to do the failover gracefully ?
if CCR is robust or smart enough then why do we need to use the Move Storage Group cmd to do the failover gracefully ?
You cannot suspend the storage group copy before failing over. That will result in the nodes not being in sync. The nodes need to be in sync for a lossless failover.
When you use the Move Storage Group command, it performs checks to ensure the storage group is healthy on both nodes and in a state that would allow failover. If you just reboot the active node, those checks are not performed, which could result in the databases not mounting on the other node.
-JJ
When you use the Move Storage Group command, it performs checks to ensure the storage group is healthy on both nodes and in a state that would allow failover. If you just reboot the active node, those checks are not performed, which could result in the databases not mounting on the other node.
-JJ
Thanks JJ.
So even after the failover from the exchange management console, do I need to suspend the replication ?
So even after the failover from the exchange management console, do I need to suspend the replication ?
No, you do not need to suspend the replication.
-JJ
-JJ
Ok so when do we need to suspend the replication ? Any scenario to use that command ?
There may be many scenarios when you would need to suspend replication. One I can think of off the top of my head is if you need to reseed a database.
-JJ
-JJ
1) Patch passive node, reboot
2) move DB + cluster resources to passive node
3) Patch old active one, reboot
4) Move DB back to original active
In my environment at that time I had Exchange, BES, SQL, Enterprise Vault and all this needed to be coordinated in the right order for everything to be OK.