CCR: Process to recover after power down of Active Node

[JPJeffery]

Peer review time!

Exchange 2007: Our Active node (PROD010) ran out of disk space (for the transaction logs) and as an emergency measure was powered off to allow the passive node (DR01) to take over. This was done WITHOUT issuing a failover command first (so, when it powered down it did so still thinking it was the active node).

We've been running off DR01 (the 'old' passive node) all day, with the only issue being that we've had to manually move log files to keep the logs drive with sufficient free space all day.

My plan to recover the situation this evening is:

[ol 1]
[li]Power on the old active node, PROD01.[/li]
[li]At this point I assume the two nodes will have a conversation and DR01 will convince PROD01 that PROD01 is now the passive node and DR01 is the active node[/li]
[li]Once I've checked that DR01 is the Active/Quorum Owner, run Update Storage Group Copy to get PROD01 up to date with all the day's emails (I expect this to take a wee while!)[/li]
[li]Once the Copy Status is confirmed as OK for all Storage Groups, failback to PROD01 and Move Group in Cluster Admin to revert the Active/Quorum Owner back to PROD01[/li]
[/ol]
Does this sound right?

JJ
[small][purple]Variables won't. Constants aren't[/purple]
There is no apostrophe in the plural of PC (or PST, or CPU, or HDD, or FDD, or photo, or breakfast...and so on)[/small]

 

[JPJeffery]

Well, this is clearly one of the quieter Tek-Tips forums.

For the record, my approach listed above worked a treat.

JJ
[small][purple]Variables won't. Constants aren't[/purple]
There is no apostrophe in the plural of PC (or PST, or CPU, or HDD, or FDD, or photo, or breakfast...and so on)[/small]