I have been tasked with testing DR, and want to get input before I pull the trigger.
Production mailbox server is a Single Copy Cluster with two nodes. Stand-alone server is at the DR site. SCR is running and functional.
I need to failover to the DR site, test, and go back - all while leaving the production mailbox DBs intact. Re-seeding all DBs is not feasible.
The link here was very helpful, but it assumed there was a SCC in place at the DR site. Can anyone look at my procedures and tell me if this will work?
1- Dismount Production mailbox DBs and shut down passive, then active nodes in the SCC.
2- Run Restore-StorageGroupCopy -force on the SCR Target server to make the target db mountable.
3- If necessary, run eseutil /r to get target db into Clean Shutdown.
4- Use Move-StorageGroupPath and MoveDatabasePath to point the target server storage groups to the SCR data and logs.
5- Set the target database to be restorable.
6- Mount the target database.
7- Point user(s) to the target databases using "Move-Mailbox -ConfigurationOnly".
8- Setup Send/receive connectors for target Data Center.
9- Point MX records to DR Site (a few days before test, set ttl very low so it can be switched quickly)
10- Test Send/receive and mailbox availability.
11- When done testing, dismount the target database.
12- Use Move-StorageGroupPath and MoveDatabasePath to point the target server storage groups/mailbox paths to another location.
13- Send/receive connectors back to Production Data Center.
14- Point user(s) back to the source database using "Move-Mailbox -ConfigurationOnly"
15- Point routing and MX records back to Production Site
16- Power up Active, then Passive SCC nodes.
17- Finally, will I need to disable and re-enable SCR on all SG's?
< Message edited by jb7102 -- 25.May2011 4:17:15 PM >
Well, we thought we had it working, but when re-enabling SCR to original configuration, it SAID it was healthy. But when we looked closer, ReplayQueueLength was growing..In reality, it seems the two .edb files were different enough to cause replay not to occur in the SCR target.
So we're back to square one...
For our DR test, we can failover to DR site, test, and failback to the production site - but we have to re-seed the target when we are done. Sucks...
< Message edited by jb7102 -- 25.May2011 4:21:38 PM >
|
My Subscription |
My Forums |
Address Book |
Member List |
Search |
FAQ |
Ticket List |
Log Out
Printable Version
All Forums >> [Microsoft Exchange 2007] >> High Availability >> Test DR in Production Environment 
Test DR in Production Environment - 
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread
