Issue:
After MS Exchange 2007 SP1 Installed, new error is displayed in the Event Log:
Error ID: 9871 - There is an online maintenance overlap for database <DATABASE>. Please revise the online maintenance schedule.
Workarounds/Solution Information:
The default scheduling of Exchange 2007 SP1 created a "conflict" in SBS 2008 for a small company of about 15 users.
To view the schedule:
In the Exchange Management Shell, issue the command:
GetMailboxDatabase -Identity <databasename> | FL
or
In the Exchange Management Console:
Server Configuration --> Mailbox --> Database Management Tab (lower window) --> Storage Group name --> Mailbox database
Right click the database --> Choose Properties --> General tab --> Maintenance Schedule Section (bottom portion of tab)
ex:
Run daily from 2:00 A.M. to 6:00 A.M.
If you noticed your maintenance is entering your backup window, change the interval/schedule to a free time range.
_____________________________
Notes:
The Event Log will log these as ESE events.
700 indicates the Information Store ESE is beginning its pass.
701 is used when database On-line Defragmentation completes in one pass.
703 is used when multiple passes are required. It's logged when a backup interrupts or the maintenance window expired (not long enough).
704 is used to tell you that it was interrupted/terminated.
In our case, we noticed that the scan returned a complete 701 before the termination 703 error. Evidently the ESE task, although complete is still running and is killed by the window closing rather than shutting down itself - a bug?
What we did...
- Confirm that you are seeing that message that all storage groups actually complete with 701 statuses. The body of those status messages will tell you pages saved, deleted documents purged, etc, as it completes all the passes. You should also see a message with something like, I've completed 471 passes in 471 days to indicate you are completing your passes and not really doing a 703 status. In addition, you'll get a 1221 saying that you saved so many bytes in <database>.
- We noticed our 1st storage group was x minutes. The second was only 1.5 minutes.
- Look to see if the maintenance is looping any services and starting them again. In our case, we noticed that ESE was looping over-and-over. In other words, it always finished but would keep looping until the window closed. So the first set of passes took under an hour, it just error out because the defrag part looped several times and then was killed. In our case, we were seeing the successful 1221 message for both datastores, it was re-running over and over until the maintenance window ended. Stated differently we were getting 703 at the beginning of the maintenance window saying it was done. Then it would loop several times during the 4 hour maintenance window, but not finish the last loop and result with the 703 warning again.
previous page
|