June 8, 2011 at 11:22 am
I ran into this issue again today when my principal server decided to do a restart today. I was curious if anyone else has expereinced a similar issue. I never have this issue with a restart of the partner.
I am running the following:
Microsoft SQL Server 2005 - 9.00.4053.00 (Intel X86) May 26 2009 14:24:20 Copyright (c) 1988-2005 Microsoft Corporation Enterprise Edition on Windows NT 5.2 (Build 3790: Service Pack 2)
The SQL Edition is Enterprise and I'm using AWE.
The server has 3 instance with database mirroring running on 2 of the instance.
On the first instance I have 6 databases mirrored
On the second instance I have 2 databases mirrored
out of the 8 databases I only have the issue with 3 databases on the first instance and it's always the same 3.
The following error is returned in the error log when it's attempting to recover the databases:
Bypassing recovery for database 'DATABASENAME' because it is marked as an inaccessible database mirroring database. A problem exists with the mirroring session. The session either lacks a quorum or the communications links are broken because of problems with links, endpoint configuration, or permissions (for the server account or security certificate). To gain access to the database, figure out what has changed in the session configuration and undo the change.
My workaround has been to do the following during my maintenance cycles.
Break the mirror (Run ALTER DATABASE DATABASENAME SET PARTNER OFF on the partner instance)
Wait for the database to recover on the principal instance
Restablish the mirror (Run ALTER DATABASE DATABASENAME SET PARTNER= N'TCP://SERVER:5021' on the partner then run ALTER DATABASE DATABASENAME SET PARTNER= N'TCP://SERVER:5021' on the principal)
Once the database mirror is syncronized then set it to high performance (Run ALTER DATABASE DATABASENAME SET SAFETY OFF on the principal)
I was just curious is anyone else has seen this behavior.
June 9, 2011 at 11:48 am
No, but only because I don't use DB Mirroring at the moment. But it sounds like a resource issue since you're mirroring 6 dbs on that one instance.
Have you run PerfMon and a server-side trace against that instance to verify that there are no bottlenecks causing the restart? What do the logs (SQL Server and Event) tell you about what's going on around that time of the day?
July 5, 2011 at 1:36 pm
I have not verified this. Howerver, this looks to match the issue that I am experiencing.
Article ID: 979042 - Last Review: February 17, 2010 - Revision: 2.2
FIX: The principal database is not recovered if the database has a large number of virtual log files in SQL Server 2005 or in SQL Server 2008
source: http://support.microsoft.com/kb/979042
Inital research has validated that the databases in question do have a large number of virtual log files and I'm not at a patch level higher than 9.00.4285 HF8.
July 7, 2011 at 4:57 am
Let us know if this turns out to be the issue.
August 18, 2011 at 11:46 am
SQL Server 2005 SP4 resolved this issue.
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply