May 11, 2009 at 11:33 am
I have three servers with database mirroring set up in high safety with automatic failover mode. This weekend I had an unexpected failover on all databases from the primary to the mirror. Searching through the server logs I found entries like this one for each mirrored database on the principal server at the time when the failover seemed to have occured.
Event Type:Information
Event Source:MSSQLSERVER
Event Category:(2)
Event ID:1442
Date:5/10/2009
Time:6:31:16 PM
User:N/A
Computer:TBAS1
Description:
Database mirroring is inactive for database 'Personnel'. This is an informational message only. No user action is required.
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.
Data:
0000: a2 05 00 00 0a 00 00 00 ¢.......
0008: 06 00 00 00 54 00 42 00 ....T.B.
0010: 41 00 53 00 31 00 00 00 A.S.1...
0018: 00 00 00 00 ....
They were followed by several of these entries.
Event Type:Error
Event Source:MSSQLSERVER
Event Category:(2)
Event ID:1404
Date:5/10/2009
Time:6:31:16 PM
User:N/A
Computer:TBAS1
Description:
The command failed because the database mirror is busy. Reissue the command later.
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.
Data:
0000: 7c 05 00 00 10 00 00 00 |.......
0008: 06 00 00 00 54 00 42 00 ....T.B.
0010: 41 00 53 00 31 00 00 00 A.S.1...
0018: 00 00 00 00 ....
Then, a minute later these types of entries for each mirrored database.
Event Type:Information
Event Source:MSSQLSERVER
Event Category:(2)
Event ID:1480
Date:5/10/2009
Time:6:31:23 PM
User:N/A
Computer:TBAS1
Description:
The mirrored database "Fleet" is changing roles from "PRINCIPAL" to "MIRROR" due to Role Syncronization.
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.
Data:
0000: c8 05 00 00 0a 00 00 00 È.......
0008: 06 00 00 00 54 00 42 00 ....T.B.
0010: 41 00 53 00 31 00 00 00 A.S.1...
0018: 00 00 00 00 ....
I have no scheduled tasks which would have caused a server shutdown, and I can't find any record in the logs of the server having gone offline. The stats on my network show no errors to the switch ports that any of these three servers are connected to. Does anyone have any idea what could have caused this unexpected failover? Can you point me in the right direction?
May 18, 2012 at 6:38 am
I had the same issue with the DB Mirror process last month, causing an auto fail over, which caused severe issues. I just had the same error last week, but luckily, it did not auto fail over. Any ideas on this? This appears to be network communication issue between Primary and Mirror?
Thanks!
Scott
May 18, 2012 at 6:49 am
It hasn't happened again, only the one time. I never figured out what caused it.
May 18, 2012 at 1:21 pm
We had the same issue. I want to say it is the communication between the witness and the Principal. If there is a small delay that might cause a timeout for a status check it automatically brings up the Mirror in the Principle role.
We could never get a good way to keep this from happening. Might have been that our witness was at the mirror site and had to go across a slower WAN.
We instead removed the witness and automatic failover as we will know once the server goes down and can remote to our network to manually failover.
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply