What does it mean when message says Quolog.log will be reset
Windows Server 2003 R2 cluster. My %SystemRoot%\Cluster\ClusDB file is 6656KB before adding 3rd node, 7936KB after adding 3rd node. Cluster quorum log size set to 16384KB. File and Print only.
Why did I get the following message (isn't 16384KB large enough)?, What does it mean the log will be 'reset'? There were no following messages, should I assume the 'attempt' to reset the log was successful (and what would success be)? To what new value should I adjust the maximum size to be? I changed it to 24384KB.
Event Type: Warning
Event Source: ClusSvc
Event Category: Log Mgr
Event ID: 1017
Date: 11/5/2009
Time: 10:14:30 PM
User: N/A
Computer: FPNODE01C
Description:
The log file Q:\MSCS\quolog.log exceeds its maximum size. An attempt will be made to reset the log, or you should use the Cluster Administrator utility to adjust the maximum size.
Thank you.
Answers
- Well, I can't say that I agree with Tim on this one. Having the quorum log reset does not indicate that you have a corrupt quorum. I don't have all the specifics on how this works, but this is my understanding of what this message means:
The quolog.log is the log file for the quorum database. It tracks changes that were made to the clusdb since the database was last updated. The actual cluster database is stored in the chk* file on the quorum disk and locally on each node. The quolog.log is updated for all changes that have been made to the cluster and once this file reaches the "Reset quorum log at:" value, it applies these changes to the chk* file with these updates and then resets the quolog.log file back to its minimal size.
Basically, this reset is not a big deal and shouldn't really have any affect on the cluster. The only time it could affect your cluster would be if you are resetting constantly.
As a general rule, you'll want to have the Reset Quorum Log at value set to twice the size of the clusdb. More details on this can be found in MSKB 257897
Hope this helps.
Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner- Proposed As Answer byTim Quan - MSFTMSFT, ModeratorTuesday, November 10, 2009 2:20 AM
- Marked As Answer byGeorge Perkins Thursday, November 12, 2009 3:29 PM
All Replies
Hi George,
It seems that this log file is corrupted.
First I suggest increasing the maximum size. If the issue persists, to correct the problem, please use the "noquorumlogging" selection at starting the Cluster service. the selection allows the cluster to start without quorum logging. You may then access the quorum disk and remove the corrupted Quolog.log file.
1.If the Cluster service was running, use Control Panel on both nodes to stop the Cluster service.
2.On 1 node, use the Services tool on Control Panel to specify "-noquorumlogging" as the startup parameter for the Cluster service and start the service.
3.On the quorum disk, run Chkdsk. If the disk does not display corruption, the log document may be corrupted. on the case, remove the Quolog.log document and any .tmp documents located on the MSCS Directory on the quorum disk.
4.In Services, stop the Cluster service, and then start the Cluster service without startup parameters. After the service starts, you may start the service on the other node.
Related information:
http://support.microsoft.com/kb/257905/en-us
Hope it helps.
Tim Quan - MSFT
- Well, I can't say that I agree with Tim on this one. Having the quorum log reset does not indicate that you have a corrupt quorum. I don't have all the specifics on how this works, but this is my understanding of what this message means:
The quolog.log is the log file for the quorum database. It tracks changes that were made to the clusdb since the database was last updated. The actual cluster database is stored in the chk* file on the quorum disk and locally on each node. The quolog.log is updated for all changes that have been made to the cluster and once this file reaches the "Reset quorum log at:" value, it applies these changes to the chk* file with these updates and then resets the quolog.log file back to its minimal size.
Basically, this reset is not a big deal and shouldn't really have any affect on the cluster. The only time it could affect your cluster would be if you are resetting constantly.
As a general rule, you'll want to have the Reset Quorum Log at value set to twice the size of the clusdb. More details on this can be found in MSKB 257897
Hope this helps.
Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner- Proposed As Answer byTim Quan - MSFTMSFT, ModeratorTuesday, November 10, 2009 2:20 AM
- Marked As Answer byGeorge Perkins Thursday, November 12, 2009 3:29 PM
JohnToner, thank you for correcting me. KB257897 is very clear.
- I was already familiar with KB 257897 and had already established the correct values for my quolog.log file (that is, quorum log should be at least 2x the size of the ClusDB). This was the case when I received the warning message (see the first entry of this thread). It is unfortunate that the MS developers choose to raise the level of the message to "warning" for what is normal operations. This message should be informational only. Would have saved me anxiety and the trouble to try to find more information.
Thank you Tim Quan and John Toner for responding to the forum.
