28 martie 2012 11:09
Can anyone assist me? One of my SQL nodes failed over today, and Ive no idea why. I cant see in the cluster log what exactly happens, can anyone see point me in the right direction as to the cause?
- Editat de Pilly170 28 martie 2012 11:10
28 martie 2012 13:33
Have you checked the SQL log at the same time where the failover occured ?
What OS and SQL version are you running ?
28 martie 2012 13:48
OS = 2008 R2
SQL = 2008R2
Looks like there was a backup of a single database happening at that time
30 martie 2012 15:40
The clusterlog that was uploaded doesn't contain any info regarding the failover. It seems that it has been overwritten. Please notice that the cluster log contains data in GMT.
Here is a nice blog from Jeff which explains the cluster log:
Understanding the Cluster Debug Log in 2008
Also, I noticed that the cluster log level is set to 5 (DBG). That means that the cluster log will quickly get filled. My suggestions:
1. Increade the cluster log size: <cluster log /size:200>
2. Set the cluster log level to 3 (default): <cluster log /level:3>
- Marcat ca răspuns de Vincent HuModerator 9 aprilie 2012 15:37