Can any one suggest me , my master sever is running on a windows platform and very frequently goes down . I figured out that unified logs are genrating at around 51 Mb every 34 minutes ,which is filling up the disk space , as well as i am also not getting alert when my catalog is getting filled up although mail alert has been configured for this .
Please suggest a way to automatically delete the unified logs every two days or so .I dont want to redirect them to any other location as have only two drives ( i.e C: & D:) in my master .
This will get me some space from disk .
We had a similar problem with the logs right after updating. In addition to setting the logs to a specific size, we had to schedule vxlogmgr to run frequently to clean up. I am running on Unix but it should not be that much different on Windows.
Check your version of EMM against the patch notes. On Unix we had a memory problem that was killing the EMM Database until we updated to 6.5.4. Not sure if this is the same on Windows but you may be hitting a similar bug.
Also check your offline catalog backup schedule. If you have it set to run everytime you are not running any backups, it will also shut down the EMM database while the cold backup runs.
You need to make really sure your catalog space never runs out. If it does you are looking at a lot of corruption in the catalog that can cause havok with your entire setup. Anytime you have the catalog fill up you need to run the bpdbm -consistency checks.