March 3, 2008 at 12:37 pm
The tempdb log grew to 25 GB and filled up, which caused an issue filling up the D: drive.
The errors were as follows...
Operating system error 112 (There is not enough space on the disk.) encountered.
The transaction log for database 'tempdb' is full. To find out why space in the log cannot be reused, see the log_reuse_wait_desc column in sys.sysdatabases.
My issue is that I could not open the Activity Monitor to figure out which processes were running and if I ran anything in the Query window, I would receive an error, also.
I had to have someone reboot the server, because I am not an Administrator, ( I am only the DBA!)
Once the server is rebooted, everything is okay, of course, because tempdb is deleted and rebuilt.
I am being asked what caused the issue in the first place.
I can not figure out what caused this issue, can I?
Al Cook :unsure:
March 3, 2008 at 12:47 pm
first off you should have all the rights you need to stop and start services - as you don't need to reboot - a service stop and start does the trick - although not a good way to solve your problem. Lots of things fill tempdb, index rebuilds, rolled back transactions, checkdb, developers making mistakes in code. The answer is that you should have been able to see what the problem wa sif the transactions/query was still there - query the dmvs.
You could limit the growth ofd tempdb, that will throw an error which should hopefully detect what or who caused the problem.
After the fact tricky - check all the logs there will be some errors which might help.
[font="Comic Sans MS"]The GrumpyOldDBA[/font]
www.grumpyolddba.co.uk
http://sqlblogcasts.com/blogs/grumpyolddba/
March 3, 2008 at 12:56 pm
The tempdb is created every time you start the database egine. You need to make sure you dont have any outstanding transactions. Check your error logs and the event view to see if anything abnormal happened. The tempdb does a number of things from temp tables, groupings, aggregation etc. Make sure that no one is making huge temp tables, rebulding indexes or both.
March 3, 2008 at 12:59 pm
All backups and reindexing jobs were completed Friday evening.
I looked at the system log, which logged the fact that the D: disk was at or near capacity and warned that some files may need to be deleted. This was Friday evening at 10:50 pm.
Then, the SQL Server log is where I read the other 2 errors stated in the original post. These were logged beginning at 11:05 am on Saturday morning.
Because all of the administration jobs were completed, I would have to say that this was due to some process (code) running by a developer or someone. I administrate the db's, but the "engineers" do there own coding and have dbo rights to the databases. This is only for this server.
It is a weird situation. I am not an Admin on the server either. If this happens again, I will try stopping and restarting SQL services.
Thanks,
Al
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply