July 9, 2009 at 5:22 am
I have a SQL 2000 server that creates a transactionlog backup every 15 minutes. I have to do this because the logfile will grow really fast if I don't back it up.
The weird thing is that the logfile should shrink when the backup is complete but in some way it doesn't.
execute master.dbo.xp_slssqlmaint N'-PlanID 94615F0C-6402-4D6E-AC2E-94CADCBFFE83 -BkUpMedia DISK -DelBkUps 4DAYS -BkUpLog "E:\DBDUMPS" -CrBkSubDir -BkExt "trn" -NOOPTOLR -Throttle 100 -BufferCount 20 -MaxTransferSize 1048576 -CompressionLevel 1 -Priority 0 -Logging 0 -Affinity 0 -WriteHistory '
I use Litespeed to create the backups because they would be too big otherwise. Litespeed shouldn't be the problem because I use it with several other SQL 2000 servers with the same SQL Server version (SQL Server 2000 SP3).
Can anybody give me a lead what possible could be the problem.
Thanx in advance
July 9, 2009 at 5:42 am
use dbcc shrinkfile
or backup log with truncate only option or set recovery model to simple
July 9, 2009 at 5:49 am
Its a common misconception that it's a good idea to shrink the transaction log. If you do, it will just grow again.
If you have backups every 15 minutes, that's fine. The transaction lof itself will settle at a size and probably stay there. Backing it up just leaves more space for the next transaction to be logged.
The only time it's worth shrinking it is if you took some unusual activity that made it grow much bigger than it would in everyday use.
Tim
.
July 9, 2009 at 8:50 am
ramuvanparti (7/9/2009)
or backup log with truncate only option or set recovery model to simple
That's very bad advice to give someone since you don't explain what the consequences are of doing either of those.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
July 9, 2009 at 9:33 am
Go with Tim's suggestion. Just run the log backups, don't shrink it.
And don't mess with the recovery model unless you know what you are doing. Certainly don't truncate it if you need this data and want to be able to recover to a point in time.
July 10, 2009 at 12:26 am
I know the consequenses about setting the recovery mode etc and I don't want to set it to simple because I want to be able to restore to a point in time.
The weird thing is that the logfile grows from 100MB to 45GB overnight.
I'll create a job that will shrink the file but it's weird anyway because I have this on no other server. I'll have to work around it fortunately the server is planned to be upgraded to SQL 2005 maybe the problem is solved after the upgrade.
Thanx for the advise
July 10, 2009 at 12:36 am
sanderstad (7/10/2009)
The weird thing is that the logfile grows from 100MB to 45GB overnight.
Some activity is happening at night that's consuming heavy log file space? I suspect it's index rebuilding? can u check out what's happening during night that's causing massive log growth?
July 10, 2009 at 12:45 am
sanderstad (7/10/2009)
The weird thing is that the logfile grows from 100MB to 45GB overnight.
There's some operation overnight that requires 45 GB or so of log space.
Rebuilding all indexes?
Doing large data archiving?
Loading lots of data?
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
July 10, 2009 at 2:15 pm
sanderstad (7/10/2009)
I know the consequenses about setting the recovery mode etc and I don't want to set it to simple because I want to be able to restore to a point in time.The weird thing is that the logfile grows from 100MB to 45GB overnight.
I'll create a job that will shrink the file but it's weird anyway because I have this on no other server. I'll have to work around it fortunately the server is planned to be upgraded to SQL 2005 maybe the problem is solved after the upgrade.
Thanx for the advise
Shrinking the file everynight - just so it can grow again to 45GB is a waste of time and is causing you to suffer through performance issues. You shrink the file down to 100MB and probably have a very small increment for growing the file (default is either 1MB or 10% growth). With that you end up with a lot of autogrow events - every night.
If that system needs is growing the log to 45GB every night, here is what I would do:
1) shrink the file one time to 100MB
2) right after shrinking the file - manually grow the file to 50GB in one pass (if possible). If this takes too long, grow it to 50GB at 5GB intervals.
3) Leave it alone after that
Jeffrey Williams
“We are all faced with a series of great opportunities brilliantly disguised as impossible situations.”
― Charles R. Swindoll
How to post questions to get better answers faster
Managing Transaction Logs
Viewing 9 posts - 1 through 8 (of 8 total)
You must be logged in to reply to this topic. Login to reply