Our DB Log File is getting full out of nowhere. We tried checking all the running tasks and reports nothing seems like affecting the Log File. Our fix is to run a shrink and truncate on our MSQL Server and that helps free up the space. Hoping someone can assist or experience this kind of issue and share their fix. Thank you!
couple of other things to check, provisioning transactions? Make sure that you are running Perf Maintenance task with Prune provisioning transactions.
Review this article too
Growth of tempDB in Microsoft SQL server - Compass (sailpoint.com)
1 Like
Our TempDB is working fine since we already have a scheduled job on SQL server if its hits threshold it will clean up the data.
We have separate drive on the DB server that stores the logs and that’s the one getting full in just 2 or 3 days.
@pbobadil
Did you check with your DBA if there are any long running queries from SailPoint application , if so from which host and what process is causing that, so you can analyze accordingly
Also regular backing up of log file can also help but in your case, it is filling very fast so I would suggest to identify root cause
Below steps can be used for regular backup
– START GENAI@CHATGPT4
BACKUP LOG [YourDatabaseName]
TO DISK = ‘PathToStoreLogBackup\YourDatabaseName_LogBackup.bak’
WITH NOFORMAT, NOINIT,
NAME = ‘YourDatabaseName-Transaction Log Backup’,
SKIP, NOREWIND, NOUNLOAD, STATS = 10
– END GENAI@CHATGPT4
Below is command for shriking
– START GENAI@CHATGPT4
USE [YourDatabaseName];
GO
DBCC SHRINKFILE (YourDatabase_Log, 1);
GO
– END GENAI@CHATGPT4
Please work with your DBA for above
1 Like
You can also take a look at:
– Remold
1 Like
This is the question I was going to ask.
A lot of the time, this is caused by big Identity Refresh jobs with a small number of partitions. Each partition thread keeps a session open the entire time, meaning that all of the changes made during that transaction will end up in transaction log space for database consistency. (In Oracle, this is what TEMP space is for.)
In big environments (several million users), I usually start refreshes with hundreds of partitions. Only a few will be running at any given time, but they finish up quickly and release any transaction logs waiting on them.
1 Like
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.