High Log flush time???

  • Hi,

    We are using Spot light for Monitoring SQL Server 2005 and we are getting the below error frequently.

    What this error referencing to? What is the corrective action to get rid of this? Is this a serious issue? We are getting this for all most all databases

    TimeConnectionActionDetailsSeverityAlarm

    05/20/2009 4:01:32 PM ABCAlarm raisedthe database Search_DB has a high Log Flush Wait time (515 ms).LowDatabases - Log Flush Wait Time Alarm.

  • klnsuddu (5/21/2009)


    Hi,

    We are using Spot light for Monitoring SQL Server 2005 and we are getting the below error frequently.

    What this error referencing to? What is the corrective action to get rid of this? Is this a serious issue? We are getting this for all most all databases

    TimeConnectionActionDetailsSeverityAlarm

    05/20/2009 4:01:32 PM ABCAlarm raisedthe database Search_DB has a high Log Flush Wait time (515 ms).LowDatabases - Log Flush Wait Time Alarm.

    This means SQL server cannot write the log files onto the disk in a timely manner. Do you have DB mirroring set up by any chance to this database?

    What is your physical disk layout and also could I ask you what is the frequency of your tlog backups?

  • Is it share point by any chance and they are rebuilding indexes this could take for ages.....

  • Yes, these are share point databases and we have log backups twice a day. It is in pre-production, later we will change the log backup schedule to every 1 hour.

    This error is because of not having frequent log backups?

  • No shouldn't be because of lack of backups...1 hours is good..or could do 15 minutes if they are very big logs to keep them down a bit when backing them up. Ask the sharepoint guys if they are doing search indexes and if they are on MOSS which i think is the latest version of sharepoint.

    Also ask them if they are doing rebuilds on the app side or setting up new search indexs on the apps.

    Thread i found looks interesting.

    http://sqlcat.com/technicalnotes/archive/2008/12/09/diagnosing-transaction-log-performance-issues-and-limits-of-the-log-manager.aspx

  • Hi,

    We have share point databases in sql server 2005 EE x64 and we do NOT have mirroing setup.

    and No Index rebuilds and Integrity checks happening. But Still we are frequently getting the high Log flush wait itme alarms? is this a critical to databases?

    how to rectify the cause of this alarm?

  • Put on profiler and capture what is running at this time.

  • I know this thread is old (very old but it may help someone else)

    But with a high log flush I would look at the autogrowth settings. SQL 2008 has sensable growth defaults but SQL 2000 and 2005 use 10% growth settings.

Viewing 8 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic. Login to reply