Default Trace

  • For how many days does the default trace files be available at the default folder.

    If we need information from a much older date is it possible.

    Please share your inputs.

    M&M

  • The default trace keeps 5 files, and they roll over at about 20MB. This means the time frame that is available varies according to how many transactions you have going on.

  • Adam Angelini (6/17/2010)


    The default trace keeps 5 files, and they roll over at about 20MB. This means the time frame that is available varies according to how many transactions you have going on.

    Thank you very much Adam

    M&M

  • Today, I got to know of the black box trace too which can be configured in SQL Server. Infact, it is a feature that is available in SQL Server 2000 itself. Please find some related links below.

    http://blogs.msdn.com/b/sqlserverstorageengine/archive/2007/06/03/sql-server-s-black-box.aspx

    http://www.sqlservercentral.com/articles/Administration/blackbox/954/

    Both these default and black box trace captures different set of events. Would it be enough to analyze

    performance issues on SQL Server with these trace files itself.

    Now, with both the default trace and the black box trace enabled, what is the use of running profiler trace

    when there is a issue.

    M&M

  • default trace and blackbox trace are recording specific events- they are basically the same as running a profiler trace, but without the flexibility of specifying exactly which events you want to track and filtering those events based on certain values (i.e. where databasename= thedatabaseIcareabout).

    You can use profiler to capture more or less events that the default trace does in order to retreive exactly the information you are looking for.

    Maybe tell us about the problem you are investigating- maybe we can point you in the right direction.

  • Thansk NJ.

    On some servers which I support, there are several cases where we need to do RCA (root cause analysis) when there is performance issue like slowness etc.

    Usually during these times, our partners request us to run profiler trace and share with them the information. But instead, I thought, I will configure the black box trace and the default trace on those servers and share the trace files or query the trace files to assist on performance issues.

    I was thinking of using the black box and default trace in place of the profiler trace.

    As both these black box and default collect different events, I thought it is more than enough to have these traces enabled and use them for our analysis.

    Do you agree?

    M&M

Viewing 6 posts - 1 through 5 (of 5 total)

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