I believe SQL Doctor uses server side Traces to get information (at least I think it does). Is it possibly that the size of information it is trying to collect can cause the trace file to grow and “crash”.
I have 1 TB of data across server databases on a SQL 2012 Enterprise box. Every time I use SQL Doctor on it, I get a —
Trace ID ‘4’ was stopped because of an error. Cause: Stopping the trace because the current trace file is full and the rollover option is not specified.. Restart the trace after correcting the problem
Can the code be re-written to accommodate for growth and multiple files on this trace?
You’re right. During some portions of analysis we do run a trace on the workload. That error message would suggest that we don’t have trace rollover set for these traces, though, so they could fill up and only show us a narrow portion of the overall workload.
I’ll take this in to Support for further investigation. If we make a change in the code that accommodates multiple rollover files, we can update you when a release that includes that comes out.
The nice thing is it doesn’t crash the application, it just doesn’t give a “complete” picture of the server.
Thank you sir!
Has there been any movement on this trace file issue?
If the error occurs, what does that mean for the data collected and its analysis? Does it still give the complete picture?
Thanks in advance…
yup, still waiting on a possible solution for this problem.
I guess this “functional” request is not high on the priority list? That’s a shame, because as it is pretty much useless with large databases/servers.
Any progress on this “bug” in SQL Doctor?
Hi Robert, Remind me, you have licenses of DM as well, correct?
I will try to trace down where this landed and update you. Do you happen to have a support case open on the issue I can reference?
yes, I do (one license) and no I haven’t opened a support ticket yet.
OK, this was opened as an enhancement request SQL Doctor as it turns out.
“The trace in SQL doctor is designed for use over a limited period of time, so it wasn’t originally configured to roll over. It’s sound like some environments are busy enough that that may be necessary. ”
Are you using SQL Doctor with more than one instance? If you want you can shoot me contact information on e-mail and I can talk to you about some roadmap items that might interest you.
Scott, I know we recently talked and discussed some “roadmap items”. Just curious, do you see SQL Doctor getting updated with this TRACE FILE fix or do I wait on the “roadmap items”.
BTW, I have opened a case (00199164) and it is in “unsupported SQL” category and the status shows “deferred development”
It will be reviewed as an enhancement for a future SQL Doctor release. Sorry to be so wishy-washy, but we want to have a release that overhauls and updates all recommendations. So timing will be determined by when we can schedule the research effort.
So depending on the timing and scope of that release, we may be able add it.