RE: 10046 not tracing everything?
- From: "Schultz, Charles" <sac@xxxxxxxxxxxxx>
- To: <oracle-l@xxxxxxxxxxxxx>
- Date: Tue, 20 Jun 2006 09:26:30 -0500
Never mind, we just realized that a completely independent process
filled up the fs mount, hence the trace file stopped before the sql was
ever submitted.
Sorry for the wasted bandwidth.
Goes to show that one can look in a lot of "good" places, but it might
be something else completely. =)
> _____________________________________________
> From: Schultz, Charles
> Sent: Tuesday, June 20, 2006 9:10 AM
> To: oracle-l@xxxxxxxxxxxxx
> Subject: 10046 not tracing everything?
>
> Good day, all,
>
> We are trying to isolate a query that is assigned a bad query plan,
> but when we turn on event 10046, the sql is not captured. I have
> verified that the session (and serial#) are valid, show up in
> dba_hist_active_session_history and that the event is triggered for
> this specific session. But even though the sql id (and corresponding
> plan hash value) are in dba_hist_active_session_history, they are not
> in the trace file. Under what circumstances will event 10046 not trace
> a query? I browsed a few metalink notes, but was not able to find one
> that listed any exceptions. Not that I think it matters much, but the
> query is not "system" query (ie, gathering dynamic stats), but a good
> old application query.
>
> charles schultz
> oracle dba
> aits - adsd
> university of illinois
>
>
Other related posts:
