Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> Re: cursor #0

Re: cursor #0

From: Egor Starostin <egorst_at_gmail.com>
Date: Thu, 1 Sep 2005 13:22:40 +0700
Message-ID: <f0fc919705083123221b8a755@mail.gmail.com>


> >There are no PARSING IN CURSOR #0 lines in the file...so what's cursor #0?
>
> Cursor #0 is used when no cursor number is available. There are two common situations for that:
>
> 1) When a commit or rollback is executed. In fact for these operations you will find a XCTEND line where no cursor number is available...
>
> 2) At the beginning of the trace file where no statement has been parsed yet.
I think that even if 'WAIT' line is printed before 'PARSING IN CURSOR' line, it should still have nonzero #<CURSOR>.

There is one more case for 'WAIT #0' lines (in addition to Niall's case of PL/SQL cursor): waits for RPC calls goes to trace file as 'WAIT #0' (probably because of 'RPC CALL'/'RPC EXEC' lines doesn't have #<CURSOR> attribute).

Paul, do you have 'WAIT #0' lines before 'PARSING IN CURSOR #13' line? And in your fragment what lines are after 'WAIT #0' lines?

-- 
Egor
http://www.oracledba.ru/orasrp/
Free Oracle Session Resource Profiler
--
http://www.freelists.org/webpage/oracle-l
Received on Thu Sep 01 2005 - 01:24:41 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US