Re: tkprof output

Repeat but changing the 12 to a zero should work. You will find, however, that any cursor that has not closed when you stop tracing will not dump its 'STAT' lines (including execution plan) to the trace file. Regards Jonathan Lewis http://www.jlcomp.demon.co.uk Now available One-day tutorials:

RE: tkprof output

sys.dbms_system.set_ev( v_seid, v_sernum, 10046, 0, '' ) ^ Cary Millsap Hotsos Enterprises, Ltd. http://www.hotsos.com Upcoming events: - Hotsos Clinic 101, Mar 25-27 Oxford - Hotsos Clinic 101, Apr 8-10 Chicago -Original Message- Sent:

Re: tkprof output

sys.dbms_system.set_ev( v_seid ,v_sernum ,10046, 0,'') Igor Neyman, OCP DBA [EMAIL PROTECTED] - Original Message - To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]> Sent: Thursday, March 13, 2003 10:49 AM > I use sys.dbms_system.set_ev( v_seid ,v_sernum ,10046,12 ,'') to

Re: tkprof output

also how can we check if one particular session is being traced (10046 ) or not . Basically I want to be sure that tracing is stopped for the session and its not fillling up disk space . -ak - Original Message - To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]> Sent: Wed

Re: tkprof output

I use sys.dbms_system.set_ev( v_seid ,v_sernum ,10046,12 ,'') to start tracing on 8.1.6 db . What should I do to stop tracing without exiting out of session . Thanks, -ak - Original Message - To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]> Sent: Wednesday, March 12, 2003 9

Re: tkprof output

Thanks Jonathan for the information . Let me go through raw -trace file once more to get info . Everytime I go through it I find something interesting . -ak - Original Message - To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]> Sent: Wednesday, March 12, 2003 8:54 AM > > Th

RE: tkprof output

An action with dep=n+1 (n>=0) for cursor #k is the recursive child of the next dep=n action for cursor #k that immediately follows in the trace data. Cary Millsap Hotsos Enterprises, Ltd. http://www.hotsos.com Upcoming events: - Hotsos Clinic 101, Mar 25-27 Oxford - Hotsos Clinic 101, Apr 8-10

Re: tkprof output

The examples you have given are SYS-recursive, The call to cdef$ is Oracle looking for some information about constraints (one possibility is that you keep breaking a PK or UK constraint and Oracle has to keep looking up the name of the constraint because it doesn't cache constraint names). The

Re: tkprof output

Thanks Jonathan, what is meaning of recursive depth ? I see calls to cdef$, seq$ tables/views does it hint something . I though procedure is using some sequence and these are internal calls to generate seq numbers . Is that rite ? -ak - Original Message - To: "Multiple recipients of l

Re: tkprof output

Any SQL within a pl/sql block is recursive SQL (user recursive, rather than SYS recursive) so this time could simply be the cost of running your application code. Unfortunately the tkprof output doesn't quote the recursive depth of the SQL - however if you identify possible suspects, you can chec

Re: tkprof output

Using sys=yes on the tkprof command line would be a good start. That way you will see which statements are generating all the recursive SQL. Jared "AK" <[EMAIL PROTECTED]> Sent by: [EMAIL PROTECTED] 03/11/2003 11:29 AM Please respond to ORACLE-L To: Multiple recipients of

Re: tkprof output

Run tkprof with "SYS = YES" option and find recursive calls that contribute to this time. Igor Neyman, OCP DBA [EMAIL PROTECTED] - Original Message - To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]> Sent: Tuesday, March 11, 2003 2:29 PM > I am running tkprof on event 100

RE: Tkprof output

Multiple recipients of list ORACLE-L > Subject: RE: Tkprof output > > Paul (or anyone) - > Is there a way to set the level within the session as: >alter session set sql_trace = true > Or do I have to start the session, start a SQLPLUS session, find th

RE: Tkprof output

Paul (or anyone) - Is there a way to set the level within the session as: alter session set sql_trace = true Or do I have to start the session, start a SQLPLUS session, find the session_id and run the procedure to set it on another session? Dennis Williams DBA Lifeto

RE: Tkprof output

Set you 10046 trace at level 4 or 12, and then check the trace file. Paul -Original Message- Sent: Friday, February 22, 2002 3:53 PM To: Multiple recipients of list ORACLE-L Hi Listers, I got the following statement in the tkprof output file : update EMP set "ENAME"=:V001,"EMPNO"=:V

RE: Tkprof output

Title: RE: Tkprof output A huge thanks for all those who took the time out to respond to my problem. The query ran to completion yesterday and (as I said to Lisa) I was later told that I was not to worry as it was a one off migration (well that only wasted about 1- 2 hours of my time

RE: Tkprof output

Title: Message Take a look at using array fetches to retrieve rows more efficeintly.  It appears you are fetching 1 row at a time.  Which can be grossly inefficient. Another helpful thing would to generate the TKPROF with Explan plans so you can see the rows per step and the paths chosen.  

RE: Tkprof output

Title: RE: Tkprof output Also note the very high "query" number (i.e. buffers gotten for consistent read).  That could account for a lot of the i/o, which is the proximate cause of the lng elapsed time.  From the (to use one of Lisa's favorite terms) doco (for 9i): Re

RE: Tkprof output

Are you on 8i? Then you can use BULK binding which will definitely give you some speed. IF you need example ... let me know. Raj __ Rajendra Jamadagni MIS, ESPN Inc. Rajendra dot Jamadagni at ESPN dot com Any opinion expressed h

RE: Tkprof output

Title: RE: Tkprof output Lee, This query seems suspect UPDATE VM_LIVE.SINGLE_CUSTOMER_HISTORY SCH SET VISIBLE=1 WHERE  ACXIOM_CUSTOMER_KEY = :b1  AND VERSION_NO = :b2 because of this call count   cpu    elapsed   disk  query    current    rows