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: A Tale of Two SQLs

RE: A Tale of Two SQLs

From: Poras, Henry R. <Henry_Poras_at_dfci.harvard.edu>
Date: Mon, 3 May 2004 18:08:47 -0400
Message-ID: <0D3AAB332E29BA4F9D8BE559F244A914558B2B@phsexch27.mgh.harvard.edu>


Dan,

I agree that the work appears to be going into creating consistent reads. Is your SQL just a SELECT? Any DML (updates?). Any triggers? I'm thinking 'write consistency'. Tom Kyte gave a really good talk on this at IOUG. If you do a long running UPDATE and another session updates and commits a row you haven't reached, you rollback and start the update again. CR obviously increase even without much activity on the table. My tip today is to check out http://asktom.oracle.com/pls/ask/f?p=4950:8:16912210699952507311::NO::F4950_P8_D ISPLAYID,F4950_P8_CRITERIA:11504247549852, Henry

-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org]On Behalf Of Tim Gorman Sent: Monday, May 03, 2004 4:38 PM
To: oracle-l_at_freelists.org
Subject: Re: A Tale of Two SQLs

Dan,

I think that you've hit upon the only possible explanation, especially regarding your comment that "eventually it will get a 1555 error". The flip-side of ORA-01555 is more work spent rebuilding a consistent read...

Hope that the support folks return the right answer... :-)

-Tim  

on 5/3/04 2:05 PM, Daniel Fink at Daniel.Fink_at_Sun.COM wrote:

> It was the best of sql executions; it was the worst of sql
> executions.
>
> We have a process that periodically slows down. So far, we have
> not been able to pinpoint a root cause, but we continue to
> pester the development/application support folks for more info.
> I apologize for the limited information, but I'm hoping that we
> are aiming in the right direction.
>
> We are smart enough that we captured a 10046 trace during a good
> execution as a baseline. I have tagged the lines from this trace
> file as (good). This morning, the query began slowing down
> (eventually it will get a 1555 error). We turned on tracing and
> those lines are tagged as (bad).



Please see the official ORACLE-L FAQ: http://www.orafaq.com

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.
--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to:  oracle-l-request_at_freelists.org
put 'unsubscribe' in the subject line.
--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
Received on Mon May 03 2004 - 17:33:01 CDT

Original text of this message

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