RE: SQL*Net Message to client wait on batch job
- From: Wolfgang Breitling <breitliw@xxxxxxxxxxxxx>
- To: oracle-l@xxxxxxxxxxxxx, Thomas Biju <BThomas@xxxxxxxxxx>
- Date: Mon, 23 Aug 2004 08:33:10 -0700
Are you sure that the "sql*net wait to client" are the problem? I mean they ARE
a problem, but I can't see that they are responsible for the slowdown as they
would have been there in 8i as well, unless sqr and the 8i database were
running on the same host using two_task protocol and the Oracle 9i database is
now on a different host and you have network latency to deal with.
I noticed that the one select in you trace snipped took 13 seconds elapsed and
cpu, 177,000 consistent reads and 3 physical reads to fetch 3 rows. If that is
an isolated sql it won't be a concern (although 177,000 cr to get 3 rows seems
a bit excessive) but if the sqr is doing 10s or 100s of them they add up.
Compare the plan for this, and other sql, with the plans in 8i. There were lots
of changes in the 9i optimizer and some of them may bite you, particularly in a
Peoplesoft environment.
Quoting Thomas Biju <BThomas@xxxxxxxxxx>:
> Thank you. Unfortunately, this SQR was working good in Oracle8i and not so =
> good in Oracle9i. So it is "obviously" a database issue. :-)
--
regards
Wolfgang Breitling
Oracle 7,8,8i,9i OCP DBA
Centrex Consulting Corporation
www.centrexcc.com
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at //www.freelists.org/archives/oracle-l/
FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
Other related posts:
