Re: DB CPU is much lower than CPU Time Reported by TOP SQL consumers

One of my points was that you cannot trust the module information to be correct; unless you KNOW that each statement in the top SQL can only be called from EXACTLY one module, you may be misinterpreting the stats. Bear in mind that if SQL calls PL/SQL which calls SQL, you won't see the pl/sql in the statspack report, so the fact that you can only see one pl/sql package call doesn't mean that's the only one causing a problem.

Given my comment about upper case and naming conventions in bind variables I would also like to have seen some comment from you to the effect that you could ignore the double-counting because all the other SQL statements failed to match the pl/sql requirements.

Regards

Jonathan Lewis
http://jonathanlewis.wordpress.com/all-postings

Author: Oracle Core (Apress 2011)
http://www.apress.com/9781430239543

• Original Message ----- From: "Ls Cheng" <exriscer_at_gmail.com> To: <jonathan_at_jlcomp.demon.co.uk> Cc: "Oracle Mailinglist" <oracle-l_at_freelists.org> Sent: Wednesday, November 14, 2012 9:49 AM Subject: Re: DB CPU is much lower than CPU Time Reported by TOP SQL consumers

| Hi
|
| I guess this isnt double counting because the top 1 SQL is a pl/sql
package
| running from JDBC Thin Client, I have the source code and some sql
| stataments are from this package but other top SQL statements are not
| called from this package, also the module for other top SQL statements
are
| from C programs.
|
| Adding up the top sql statements CPU usage not from the package is much
| higher than DB CPU.
|
| TIA
|
|
|
| -----
| No virus found in this message.
| Checked by AVG - www.avg.com
| Version: 2012.0.2221 / Virus Database: 2441/5393 - Release Date: 11/13/12

--
http://www.freelists.org/webpage/oracle-l