Re: Oracle 9i-10g init.ora parameters

From: joel garry <>
Date: Thu, 30 Oct 2008 11:00:51 -0700 (PDT)
Message-ID: <>

On Oct 30, 3:33am, raja <> wrote:
> Hi,
> I have a database, which was recently migrated from 9i to 10g.
> Current Version :
> I checked some of the init.ora parameters in current 10g database.
> They were as follows :
> optimizer_mode : choose
> statistics_level : typical
> timed_statistics : TRUE
> From the above init.ora parameter values, i found that they looked to
> be the values of 9i version.
> My doubt is :
> 1. If we migrate from 9i to 10g, whether the init.ora parameters will
> set according to the default values of 10g automatically
> (or)
> will we have to change these settings manually ?
> 2. What does this mean, that a 10g database (, still having
> the default values of 9i init.ora parameter values ?
> 3. Also, please add the comparison of the default init.ora parameters
> between oracle 9i and 10g ?
> Please clarify my doubts.
> Thanks in Advance.
> With Regards,
> Raja.


Since then, our enterprise app vendor suggested setting _OPTIM_PEEK_USER_BINDS=FALSE as a magic bullet to solve a problem that has nothing to do with Oracle, but it made no apparent different in my test system (though I can envision some situations where it may benefit, and perhaps more where it is not what we want). The problem program needed an index in the part of the program where it does talk to Oracle, a different problem that was magnified, perhaps because optimizer_index_caching was no longer set, perhaps because the hardware is different (but the storage is the same), perhaps there are some different issues when full scanning a table in the SGA in 10g, perhaps because the data added since conversion is not physically distributing the same way, still don't know exactly why (and tracing wasn't entirely helpful), etc...

DO NOT use the same settings in the init.ora, in case it isn't clear.


-- is bogus.
"I came. I saw. I concurred."  Irvine H. Page
Received on Thu Oct 30 2008 - 13:00:51 CDT

Original text of this message