Re: The Oracle 9.2.0.4 Optimizer - are there problems with it?

  • From: "Juan Cachito Reyes Pacheco" <jreyes@xxxxxxxxxxxxxxxx>
  • To: <oracle-l@xxxxxxxxxxxxx>
  • Date: Fri, 2 Apr 2004 18:09:12 -0400

Did you tried setting OPTIMIZER_FEATURES_ENABLE
to 9.2.0.3


Juan Carlos Reyes Pacheco
OCP
Database 9.2 Standard Edition
----- Original Message ----- 
From: "Powell, Mark D" <mark.powell@xxxxxxx>
To: <oracle-l@xxxxxxxxxxxxx>
Sent: Friday, April 02, 2004 5:52 PM
Subject: RE: The Oracle 9.2.0.4 Optimizer - are there problems with it?


Anytime you upgrade Oracle a fix or modification to the CBO may have an
undesirable side effect on one or more queries.  Depending on when the
object statistics were last updated it may just be time to update the
statistics, or it might just be time to sit down and rework a couple of
queries to take advantage of a couple of advances in Oracle technology.  So
why not just find and tune the queries in question?

If it is in fact possible to set the database that far back you are liable
to end up with a new set of customers complaining that their queries now run
slower or that some piece of code broke because it was using a version 9
feature made unusable by setting compatibility back.  In the long run it is
probably best to at least attempt to modify the SQL first.  There were
around 8 to 12 underbar parameters changed from a default of off to being on
by 9.2 and I have seen a few posts on metalink about setting one of these
specific parameters to solve problems with specific types of queries.  But
tracking the SQL to one of these problems will first require you attempt to
tune it.

Good luck with whatever choice you make.


-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx]On Behalf Of Wales, Stephen (RTSI)
Sent: Friday, April 02, 2004 4:10 PM
To: oracle-l@xxxxxxxxxxxxx
Subject: The Oracle 9.2.0.4 Optimizer - are there problems with it?


We recently (2 weeks ago) threw the 9.2.0.4 patch onto our production =
9.2.0.2 databases.

Since that time, users have been complaining about long running queries =
are running even longer.  Now I'm sure that part of that is poor query =
structure (I've analyzed and tuned 2 of them this week already) but the =
users want to know why their queries are going so slow...

I've seen some notes on Metalink referencing a parameter =
optimizer_features_enable and talk of setting it to 8.1.7 - is this =
worth doing?  Anyone got any input on this?  Or had similar problems?  =
Or know of a solution?

Would appreciate any advice.

Thanks
Steve
----------------------------------------------------------------
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
-----------------------------------------------------------------
----------------------------------------------------------------
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
-----------------------------------------------------------------


----------------------------------------------------------------
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: