Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: The init.ora parameter CURSOR_SPACE_FOR_TIME

Re: The init.ora parameter CURSOR_SPACE_FOR_TIME

From: Vimal <vthangavel_at_lucent.com>
Date: 8 Sep 1999 13:36:06 GMT
Message-ID: <01bef9ff$1daff4c0$f0461287@ga3699_vthangavel.lucent.com>


Joe,

Please set to CURSOR_SPACE_FOR_TIME to FALSE in the init.ora file.

It has to be set to TRUE only if there are no cache misses ( i.e. if the reloads value is zero ).

Vimal.

Joseph Li <lijoecyn_at_netspace.net.au> wrote in article <37D763B5.1E76_at_netspace.net.au>...
> The platform is SUN Sparc Solaris 2.6, version of Oracle is 7.3.4.0.1 .
> I am running an application which demands CURSOR_SPACE_FOR_TIME be set
> to TRUE, or it could be set to TRUE by the previous DBA. The instance
> has been up since July 27 1999. I look at the pins and reloads values
> in V$LIBRARYCACHE; sum(pins) is 38855892, sum(reloads) is 486399, so
> sum(reloads) is more than 1% of sum(pins). The reloads came from the
> SQL AREA, TABLE/PROCEDURE, BODY namespaces, mostly from SQL AREA.
> Reading the tuning guide, I came to the conclusion that if
> CURSOR_SPACE_FOR_TIME is TRUE then there should be ZERO reloads, so I
> must be missing something. Can somebody offer a explanation ?
>
Received on Wed Sep 08 1999 - 08:36:06 CDT

Original text of this message

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