Re: EXPLAIN PLAN : better with RULE!

From: Telemachus <telemachus_at_ulysseswillreturn.net>
Date: Wed, 24 Jul 2002 10:42:55 +0100
Message-ID: <AEu%8.4476$zX3.3794@news.indigo.ie>

There's an old guy somewhere who lives behind a waterfall somewhere in Patagonia
who's can fully interpret 10053 trace.

Evdbt.com has a good starter for it ... but oracle support are usually reticent about giving details on the biblical interpretation of the scripture.

"Connor McDonald" <connor_mcdonald_at_yahoo.com> wrote in message news:3D3DC4B6.32B9_at_yahoo.com...
> Steve Mitchell wrote:
> >
> > "Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote in message
> > news:1027158123.29738.1.nnrp-14.9e984b29_at_news.demon.co.uk...
> > >
> > > Rather than just fiddling with parameters though, the
> > > next thing to do is look at the indexes and columns
> > > that the RBO has used in its plan, and examine the
> > > statistics stored in user_indexes, user_tab_columns,
> > > and user_histograms to see why CBO doesn't like
> > > them. Also check the stats against the real data
> > > to see if the stats are misleading.
> > >
> >
> > Great idea,
> >
> > Do you know where I might find some reference that would help me
> > understand the statistics stored in user_indexes, user_tab_columns, and
> > user_histograms? I would like to better understand how these
> > fit together to get an idea of why the CBO is ignoring indexes that are
> > obviously a better choice.
> >
> > > --
> > > Jonathan Lewis
> > > http://www.jlcomp.demon.co.uk
> > >
> >
> > --steve

>

> Once you've toured the manuals, the next logical place is a 10053 trace
> - which gives the CBO in most of its gorey detail...

>

> hth
> connor
> --
> ==============================
> Connor McDonald

>

> http://www.oracledba.co.uk

>
> "Some days you're the pigeon, some days you're the statue..."
Received on Wed Jul 24 2002 - 04:42:55 CDT