Dr NoName wrote:
Hi all,
I got another problem with postgres. This time it
refuses to use the indexes. Check this out:
If you would like help, first try posting queries and EXPLAIN ANALYZE dumps.
Also you may have better luck with the Performance list
([EMAIL PROTECTED]).
Best Wishes,
Chr
than others.
Sincerely,
Joshua D. Drake
Bob Pawley
- Original Message - From: "Jaime Casanova"
<[EMAIL PROTECTED]>
To: "Dr NoName" <[EMAIL PROTECTED]>
Cc: "Ragnar Hafstað" <[EMAIL PROTECTED]>;
Sent: Tuesday, August 02, 2005 12:57 PM
riginal Message -
From: "Jaime Casanova" <[EMAIL PROTECTED]>
To: "Dr NoName" <[EMAIL PROTECTED]>
Cc: "Ragnar Hafstað" <[EMAIL PROTECTED]>;
Sent: Tuesday, August 02, 2005 12:57 PM
Subject: Re: [GENERAL] indexes are fucked
On 8/2/05, Dr NoName <
On 8/2/05, Dr NoName <[EMAIL PROTECTED]> wrote:
> Thanks Ragnar. That solved the problem. I still would
> like some explanation about this voodoo. Most
> importantly, how can I estimage a "good" statistics
> number?
>
> thanks,
>
> Eugene
http://www.postgresql.org/docs/8.0/static/planner-stats.h
Thanks Ragnar. That solved the problem. I still would
like some explanation about this voodoo. Most
importantly, how can I estimage a "good" statistics
number?
thanks,
Eugene
--- Dr NoName <[EMAIL PROTECTED]> wrote:
> > the seqscan is cheaper when a large enough
> > proportion
> > (maybe 5%) of
> the seqscan is cheaper when a large enough
> proportion
> (maybe 5%) of rows are retrieved, and indeed the
> cost
> is estimated at 39014
very good explanation. thank you.
> try to increase statistics for this column:
>
> ALTER TABLE render ALTER COLUMN person_id
> SET STATISTICS 1000;
> AN
On Tue, 2005-08-02 at 10:50 -0700, Dr NoName wrote:
> > What is the output of these:
> >
> > set enable_seqscan = off;
> > explain SELECT render.* FROM render WHERE person_id
> > = 432;
>
>
> QUERY PLAN
>
This is really the only thing I can think to suggest;
Have you tried 'SET enable_seqscan TO OFF;' and then tried the query
again? This happens to me now and then where an index is a lot faster
but the planner just doesn't want to use it. I've got an option in my
code to turn off 'enable_seqsca
> What is the output of these:
>
> set enable_seqscan = off;
> explain SELECT render.* FROM render WHERE person_id
> = 432;
QUERY PLAN
--
On Tue, 2005-08-02 at 10:04 -0700, Dr NoName wrote:
> I got another problem with postgres. This time it
> refuses to use the indexes. Check this out:
[snip]
> siam_production=> explain SELECT render.* FROM render
> WHERE person_id = 432;
>QUERY PLAN
>
Hi all,
I got another problem with postgres. This time it
refuses to use the indexes. Check this out:
siam_production=> \d render
Table
"public.render"
Column|Type |
Modifiers
--
11 matches
Mail list logo
