From: | Jonathan Vanasco <postgres(at)2xlp(dot)com> |
---|---|
To: | Ragnar <gnari(at)hive(dot)is> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: queries stop using indexes |
Date: | 2007-04-02 14:23:04 |
Message-ID: | 9753DF7D-5B36-4CB4-9EC3-FFB0FC3CB432@2xlp.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Apr 2, 2007, at 7:40 AM, Ragnar wrote:
> if you analyze the table after recreating the index, does it then stop
> using it?
>
> possibly you need to increase statistic targets on some or all
> of the index columns.
>
> a EXPLAIN ANALYZE could help figuring this out.
no, analyzing the virgin index is fine.
heavy use on the index just seems to convince the planner to not use
it -- and no amount of analyze or vaccum analyze will change that.
i'll post some verbose debugging on the queries and analyze the next
time the index blows
// Jonathan Vanasco
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
| SyndiClick.com
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
| FindMeOn.com - The cure for Multiple Web Personality Disorder
| Web Identity Management and 3D Social Networking
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
| RoadSound.com - Tools For Bands, Stuff For Fans
| Collaborative Online Management And Syndication Tools
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
From | Date | Subject | |
---|---|---|---|
Next Message | Carlos Oliva | 2007-04-02 14:32:43 | Re: Unexplained case insensitive results |
Previous Message | Jaime Silvela | 2007-04-02 14:21:30 | Re: time series query |