Re: PostgreSQL doesn't use indexes even is enable_seqscan

From: Stephan Szabo <sszabo(at)megazone23(dot)bigpanda(dot)com>
To: Hans-Juergen Schoenig <hs(at)cybertec(dot)at>
Cc: <pgsql-general(at)postgresql(dot)org>
Subject: Re: PostgreSQL doesn't use indexes even is enable_seqscan
Date: 2002-07-08 16:10:23
Message-ID: 20020708090803.S27154-100000@megazone23.bigpanda.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Sun, 7 Jul 2002, Hans-Juergen Schoenig wrote:

> I have a severe problem with PostgreSQL 7.2.1.
> I have a table containing 500mio records (for testing purposes).
>
> I have indexed the table:
>
> CREATE UNIQUE INDEX idx_one_id ON one(id);
> CREATE INDEX idx_one_xmod ON one(xmod);
>
> The index was created properly but somehow it isn't used:
>
> cluster=# \d one
> Table "one"
> Column | Type | Modifiers
> --------+---------+-----------
> id | bigint |
> even | boolean |
> xmod | integer |
> Indexes: idx_one_xmod
> Unique keys: idx_one_id
>
> cluster=# SET enable_seqscan TO off;
> SET VARIABLE
> cluster=# SELECT * FROM one WHERE id=300000;

You'll need to quote or explicitly cast the
constant to bigint for the indexes to get used.
It's a problem with the way numeric constants are
handled. They'll get prematurely forced to int4
if they fit, and won't promote for index use.

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message terry 2002-07-08 16:10:24 Re: Bug in 7.2.1? -- ERROR: Adding columns with defaults is not implemented.
Previous Message Peter Alberer 2002-07-08 16:10:05 Speeding up subselect ?