Re: Index not used for simple query, and yes I ran vacuum analyze

On Wed, 26 Sep 2001 14:08:28 -0400, Tom Lane <tgl@sss.pgh.pa.us> said:
>
> Matthew Braithwaite <mab-lists@braithwaite.net> writes:
>> I RTFM but I'm still confused.  I have a table, headers
>>> mab=> \d headers
>>> Table "headers"
>>> Attribute |  Type   |                        Modifier
>>> -----------+---------+---------------------------------------------------------
>>> header_id | integer | not null default nextval('headers_header_id_seq'::text)
>>> part_id   | integer | not null
>>> key       | text    |
>>> value     | text    |
>>> Index: headers_pkey
>
>> I have indexed the table by part_id:
>
>>> mab=> \d headers_ref_idx
>>> Index "headers_ref_idx"
>>> Attribute |  Type
>>> -----------+---------
>>> part_id   | integer
>>> btree
>
> Um, it sure looks like that index is not on that table.

Whoof.  I think you are right; I had erroneously used that index name
in two places in my schema.  I have it working now; next time I'll
post to pgsql-idiot. :-)

> Note that the "\d headers" output shows only one index, named
> headers_pkey.

Oh!  I didn't know that all of a table's indices were supposed to show
up there.  Thanks, that's handy to know.