Weird index problem - Mailing list pgsql-general
From | Ole Gjerde |
---|---|
Subject | Weird index problem |
Date | |
Msg-id | Pine.LNX.4.05.9907132215460.24176-100000@snowman.icebox.org Whole thread Raw |
In response to | Administration Wizards ... ("Keith R. Davis" <keidav@whidbey.com>) |
Responses |
More on Weird index problem
|
List | pgsql-general |
Hey, I'm having a very weird thing happening here. I've had this one table for quite some time, but what I haven't noticed until now is that only 3 out of 9 indexes seems to be working on it. OS: Red Hat Linux 6.0 / Linux 2.2.6 Arch: i386 Postgres version: CVS of 6.5 a few days before actual release Table = av_parts +----------------------------------+----------------------------------+-------+ | Field | Type |Length| +----------------------------------+----------------------------------+-------+ | itemid | int4 not null default nextval ( |4 | | vendorid | int4 |4 | | partnumber | varchar() |25 | | alternatepartnumber | varchar() |25 | | nsn | varchar() |15 | | description | varchar() |50 | | condition | varchar() |10 | | quantity | int4 |4 | | rawpartnumber | varchar() |25 | | rawalternatenumber | varchar() |25 | | rawnsnnumber | varchar() |15 | | date | int4 |4 | | cagecode | varchar() |10 | +----------------------------------+----------------------------------+-------+ Indices: av_parts_altpartnum_index av_parts_itemid_key av_parts_nsn_index av_parts_partnumber_index av_parts_rawalternatenumber_ind av_parts_rawaltnum_index av_parts_rawnsn_index ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ av_parts_rawpartnumber_index av_parts_vendorid_index This is the one I will use as an example. This is an index on rawnsnnumber (varchar 15). Output of a simple select on that field: parts=> explain select * from av_parts where rawnsnnumber = '123456'; NOTICE: QUERY PLAN: Seq Scan on av_parts (cost=194841.86 rows=3206928 width=124) EXPLAIN This doesn't really make sense. There is an index on that field, and I have just done a vacuum on the table. The index on partnumber, itemid and vendorid is being used properly, all others are not. Can someone explain this? Thanks, Ole Gjerde
pgsql-general by date: