OK, I posted the current schema and before and after EXPLAIN output.
The problem query is "select * from vtb_allocpayroll". It is messy, but works fine until I analyze things...
http://www.pakrat.com/pgtrouble
I will look into the pg_stats thing I saw mentioned, but am not sure if it applies since I am way back at 7.1.3.
planning=# select version();
version
--------------------------------------------------------------
PostgreSQL 7.1.3 on i386--netbsd, compiled by GCC egcs-1.1.2
(1 row)
Ian A. Harding
Programmer/Analyst II
Tacoma-Pierce County Health Department
(253) 798-3549
mailto: iharding@tpchd.org
>>> Neil Conway <nconway@klamath.dyndns.org> 05/17/02 04:08PM >>>
On Fri, 17 May 2002 10:06:32 -0700
"Ian Harding" <ianh@tpchd.org> wrote:
> I can forward my schema and another example to the group if anyone wants.
Yes, I'd be interested in seeing that, as well as the output of
EXPLAIN ANALYZE for both the good and bad query plans.
Cheers,
Neil
--
Neil Conway <neilconway@rogers.com>
PGP Key ID: DB3C29FC