Tom Lane [tgl(at)sss(dot)pgh(dot)pa(dot)us] wrote:
> This seems like kind of a stupid plan anyway (which PG version was this
> exactly?) but certainly the big issue is the catastrophically bad
> rowcount estimate for the indexscan. Do you have ANALYZE stats for
> ts_stats_transetgroup_user_daily at all (look in pg_stats)?
postgres 8.3.5. Yes, here's a count(*) from pg_stats:
cemdb=# select count(*) from pg_stats where
tablename='ts_stats_transetgroup_user_daily';
count
-------
186
(1 row)
Thanks,
Brian