Re: Performance Killer 'IN' ?

[Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>]

Kai Hessing <kai(dot)hessing(at)hobsons(dot)de> writes:
> Index Scan using phon_phon_idx on phon  (cost=0.00..5193.83 rows=530
> width=148) (actual time=0.146..0.146 rows=0 loops=1)
> ...
> Seq Scan on phon  (cost=0.00..1573304.58 rows=105931 width=148) (actual
> time=369563.565..369563.565 rows=0 loops=1)

You need to look into the discrepancy between estimated and actual row
counts.  (I suppose the reason you're showing 0 rows here is that you
already did these UPDATEs and so none of the rows in question pass the
status filter --- but how many rows are there matching the phon index
conditions?)  Perhaps a larger statistics target for the phon column
would be a good idea.

			regards, tom lane