| From: | David Mitchell <david(dot)mitchell(at)telogis(dot)com> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Strange slow database |
| Date: | 2005-11-18 03:43:55 |
| Message-ID: | 437D4DFB.6000408@telogis.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Tom Lane wrote:
> I speculate that you did a VACUUM FULL on it recently.
You speculate right, we do a vacuum full every sunday night as a safety
net. So the vacuum full was a week old.
>
> The condition of the indexes suggests strongly that you've not been
> vacuuming pg_attribute often enough (and perhaps not any of the other
> system catalogs, either?). Heavy use of temp tables will cause
> pg_attribute and pg_class to bloat if you don't keep after them.
That's interesting. We stopped using temporary tables because they were
causing us lots of trouble, including table bloat. We vacuum
pg_attribute (and every other table with an entry in pg_tables) every
ten minutes. What other that temp tables could bloat pg_attribute? We
use refcursors to return data from our stored procedures. We also have a
few stored procedures that return SETOF.
Thanks for helping us out on this.
--
David Mitchell
Software Engineer
Telogis
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2005-11-18 04:45:07 | Re: Trouble with recursive trigger |
| Previous Message | Greg Stark | 2005-11-18 03:33:42 | Re: Partial foreign keys, check constraints and inheritance |