Look through the mailing list archives for a program called "reindexdb"
that will likely fix your problems. There was a whole thread on this not
more than a month or so ago.
Basically, large updates can cause uncontrolled index growth that vacuum
doesn't fix.
On Tue, 18 Jun 2002, Bjoern Metzdorf wrote:
> Hi,
>
> I have a 3 GB (fs based) large pgdata directory. I regularly do vacuums
> every 15 minutes and vacuums with analyzing every night.
>
> After dumping the whole db (pg_dump -c db), dropping and creating the db,
> reinserting the dump and vacuuming again, my pgdata directory only contains
> 1 GB. The dump had no errors, all data has been saved and reinserted.
>
> The xlogs/clogs didnt take up 2 GB, so I am wondering what has happened.
>
> Shouldn't the vacuuming take care of this?
>
> A (desired) sideeffect is, that the postmaster runs much faster now. Queries
> get executed much faster.
>
> If I compare the relpages from before and after, I see the difference there
> also.
>
> Any hints?
>
> Greetings,
> Bjoern
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 5: Have you checked our extensive FAQ?
>
> http://www.postgresql.org/users-lounge/docs/faq.html
>
--
"Force has no place where there is need of skill.", "Haste in every
business brings failures.", "This is the bitterest pain among men, to have
much knowledge but no power." -- Herodotus
