Re: is pg_autovacuum so effective ?

Gaetano Mendola <mendola(at)bigfoot(dot)com> writes:
> Matthew T. O'Connor wrote:
>
>> Well without thinking too much, I would first ask about your FSM
>> settings? If they aren't big enought that will cause bloat. Try
>> bumping your FSM settings and then see if you reach steady state.
>
> FSM settings are big enough:
>
> max_fsm_pages | 2000000
> max_fsm_relations | 1000
>
> at least after a vacuum full I see that these numbers are an overkill...

When you do a VACUUM FULL, the FSM is made irrelevant because VACUUM
FULL takes the time to reclaim all possible space without resorting to
_any_ use of the FSM.

If you VACUUM FULL, then it's of little value to bother having a free
space map because you're obviating the need to use it.

In any case, the FSM figures you get out of a VACUUM are only really
meaningful if you're moving towards the "equilibrium point" where the
FSM is large enough to cope with the growth between VACUUM cycles.
VACUUM FULL pushes the system away from equilibrium, thereby making
FSM estimates less useful.
--
"cbbrowne","@","ca.afilias.info"
<http://dev6.int.libertyrms.com/>
Christopher Browne
(416) 673-4124 (land)