Mike Goldner <mgoldner(at)agmednet(dot)com> writes:
> First of all, my max_fsm_pages is obviously way off. However, every
> time I increase my max_fsm_pages the next vacuum says that it requires
> more. Will there every be a plateau in the requested pages?
We realized recently that this can happen if you have a single relation
that all by itself needs more FSM space than you've got; any per-table
space beyond max_fsm_pages is disregarded in the calculation of the
total system-wide space needed, and so you get this inflationary effect
where the reported requirement keeps going up when you increase the
setting. There's a fix in for 8.2.
> So, all indications point to postgres blocking all access during the
> entire vacuum.
You haven't said anything I consider credible evidence of that. But try
looking in pg_locks next time.
regards, tom lane
In response to
pgsql-admin by date
|Next:||From: Tom Lane||Date: 2006-10-25 22:49:48|
|Subject: Re: CLUSTER using more memory than expected |
|Previous:||From: Mike Goldner||Date: 2006-10-25 20:00:29|
|Subject: Re: Vacuum Problems (locking)|