Mime-Version: 1.0 (Apple Message framework v622)
In-Reply-To: <4574.1114030270@sss.pgh.pa.us>
References: <6f4567d4e64297b356e2d71ec392cdae@richardplotkin.com>
	<4574.1114030270@sss.pgh.pa.us>
Content-Type: text/plain; charset=US-ASCII; format=flowed
Message-Id: <b155378a669f04e74f597d4b105c2a10@richardplotkin.com>
Content-Transfer-Encoding: 7bit
From: Richard Plotkin <richard@richardplotkin.com>
Subject: Re: Disk filling, CPU filling, renegade inserts and deletes? 
Date: Wed, 20 Apr 2005 14:06:03 -0700
To: pgsql-performance@postgresql.org

Hi Tom,

> Q: what have you got the FSM parameters set to?

Here's from postgresql.conf -- FSM at default settings.
# - Memory -

shared_buffers = 30400          # min 16, at least max_connections*2, 
8KB each
work_mem = 32168                # min 64, size in KB
#maintenance_work_mem = 16384   # min 1024, size in KB
#max_stack_depth = 2048         # min 100, size in KB

# - Free Space Map -

#max_fsm_pages = 20000          # min max_fsm_relations*16, 6 bytes each
#max_fsm_relations = 1000       # min 100, ~50 bytes each

# - Kernel Resource Usage -

max_files_per_process = 750     #1000   # min 25
#preload_libraries = ''


> Q: what exactly is bloating?  Without knowing which tables or indexes
> are growing, it's hard to speculate about the exact causes.  Use du and
> oid2name, or look at pg_class.relpages after a plain VACUUM.

This I do not know.  I've disabled the cron jobs and will let the 
system bloat, then I will gather statistics (I'll give it 12-24 hours).

> It's likely that the real answer is "you need to vacuum more often
> than every six hours", but I'm trying not to jump to conclusions.

That could be it, except that I would expect the problem to then look 
more like a gradual increase in CPU usage and a gradual increase in use 
of disk space.  Mine could be an invalid assumption, but the system 
here looks like it goes from no problem to 100% problem within a 
minute.

Thanks again!
Richard