From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Markus Wollny" <Markus(dot)Wollny(at)computec(dot)de> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: One source of constant annoyance identified |
Date: | 2002-06-27 14:40:48 |
Message-ID: | 25540.1025188848@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Markus Wollny" <Markus(dot)Wollny(at)computec(dot)de> writes:
> [ turned off persistent connections ]
> Now there's still the odd 250MB backend lingering around for some time,
> but it's not four or five of them any more, wich is a big gain when
> there's 250MB swap around more often than not.
So you haven't really solved the problem --- somewhere there is a query
being issued that ramps the backend up to a lot of memory. All you've
done is ensured that the backend won't hang around very long. The
persistent connection isn't really at fault, except in that it causes
backends to keep being used after their memory usage has become bloated.
Although this might be enough to eliminate your immediate performance
problem, you should keep looking. Starting lots more backends than you
need to is a performance hit, so turning off persistent connections is
really only a stopgap not a desirable answer. And I'm still wondering
whether you've exposed a fixable memory leak bug. We need to identify
exactly what query is causing the backends to eat memory.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Gregory Wood | 2002-06-27 15:05:53 | Re: pg_dump / consistent snapshot / backup |
Previous Message | Curt Sampson | 2002-06-27 14:35:44 | Re: Shared Memory Sizing |