From: | Doug McNaught <doug(at)mcnaught(dot)org> |
---|---|
To: | Andy Harrison <ah11(at)mlz(dot)us> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: postmaster processes |
Date: | 2003-09-04 16:52:07 |
Message-ID: | m3iso8ts20.fsf@varsoon.wireboard.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Andy Harrison <ah11(at)mlz(dot)us> writes:
> Could anyone explain about the max_connections in postgres?
>
> We ran into the connection limit on one of our production servers.
> One reason is that we have more apps that use postgres running on it
> (RT3 and Webcollab were added not long ago). Webcollab, for
> instance, seems to keep 7 postmaster processes all to itself. Is
> this normal? Does postgres work similarly to how apache handles
> spawning spare servers?
No, it doesn't. If a PG backend is hanging around, it means an
application is keeping it open. Most likely the app has a connection
pooling library that hangs onto connections rather than incur the
overhead of creating new ones for every request.
> I notice also on my test server, that 7 postmaster processes for
> webcollab are running, even though I'm the only one who has used it,
> and not all that heavily, and they haven't been touched for two
> days.
See above. The app is in control--see if it has settings to control
the connection pool.
-Doug
From | Date | Subject | |
---|---|---|---|
Next Message | Jonathan Bartlett | 2003-09-04 17:06:42 | Re: Replaceing records |
Previous Message | Adam Kavan | 2003-09-04 16:32:42 | Re: pg_autovacuum |