From: | Doug McNaught <doug(at)mcnaught(dot)org> |
---|---|
To: | Stephen J Peck <peck(at)advancen(dot)com> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Backend start-up failed FATAL : non-superuser |
Date: | 2004-01-09 02:54:26 |
Message-ID: | 87llohygpp.fsf@asmodeus.mcnaught.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Stephen J Peck <peck(at)advancen(dot)com> writes:
> I am currently using PostgreSQL 7.3.2 on i386-pc-linux-gnu, compiled by
> GCC gcc (GCC) 3.2.3.
You should probably upgrade to 7.3.5, which fixes a couple of quite
nasty bugs in the version you're running. It's an iun-place
upgrade--no dump/reload required.
> I'm getting a 'Backend start-up failed FATAL : non-superuser connection
> limit exceeded' error and was wondering as too why this occurs.
>
> The only solution I have found to reset the connections is to actually
> restart Postgresql.
>
> Could this be that postgres is not releasing connections correctly, for
> it seems that it occurs only when there have been frequent restarts of
> the application we are working on.
It's much more likely that your app is not releasing the
connections--PG is pretty good about exiting the backend when its
connection is dropped.
When you get the "start-up failed" message, see if your application
has any processes or threads still running. Use 'netstat' and see how
many active connections there are on the ports/sockets used by
Postgres (you didn't mention whether you're using tcp/up or Unix socket
connections). Use 'lsof' (or equivalent) if need be to find out what
processes have these connections open.
> If someone could give me a quick run down of some of the possible causes
> of this error, or just a basic explanation of why it occurs I would
> really appreciate it.
The error message is perhaps not clear enough? Seems straightforward
to me...
-Doug
From | Date | Subject | |
---|---|---|---|
Next Message | D. Dante Lorenso | 2004-01-09 03:13:41 | Natural upgrade path for RedHat 9? |
Previous Message | Tom Lane | 2004-01-09 02:27:28 | Re: full db locking during dump |