First, this isn't a bug... this is more of a discussion for -general or
-admin.
On Wed, Apr 27, 2011 at 9:44 AM, Kevin Grittner <Kevin.Grittner@wicourts.gov
> wrote:
> "Suprabhat" <suprabhatm@gmail.com> wrote:
>
> > FATAL: too many connections already and there are too many
> > postgres processes on the server. We have used connection
> > pooling methods after unsuccessfully using pgpool2.
>
> Is there a thread somewhere where you were trying to get help
> configuring pgpool?
>
Agreed, pgpool is quite sensitive to being configured properly.
If you're using the pooler and you are still getting that error, it is
definitely a configuration issue.
>
> > Probably we are doing some basics wrong.
> >
> > Total number of connections we are trying to make is something
> > around 300 concurrently.
>
> Well, the main point of a connection pool is to allow multiple
> client-side processes or threads to see a large number of logical
> connections to the database while those funnel down to a small
> number on the database side. 300 active users might do quite well
> on a connection pool which maintains 10 or 20 connections to the
> database. It sounds like you haven't used the correct settings to
> achieve that.
>
>
In all honesty, if you're just looking for a connection pool, I personally
prefer pgbouncer. That being said, pgpool provides a lot of other features
(load-balancing, replication,etc...).
-Kevin
>
> --
> Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-bugs
>
