And also you can monitor by scheduling below command in cron. It will
collect the detailed data, so that we came to know where the connections
are coming.
[postgres@local~]$ crontab -l
* * * * * /opt/postgres/9.3/bin/psql -Aqt -p 5493 -c "select * from
pg_stat_activity;" >>/tmp/stats.csv
On Thu, Mar 27, 2014 at 10:20 PM, Merlin Moncure <mmonc...@gmail.com> wrote:
> On Wed, Mar 26, 2014 at 7:02 PM, David Johnston <pol...@yahoo.com> wrote:
> > Nithya Soman wrote
> >> Hi
> >>
> >> Could you please provide any method (query or any logfile) to check
> >> max connections happened during a time interval in psql DB ?
> >
> > Only if the time interval desired in basically zero-width (i.e.,
> > instantaneous). The "pg_stat_activity" view is your friend in this.
> >
> > You have numerous options, including self-coding, for capturing and
> > historically reviewing these snapshots and/or setting up monitoring on
> them.
> >
> > This presumes you are actually wondering "over any given time period how
> > many open connections were there"? If your question is actually "In the
> > given time period did any clients get rejected because {max connections}
> > were already in use." you can check the PostgreSQL logs for the relevant
> > error.
>
> There's also some useful high level statistics (including connection
> count) in pg_stat_database. For exact connection count over time
> frame, I'd turn on log_connections in postgresql.conf and grep the
> log.
>
> merlin
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general
>
