Re: Regarding the Spikes in the connection pool on our Database in the particular time period.

From: Juan José Santamaría Flecha <juanjo(dot)santamaria(at)gmail(dot)com>
To: Ramesh Penuballi <ramesh(dot)252(at)gmail(dot)com>
Cc: Pgsql-admin <pgsql-admin(at)lists(dot)postgresql(dot)org>
Subject: Re: Regarding the Spikes in the connection pool on our Database in the particular time period.
Date: 2020-09-08 10:46:50
Message-ID: CAC+AXB2dk44jKnu2D9rpysLKzbb0s_h=dZPQfpXZ=z5YRpHgpg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

On Tue, Sep 8, 2020 at 9:49 AM Ramesh Penuballi <ramesh(dot)252(at)gmail(dot)com>
wrote:

>
> We are facing an issue of sudden spikes in the connection pool in our Jira
> instance which is connected to our database server (RHEL) which is
> installed postgresql9.2 version.
>

This is a pretty old version of Postgres, at some point you should consider
upgrading.

Please find the below image which shows the increase in the connection
> pool.
>

You should size the pool based on your busy hour, 20 connections are not
that many, you can probably double that.

> These spikes are happening in our Jira instance at the particular timing
> of every day 10:30 AM IST -11:30 AM IST (i.e. 1AM EST-2AM EST) and it
> is causing the slowness of the applications.
>

The amount of connections does not look like the source of the problem, but
a symptom of slowness somewhere else. Can you log the slowest queries being
executed during that time? Something like a pgbagder report would be very
useful [1].

[1] https://pgbadger.darold.net/

Regards,

Juan José Santamaría Flecha

In response to

Browse pgsql-admin by date

  From Date Subject
Next Message Dirk Krautschick 2020-09-09 09:12:41 Actual meaning from pg_stat_replication.sync_state potential vs. sync and the usage of synchronous_standby_names
Previous Message ZongtianHou 2020-09-08 09:51:52 Re: how to clean dropped column in pg_attribute