Re: Error

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Rajesh Kumar <rajeshkumar(dot)dba09(at)gmail(dot)com>
Cc: pgsql-admin(at)lists(dot)postgresql(dot)org, Holger Jakobs <holger(at)jakobs(dot)com>
Subject: Re: Error
Date: 2023-11-05 16:19:25
Message-ID: 3652551.1699201165@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

Holger Jakobs <holger(at)jakobs(dot)com> writes:
> Am 05.11.23 um 16:46 schrieb Rajesh Kumar:
>> Is it because of vacuuming?

> No, some application has called pg_terminate_backend() function or the
> server was shut down.
> see
> https://www.postgresql.org/docs/current/functions-admin.html#FUNCTIONS-ADMIN-SIGNAL

Concretely, this indicates that something sent the server process
a SIGTERM signal. At least within Postgres, that can only come
from pg_terminate_backend() or the postmaster sending it because
it was sent a SIGINT signal to do a fast shutdown.

So, if you weren't doing this intentionally, you need to look
for an external source of those signals.

A common newbie mistake that could lead to this is launching the
postmaster by hand, or in a shell script, in a way that doesn't
dissociate it from your terminal session. In that case typing
control-C or ending the session would deliver one of these
signals. Make sure your postmaster start script (a) uses nohup
and (b) redirects all three of postmaster's stdin, stdout, and
stderr away from the terminal.

Could be something else too, but in the end you've got something
signaling the server that you don't want.

regards, tom lane

In response to

  • Re: Error at 2023-11-05 15:49:40 from Holger Jakobs

Browse pgsql-admin by date

  From Date Subject
Next Message Jeff Janes 2023-11-05 16:54:12 Re: Error
Previous Message Holger Jakobs 2023-11-05 15:49:40 Re: Error