BUG #16565: Systemd service file does not create /var/run/postgresql directory across reboot

From: PG Bug reporting form <noreply(at)postgresql(dot)org>
To: pgsql-bugs(at)lists(dot)postgresql(dot)org
Cc: mcuccato(dot)vts(at)gmail(dot)com
Subject: BUG #16565: Systemd service file does not create /var/run/postgresql directory across reboot
Date: 2020-07-31 10:05:14
Message-ID: 16565-3ce4cfab67915389@postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

The following bug has been logged on the website:

Bug reference: 16565
Logged by: Marco Cuccato
Email address: mcuccato(dot)vts(at)gmail(dot)com
PostgreSQL version: 12.3
Operating system: Red Hat Enterprise Linux Server release 7.6
Description:

When the server is rebooted, PostgreSQL service doesn't start because of
this error:
FATAL: could not create lock file "/var/run/postgresql/.s.PGSQL.5432.lock":
No such file or directory

This happens because it tries to create the lock file under
/var/run/postgresql that doesn't exists. It doesn't exist because /var/run
is a symlink to ../run (/run as absolute path) that is mounted as tmpfs.
As tmpfs is not persistent, when the server is rebooted the directory
doesn't exists and must be re-created.
Doing that, the lock file will be created successfully and the service will
start up correctly.

Please fix the postgresql-12.service systemd file: make it check and
eventually create /var/run/postgresql directory.
Thanks

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Marco Cuccato 2020-07-31 13:03:03 Re: BUG #16565: Systemd service file does not create /var/run/postgresql directory across reboot
Previous Message Tom Lane 2020-07-31 01:13:45 Re: BUG #16558: `AND FALSE` increases planning time of query on 2 tables with 1000 partitions to more than 7 seconds