The following bug has been logged on the website:
Bug reference: 14967
Logged by: Patrik Martinsson
Email address: martinsson.patrik@gmail.com
PostgreSQL version: 9.3.20
Operating system: CentOS 7.4.1708
Description:
Hi,
I'm not sure where the "bug" lies here, but I just noticed that running an
yum update from postgresql93-9.3.20-1PGDG.rhel7.x86_64 ->
9.3.20-3PGDG.rhel7.x86_64 makes postgresql impossible to start through
systemd.
The issue seems to be related to the parameter "TimeoutSec=infinity" in
/usr/lib/systemd/system/postgresql-9.3.service.
Running "systemctl start postgresql-9.3.service" immediately fails with
> "Job for postgresql-9.3.service failed because a timeout was exceeded. See
"systemctl status postgresql-9.3.service" and "journalctl -xe" for details."
Running "systemctl status postgresql-9.3.service" reveals following,
$ > systemd[1]: Starting PostgreSQL 9.3 database server...
$ > systemd[1]: postgresql-9.3.service start operation timed out.
Terminating.
$ > systemd[1]: Failed to start PostgreSQL 9.3 database server.
$ > systemd[1]: Unit postgresql-9.3.service entered failed state.
$ > systemd[1]: postgresql-9.3.service failed.
The difference I see in the /usr/lib/systemd/system/postgresql-9.3.service
between these two versions are that the "TimeoutSec-parameter-value" is
changed from "300" to infinity. Changing the parameter back to 300 makes
postgresql start again as normal.
I have to admit I don't really understand whats going on here, since I'm
under the impression that setting the value to "infinity" would disable the
timeout and thus wait forever on "ExecStartPre=" to do its thing. However,
there is clearly something I'm missing.
This is not a bug in postgresql per say, but I couldn't find a better place
to put this.
Any suggestions ?
Best regards,
Patrik Martinsson
Sweden