Hi,
We're running sogo 5 on an upgraded squeeze -> jessie -> stretch ->
buster system, our init system is now systemd, and it uses a 'generated'
unit file for sogo:
root@sogo:/etc/logrotate.d# systemctl list-unit-files | grep sogo
sogo.service generated
Now, we occasionally get this midnight error from logrotate, after log
rotation:
Apr 23 00:02:14 sogo systemd[1]: sogo.service: Found left-over process 28417
(sogod) in control group while starting unit. Ignoring.
Apr 23 00:02:14 sogo systemd[1]: This usually indicates unclean termination of
a previous run, or service implementation deficiencies.
Apr 23 00:02:14 sogo systemd[1]: sogo.service: Found left-over process 28418
(sogod) in control group while starting unit. Ignoring.
Apr 23 00:02:14 sogo systemd[1]: This usually indicates unclean termination of
a previous run, or service implementation deficiencies.
Apr 23 00:02:14 sogo systemd[1]: sogo.service: Found left-over process 28419
(sogod) in control group while starting unit. Ignoring.
Apr 23 00:02:14 sogo systemd[1]: This usually indicates unclean termination of
a previous run, or service implementation deficiencies.
This causes the logrotate sogo postrotate script ("/bin/systemctl
restart sogo.service") to fail, and after that sogo is no longer
running. :-(
I now created a script to verify that sogo is running, and restart it if
it is not. But solving the issue at it's root (the failing restart)
would be much better.
Anyone here experienced the same..? And how did you solve it?
Thanks and enjoy your weekend, everybody!
MJ
--
users@sogo.nu
https://inverse.ca/sogo/lists
