Back to homepage | Back to blog | Back to archive
Written by Adrian Kjær Dalhaug at 2020-04-21 07:37:52 UTC.
I got reports of strange behaviour and figured I'd check a cronjob to see if it had any problems. Running journalctl confirmed my suspicion: The last entry in the log was a few days ago.
Coincidently I remembered that the disk was full a few days ago before I
increased the volume. Maybe crond had crashed due to full disk?
systemctl status crond said it was running just fine. Strange.
Next I checked my crontab, but it looked correct as well. So cron is running my correctly configured crontab... could it be...?
systemctl status systemd-journald. Yep! Crashed.
After restarting the service I could read output from the cronjob at its next run. Solving the initial bug reports were easy when equipped with proper logs.
Today I learned that it's probably a good idea to restart a server after a full disk to ensure every service has recovered correctly.
Comments? Mail them to: adrian@enpo.no.
Generated at 2020-04-21 07:37:53 UTC.