2

I have the following output as a user running crontab -l:

#Ansible: backup-external chaos
*/20 * * * * flock --nonblock /home/mu/.cache/backup-external.lock backup-external chaos
*/20 * * * * /home/mu/bin/ddc-save-brightness

Neither job is executed. If I run them manually, they seem to work just fine.

The Ansible snippet there comes from using Ansible to add this one job for my user.

Looking at systemctl status crond.service -l makes it clear that the service itself is running. It seems to fail to load the crontab for my user mu due to SELinux it seems:

● crond.service - Command Scheduler
   Loaded: loaded (/usr/lib/systemd/system/crond.service; enabled; vendor preset: enabled)
   Active: active (running) since Mi 2016-01-27 17:51:08 CET; 1h 43min ago
 Main PID: 1351 (crond)
   CGroup: /system.slice/crond.service
           └─1351 /usr/sbin/crond -n

Jan 27 17:51:09 martin-friese.fritz.box crond[1351]: (mu) Unauthorized SELinux context=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 file_context=unconfined_u:object_r:user_cron_spool_t:s0 (/var/spool/cron/mu)
Jan 27 17:51:09 martin-friese.fritz.box crond[1351]: (mu) FAILED (loading cron table)
Jan 27 17:51:09 martin-friese.fritz.box crond[1351]: (CRON) INFO (running with inotify support)
Jan 27 18:01:01 martin-friese.fritz.box CROND[3726]: (root) CMD (run-parts /etc/cron.hourly)
Jan 27 18:01:01 martin-friese.fritz.box anacron[3737]: Anacron started on 2016-01-27
Jan 27 18:01:01 martin-friese.fritz.box anacron[3737]: Will run job `cron.daily' in 13 min.
Jan 27 18:01:01 martin-friese.fritz.box anacron[3737]: Jobs will be executed sequentially
Jan 27 18:01:01 martin-friese.fritz.box run-parts[3741]: (/etc/cron.hourly) starting mcelog.cron
Jan 27 18:14:01 martin-friese.fritz.box anacron[3737]: Job `cron.daily' started
Jan 27 19:01:01 martin-friese.fritz.box CROND[2681]: (root) CMD (run-parts /etc/cron.hourly)

This is on Fedora 23 and I did not change the SELinux policy, so it is probably enforcing strictly.

What do I have to change in order to get the jobs running again?

2

This was resolved in this bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1298192

Please make sure you have the latest kernel: 4.3.3-301

  • Interesting! I am now running 4.3.3-303.fc23.x86_64 and my jobs are working again without changing anything. – Martin Ueding Jan 28 '16 at 18:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.