First I noticed lots of things taking forever, such as logging in via ssh. Then I found log messages like:
2720.scope: Failed to activate service 'org.freedesktop.systemd1': timed out (service_start_timeout=25000ms) Feb 22 09:55:56 tomh su: pam_systemd(su:session): Failed to create session: Connection timed out Then I tried to reboot, and the reboot command said there was no such device as /dev/initctl, but it was there and looked like: lrwxrwxrwx 1 root root 25 Feb 23 07:06 /dev/initctl -> /run/systemd/initctl/fifo I eventually got this to work: systemctl --force --force reboot but I have no idea how long it took to finish (I was doing all this remotely). Today I see /run/systemd/initctl/fifo is open in PID 1 and everything is talking normally. There was a PID 1 out there yesterday, but it sure seemed to be completely screwed up somehow. Remind me again why it is a good idea to have PID 1 be a complicated dbus activated grab bag of crap? Why couldn't all the complicated crap be in PID 2 where it could be restarted and let PID 1 just be the traditional init task that reaps dead PIDs? _______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
