Red Hat Bugzilla – Bug 811485
dovecot service does not properly restart after reboot
Last modified: 2012-04-11 05:32:07 EDT
Description of problem: After rebooting a server with installed dovecot. I let dovecot restart with "systemctl enable dovecot.service" after reboot. This automaticly starts the process "/usr/sbin/dovecot -F" and some child processes like dovecot/anvil, dovecot/log and dovecot/config.
Connecting to the server produces in /var/log/maillog
dovecot: master: Warning: Time moved backwards by 3526 seconds, waiting for 180 secs until new services are launched again.
dovecot: log: Warning: Time moved backwards by 3526 seconds.
dovecot: config: Warning: Time moved backwards by 3525 seconds.
dovecot: anvil: Warning: Time moved backwards by 3526 seconds.
and I can't retrieve any messages. Only after restart of the dovecot service via "/bin/systemctl restart dovecot.service" it works properly.
Version-Release number of selected component (if applicable): dovecot-2.0.19-1.fc16.x86_64
If time moved backwards, it's expected behavior (from dovecot documentation):
Dovecot v2.0 finally tries to handle this a bit more gracefully. Its behavior when time moves backwards is:
Existing imap and pop3 processes either sleep or die, just like with older versions
Master process stops creating new processes until either the original time is reached, or after a maximum wait of 3 minutes.
Other processes log a warning, but do nothing else.
Timeouts are updated so that the timeout is executed approximately at the original intended time.