Can confirm the bug. I believe it is due to slow proftpd shutdown and
start-stop-daemon not waiting for proftpd process to shut down entirely
(to remove its PID file to be exact).
In my case I'm using proftpd together with MySQL and i suspect this is
the cause for slow shutdown.
While solution proposed by Colin Dearborn seems to work fine (unless of
course shutdown takes more than 1 sec), I'd like to propose making
start-stop-daemon wait for proftpd to exit, that is:
Line 109 of /etc/init.d/proftpd:
From:
start-stop-daemon --stop --signal $SIGNAL --quiet --pidfile "$PIDFILE"
To:
start-stop-daemon --stop --signal $SIGNAL --quiet --pidfile "$PIDFILE"
--retry=10/TERM/10
(stopping proftpd using $SIGNAL which is SIGTERM at the time, waiting 10
secs for it to terminate, retrying SIGTERM and waiting again)
Note that we cannot use simple --retry=10 because this will be
interpreted by start-stop-daemon as --retry=TERM/10/KILL/10 and we
really dont want even the slightest chance to KILL proftpd at this
moment in script.
BTW:
proftpd -v:
ProFTPD Version 1.3.5rc3
lsb_release -a:
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 14.04 LTS
Release: 14.04
Codename: trusty
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1325338
Title:
proftpd service stopped by it's own
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/proftpd/+bug/1325338/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
