Hi Folks,

I'm trying to run the Trac standalone server (tracd) using supervisor, with
Nginx as the frontend.  This is all working fine except that when I try to
restart Trac with "sudo supervisorctl restart trac", it doesn't work.  It
starts a new process, but the old one is not cleanly killed, and therefore
the new process fails because the port it needs is blocked.
So the question is, how can I make sure the tracd process is really killed
when I tell supervisord to stop it?  I've tried using the pidproxy but this
didn't help.  Here's my current configuration:

trac_run.sh
----------------------
#!/bin/bash
. bin/activate
tracd --port=9000 --auth="*,/srv/users.htdigest,XXXX trac"
--env-parent-dir=/srv/trac --pidfile=/var/run/tracd.9000

supervisor_trac.conf
----------------------
[program:trac]
command=/usr/bin/pidproxy /var/run/tracd.9000 /opt/trac/trac_run.sh
directory=/opt/trac
user=www-data
autostart=true
autorestart=true
stdout_logfile=/var/log/my.domain.com_supervisord.log
redirect_stderr=true


As you can see I'm running Trac from a virtualenv, I don't know if this
makes any difference.


Thanks for any help,
Cheers,
Robin
_______________________________________________
Supervisor-users mailing list
[email protected]
http://lists.supervisord.org/mailman/listinfo/supervisor-users

Reply via email to