This may be a feature request, or it may be an RTFM. In any case, if it
doesn't belong on this list, please tell me so :-)
If an init script fails, currently the stderr output it generates doesn't
seem to be logged anywhere. So for example, monit shows me just:
Mar 26 20:58:14 localhost monit[24709]: 'testapp_mongrel_1' process is not
running
Mar 26 20:58:14 localhost monit[24709]: 'testapp_mongrel_1' trying to restart
Mar 26 20:58:14 localhost monit[24709]: 'testapp_mongrel_1' start:
Mar 26 20:58:14 localhost monit[24709]: /usr/bin/mongrel_rails
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:14 localhost monit[24709]: cluster::start
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:14 localhost monit[24709]: --clean
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:14 localhost monit[24709]: -C
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:14 localhost monit[24709]: /etc/mongrel.d/testapp.conf
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:14 localhost monit[24709]: --only
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:14 localhost monit[24709]: 10001
Mar 26 20:58:14 localhost monit[24709]:
Mar 26 20:58:44 localhost monit[24709]: 'testapp_mongrel_1' failed to start
However if I run the script *by hand*, I can see what the problem is:
# /usr/bin/mongrel_rails cluster::start --clean -C /etc/mongrel.d/testapp.conf
--only 10001
starting port 10001
!!! Prefix must begin with / and not end in /
!!! User does not exist: Rtestapp
!!! Group does not exist: Rtestapp
mongrel::start reported an error. Use mongrel_rails mongrel::start -h to get
help.
Is it possible to get monit to record the script error output somewhere?
Thanks,
Brian.
_______________________________________________
monit-dev mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/monit-dev