forked from tobi/delayed_job
-
Notifications
You must be signed in to change notification settings - Fork 954
monitor process
Smit Patel edited this page Jan 5, 2016
·
1 revision
Q: There is a command line option for a "monitor process". What is that?
A: It passes true
for :monitor
to this method: http://daemons.rubyforge.org/classes/Daemons.html#M000004
Also, delayed_job can be easily monitored with monit:
check process delayed_job with pidfile /var/www/current/tmp/pids/delayed_job.pid start program = "/var/www/current/script/delayed_job -e production start" as uid ubuntu and gid ubuntu stop program = "/var/www/current/script/delayed_job -e production stop" as uid ubuntu and gid ubuntu if mem is greater than 300.0 MB for 1 cycles then restart if cpu is greater than 50% for 2 cycles then alert if cpu is greater than 80% for 3 cycles then restart group background
To use monit with delayed_job workers you can use -i (identifier)
command line switch.
check process delayed_job_0 with pidfile /var/www/current/tmp/pids/delayed_job.0.pid start program = "/var/www/current/script/delayed_job -e production start -i 0" as uid ubuntu and gid ubuntu stop program = "/var/www/current/script/delayed_job -e production stop -i 0" as uid ubuntu and gid ubuntu if mem is greater than 300.0 MB for 1 cycles then restart if cpu is greater than 50% for 2 cycles then alert if cpu is greater than 80% for 3 cycles then restart group background check process delayed_job_1 with pidfile /var/www/current/tmp/pids/delayed_job.1.pid start program = "/var/www/current/script/delayed_job -e production start -i 1" as uid ubuntu and gid ubuntu stop program = "/var/www/current/script/delayed_job -e production stop -i 1" as uid ubuntu and gid ubuntu if mem is greater than 300.0 MB for 1 cycles then restart if cpu is greater than 50% for 2 cycles then alert if cpu is greater than 80% for 3 cycles then restart group background