"failed to stop" always after 60 seconds

I'm using Monit 5.8.1 to monitor a background job queue (Sidekiq) with the following config:

check process sidekiq
  with pidfile /var/run/sidekiq.pid
  start program = "/bin/su - chief -c 'cd /var/www/myapp && RAILS_ENV=production bundle exec sidekiq -C /var/www/myapp/config/sidekiq.yml'" with timeout 60 seconds
  stop program = "/bin/su - chief -c 'cd /var/www/myapp && bundle exec sidekiqctl shutdown /var/www/myapp/tmp/sidekiq.pid 70'" with timeout 80 seconds
  if totalmem > 1000 MB for 5 cycles then alert
  if cpu > 50% for 5 cycles then alert
  if cpu > 90% for 5 cycles then restart
  if 3 restarts within 5 cycles then timeout

The problem appears when I try to stop the process. After exactly 60 seconds, Monit always reports "failed to stop" and immediately afterwards "stop action done":

[EST Nov 12 08:56:12] info     : 'sidekiq' stop on user request
[EST Nov 12 08:56:12] info     : monit daemon with PID 22405 awakened
[EST Nov 12 08:56:12] info     : Awakened by User defined signal 1
[EST Nov 12 08:56:12] info     : 'sidekiq' stop: /bin/su
[EST Nov 12 08:57:12] error    : 'sidekiq' failed to stop
[EST Nov 12 08:57:13] info     : 'sidekiq' stop action done

I would expect it to report that after 80 seconds, if the process cannot be stopped.

I changed the daemon check interval (set daemon) which was initially set to 60, to other values like 120 or 20, but the behaviour didn't change.

Comments (3)