puma and delayed_job process does not restart after deployment

[kzkn]

@ajgon
I use v1.20.1, thank you for your support for puma 5!

Expected Behavior

Restarts puma and delayed_job process after deployment.

Actual Behavior

puma and delayed_job process does not restart after deployment.

Log

This is the syslog from the start of deployment to the end.

Jan 14 23:39:21 myserver monit[10273]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:39:21 myserver monit[10273]: Reinitializing monit daemon
Jan 14 23:39:21 myserver monit[952]: Reinitializing Monit -- control file '/etc/monit/monitrc'
Jan 14 23:39:21 myserver monit[10276]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:39:21 myserver monit[10276]: Reinitializing monit daemon
Jan 14 23:39:21 myserver monit[952]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:39:21 myserver monit[952]: 'myserver' Monit reloaded
Jan 14 23:39:44 myserver crontab[10408]: (deploy) LIST (deploy)
Jan 14 23:39:44 myserver crontab[10409]: (deploy) REPLACE (deploy)
Jan 14 23:39:44 myserver monit[10410]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:39:44 myserver monit[952]: 'puma_rails' restart on user request
Jan 14 23:39:44 myserver monit[952]: Monit daemon with PID 952 awakened
Jan 14 23:39:44 myserver monit[952]: Awakened by User defined signal 1
Jan 14 23:39:44 myserver monit[952]: Reinitializing Monit -- control file '/etc/monit/monitrc'
Jan 14 23:39:44 myserver monit[10413]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:39:44 myserver monit[952]: 'delayed_job_rails-1' restart on user request
Jan 14 23:39:44 myserver monit[952]: Monit daemon with PID 952 awakened
Jan 14 23:39:44 myserver monit[952]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:39:44 myserver monit[952]: 'myserver' Monit reloaded
Jan 14 23:39:44 myserver systemd[1]: Reloading nginx - high performance web server.

And this is the syslog when I ran monit restart manually. This works fine.

Jan 14 23:41:40 myserver sudo[10511]: pam_unix(sudo:session): session opened for user root by ubuntu(uid=0)
Jan 14 23:41:40 myserver monit[10512]: Skipping 'allow localhost' -- host resolved to [::ffff:127.0.0.1] which is present in ACL already
Jan 14 23:41:40 myserver monit[952]: 'puma_rails' restart on user request
Jan 14 23:41:40 myserver monit[952]: Monit daemon with PID 952 awakened
Jan 14 23:41:40 myserver monit[952]: Awakened by User defined signal 1
Jan 14 23:41:40 myserver monit[952]: 'puma_rails' trying to restart
Jan 14 23:41:40 myserver monit[952]: 'puma_rails' stop: '/bin/sh -c cat /run/lock/rails/puma.pid | xargs --no-run-if-empty kill -QUIT; sleep 5'
Jan 14 23:41:40 myserver sudo[10511]: pam_unix(sudo:session): session closed for user root
Jan 14 23:41:45 myserver monit[952]: 'puma_rails' start: '/bin/sh -c cd /srv/www/rails/current && MYENV="xxxxxx
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] Puma starting in cluster mode...
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] * Puma version: 5.1.1 (ruby 2.7.1-p83) ("At Your Service")
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] *  Min threads: 0
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] *  Max threads: 16
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] *  Environment: staging
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] *   Master PID: 10520
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] *      Workers: 2
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] *     Restarts: (✔) hot (✖) phased
Jan 14 23:41:46 myserver puma-rails[10521]: [10520] * Preloading application
Jan 14 23:41:49 myserver puma-rails[10521]: [10520] * Listening on unix:///srv/www/rails/shared/sockets/puma.sock
Jan 14 23:41:49 myserver puma-rails[10521]: [10520] Use Ctrl-C to stop

[kzkn]

related? #29

[willkoehler]

I've been exploring this issue in #244 culminating in #244 (comment)

It looks like @kzkn was ahead of me and has a preposed fix.

EDIT: I can confirm #252 fixes the problem for me