setup
- vagrant up
- ssh-copy-id vagrant@localhost -p 2222 # passwd: vagrant
- BRANCH=master bundle exec cap stage deploy
- ssh vargrant@localhost -p 2222 "ps -ef | grep cron-job"
- tail -f log/* # check ./log/cron.log
analytics
execute "bash -l -c 'cd /home/vagrant/test-demo/current && (nohup rvm use 2.4.2 do ruby bin/cron_job 2>&1 &)", named A.
this will fail to create daemon process
execute "bash -l -c 'cd /home/vagrant/test-demo/current && (nohup rvm use 2.4.2 do ruby bin/cron_job 2>&1 &) && sleep 2' ", named B.
this will succeed to create daemon process
- ssh execute cmd
A, ssh exit before nohup complete to create daemon process, so we can add a sleep for 2 seconds to prevent ssh from exiting before nohup complete to spawn daemon process.
note: you should use eye to spawn daemon process and monitor them.