Detailed bug description:
We have manifest which is installing puppetmaster package.
During the deployment we are stopping puppetmaster service, but each puppet apply, it tries to stop service again, but service is already stopped.
For example:
root@node-3:~# cat test.pp
service { 'puppetmaster':
ensure => stopped,
enable => false,
}
root@node-3:~# puppet apply test.pp
Notice: Compiled catalog for node-3.test.domain.local in environment production in 0.25 seconds
Notice: /Stage[main]/Main/Service[puppetmaster]/ensure: ensure changed 'running' to 'stopped'
Notice: Finished catalog run in 0.49 seconds
root@node-3:~# service puppetmaster status
* master is not running
root@node-3:~# echo $?
3
root@node-3:~# puppet apply test.pp
Notice: Compiled catalog for node-3.test.domain.local in environment production in 0.27 seconds
Notice: /Stage[main]/Main/Service[puppetmaster]/ensure: ensure changed 'running' to 'stopped'
Notice: Finished catalog run in 0.54 seconds
root@node-3:~# apt-cache policy puppetmaster
puppetmaster:
Installed: 3.8.5-2~u14.04+mos1
Candidate: 3.8.5-2~u14.04+mos1
If I am using upstream init script from 3.8.5 branch, it works well.
Attached diff of init script from package with upstream init script
reassign to mos team. because its linux package.