each puppet run will try to enable the service

[ltutar]

I wanted to mention the following behaviour.
elasticsearch-logstashforwarder (v0.1.1)
PE2015.2
CentOS 7

Each puppet run prints the following notice information:
Notice: /Stage[main]/Logstashforwarder::Service/Logstashforwarder::Service::Init[logstash-forwarder]/Service[logstash-forwarder]/enable: enable changed 'false' to 'true'

Debug: Caching connection for https://puppetmaster.home:8140
Debug: Executing: '/bin/systemctl is-active logstash-forwarder'
Debug: Executing: '/bin/systemctl show logstash-forwarder --property LoadState --property UnitFileState --no-pager'
Debug: Executing: '/bin/systemctl unmask logstash-forwarder'
Debug: Executing: '/bin/systemctl enable logstash-forwarder'
Notice: /Stage[main]/Logstashforwarder::Service/Logstashforwarder::Service::Init[logstash-forwarder]/Service[logstash-forwarder]/enable: enable changed 'false' to 'true'
Debug: /Stage[main]/Logstashforwarder::Service/Logstashforwarder::Service::Init[logstash-forwarder]/Service[logstash-forwarder]: The container Logstashforwarder::Service::Init[logstash-forwarder] will propagate my refresh event
Debug: Logstashforwarder::Service::Init[logstash-forwarder]: The container Class[Logstashforwarder::Service] will propagate my refresh event
Debug: Class[Logstashforwarder::Service]: The container Stage[main] will propagate my refresh event

[electrical]

Its very possible there are some ugly bugs here and there since i haven't done any real work on the module lately.
We are working on the replacement of it called filebeat.
An official module for it is on the way and should be ready when it goes GA.
That said, i'll see if i can fix this easy and fast so at least it will work.

[ltutar]

Thank you in advance

[ozbillwang]

nice to see some progresses. Waiting for the updates.

[rjhornsby]

I'm trying to figure out if this is a more general issue with something in Puppet 4. I have a very small custom module for Splunk which just sets a couple of config file options and enables the service. It was fine under Puppet 3. Same module under Puppet 4, it does exactly as described here - always says enable changed 'false' to 'true'. Still hunting for a solution. Issue is probably not specific to puppet-logstashforwarder.

[ltutar]

I do not see this behaviour if I keep using Puppet 4 on CentOS 6. I had this with Puppet 4 on CentOS 7. What is your OS?

[rjhornsby]

What is your OS?

Hmm. RHEL 7, aka CentOS 7.

Edit: One of the big changes between RHEL 6 and RHEL 7 was systemd style services over init scripts. I wonder if there's something in Puppet 4 that's not always handling old-style init scripts under RHEL 7 properly. I know, for example, the Splunk forwarder I'm using is still based on an old-style init script carried over from our RHEL 6 installations.

[ltutar]

Can you easily test your setup Puppet 4 en CentOS 6?

[rjhornsby]

Confirmed your findings, @ltutar. I cannot reproduce the issue on a RHEL 6 system.

[rjhornsby]

This looks like PUP-5296[1]. Adding provider => 'redhat' to the service resource declaration is a confirmed work-around for RHEL 7. This may resolve the issue with puppet-logstashforwarder as well.

[1] https://tickets.puppetlabs.com/browse/PUP-5296

[ltutar]

1. Did you test the work around with success?
2. Today is PE2015.3.0 released.
   [root@puppetmaster ~]# puppet --version
   4.3.1
   I do not know if it also has this unwanted behaviour.