Ewoud Kohl van Wijngaarden

> It does not make sense to have both Integer and Stdlib::Port since it is defined as Integer[0,65535] - should probably remove Integer from the variant unless it is possible...

I'm surprised I didn't link it, but https://github.com/puppetlabs/puppetlabs-stdlib/pull/1415 was my attempt to address it.

IIRC there's a check in rubocop-rspec that finds these. Clearly there's a need for it.

I don't see why. The commit you refer to comes from a time when Puppet had vendored SemanticPuppet and you could only load one or the other. The vendored implementation...

@ThomasMinor you're pinging the non-Perforce people but IMHO they are responsible for the `CONTRIBUTING.md` file. Given their [recent blog post](https://www.puppet.com/blog/open-source-puppet-updates-2025) I'm not sure if they actually want people to contribute...

> Hmm, are you saying specifying a commit hash can not be unique in a git repository? Because that would be news to me. No, that's not what I'm saying...

https://github.com/theforeman/foreman_ansible/pull/731 already pulled so I opened https://github.com/theforeman/foreman_ansible/pull/733 for the ignore part.

So that's pretty much what I suspected. I think it's something that we should tackle properly. This now leads to 2 service restarts: once when the package is updated and...

> Check if configuration files for modules are not missing with: rpm -qV httpd As a regression test, it would be good to have this in acceptance tests. That makes...

This often happens if you don't generate types in your deployment. See https://www.puppet.com/docs/puppet/8/environment_isolation.html