Puppet not correctly configured after new host deployment (path issue)

Hello,

i deployed a new installation of Foreman and Katello. After some initial configuration almost everything worked as i expected. Enrolling CentOS VMs in VMware over PXE with FreeIPA Integration works like a charm. One issue i have is with Puppet. Puppet get successfull installed in the kickstart phase but something went wrong due to wrong path of Puppet binary. Puppet is not running after deployment, the config directory has a blank config file without Server Configuration. Starting Puppet agent by hand with server parameter works fine.

here a little snippet of the log after a deployment of CentOS 7

Running transaction

Installing : puppet-agent-5.5.10-1.el7.x86_64 1/1
Uploading Package Profile
Verifying : puppet-agent-5.5.10-1.el7.x86_64 1/1

Installed:
puppet-agent.x86_64 0:5.5.10-1.el7

Complete!
Uploading Enabled Repositories Report
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
Loaded plugins: fastestmirror, product-id, subscription-manager
/tmp/ks-script-MSjeWj: line 263: /etc/puppet/puppet.conf: No such file or directory
Created symlink /etc/systemd/system/multi-user.target.wants/puppet.service, pointing to /usr/lib/systemd/system/puppet.service.
/tmp/ks-script-MSjeWj: line 291: /usr/bin/puppet: No such file or directory
Informing Foreman that we are built

I installed Foreman and Katello together running this Version:

rpm -qa foreman katello puppet*
puppetserver-5.3.7-1.el7.noarch
katello-3.11.0-0.1.rc2.el7.noarch
puppet5-release-5.0.0-4.el7.noarch
puppet-agent-oauth-0.5.1-3.el7.noarch
foreman-1.21.0-1.el7.noarch
puppet-agent-5.5.10-1.el7.x86_64

Can anyone help me how to fix this?

In our kickstart templates we have a parameter enable-puppet5. Without it we still assume the non-AIO style packaging. That may be breaking some things. Try setting the value enable-puppet5 to true.

Patches to update this would be very welcome.

works! Thank you!

I’ve got the same problem but with:

foreman-1.20.2-1.el7.noarch
/opt/puppetlabs/bin/puppet agent --version
4.10.12
/opt/puppetlabs/bin/puppetserver --version
puppetserver version: 2.8.1

Since I’m quite new to foreman, can you point me in the right direction on how to fix this?
I did not find out where to set that boolean. I’m using the ‘Kickstart Default’ Template btw.

I accidently installed puppet 4 at first because i used the foreman install instruction from the katello page. I had the same problem like you and checked the puppet version. Because puppet 4 is quite old i updated to puppet 5 and still had problems getting puppet correctly installed after host deployment. the enable-puppet5 and if im remember right enable-puppet-5-repository settings in my hostgroup solved this :slight_smile:

I also followed the installation instructions for katello, which point to the puppet-release-pc1(puppet4).
I’ll update to puppet5 next month … most likely. Thanks for helping out. I got it to work now.
setting enable-puppet5 to true in the parameters for the hostgroup did the trick for me :slight_smile: