9.3. Troubleshooting Issues when Configuring Hosts

9.3.1. Diagnosing Communication Issues with Hosts

Each time the installer runs a Actions::Staypuft::Host::PuppetRun subtask on a host, the installer starts the Puppet Agent on the host through SSH. Ensure the installer resolves the host's domain name:
[root@installer ~]# ping macd050992f6aa3.example.com
If the Actions::Staypuft::Host::PuppetRun subtask is in progress, check if puppet is running on the host:
[root@macd050992f6aa3 ~]# ps -aux | grep "puppet"
If the installer cannot resolve the host's domain name or Puppet Agent isn't running on the host, it means the installer has failed to reach to host. Check your installer's network and DNS settings to correct this issue.

9.3.2. Reviewing Host Reports

If the host fails configuration at a Actions::Staypuft::Host::AssertReportSuccess subtask, check the host reports for any commands that reported errors.
Host reports provide information on each step in Puppet's configuration process, including if the step passed or failed. This helps identify specific points of error in the configuration process.
Navigate to HostsAll hosts and select the Name of the faulty host. Click the Reports button from the Details sections. This displays a list of reports from each time a Puppet configuration ran on that host.
Select the latest report to view errors that occurred on the latest configuration attempt. Scroll down the report until you find the step that caused the error. Identifying this step can help with implementing a solution to fix the configuration process the next time Puppet runs. For help with diagnosing such errors, contact a Red Hat Technical Support representative.

9.3.3. Triggering Puppet Agent

After performing any solutions or workarounds on your host, trigger the Puppet Agent to run again and see if the error still occurs. Navigate to HostsAll hosts, select the Name of the host, and click Run puppet. Puppet Agent attempts to configure the host again and reports back to the installer if the error reappears or any new errors appear.
Alternatively, you can run Puppet Agent manually on the host. Log into the host and enter the following command on the host to trigger the Puppet Agent to run:
# puppet agent -t
The -t (or --test) option enables the most common options for performing Puppet Agent configuration and reporting the results.
The Puppet Agent attempts to configure the host again, reports back to the installer if the error reappears or any new errors appear, and displays these results directly on the host's terminal screen.