When you get connection type
errors running plans, tasks, or the orchestrator in Puppet Enterprise, you can resolve it by fixing common orchestration service communication issues.
Error messages
is either disconnected or does not have a connection type that supports running tasks/plans
does not have a connection type that supports running Puppet
Version and installation information
PE version: All supported versions
Solution
Resolve these errors by checking for and fixing orchestrator service issues.
Restart the pxp-agent on the target node
If pxp-agent service isn’t running, a connection from the orchestrator can't be initiated. Restarting the service on the target node can resolve the issue.
- To stop the target node from running:
puppet resource service pxp-agent ensure=stopped
- To restart the target node:
puppet resource service pxp-agent ensure=running
Note: If restarting the pxp-agent service only works temporarily, contact your organization’s network team to ensure the settings on your load balancer allow for a persistent idle connection.
Ensure Port 8142 is open
The orchestration service uses port 8142 for communication between the primary server and the agent.
- Test communication between the primary server and agent on port 8142 using telnet.
- If port 8142 isn’t open, work with your organization’s networking team to open it.
Review the logs and search for errors
Check the orchestration service and related component logs on these nodes for errors:
- On the target node:
/var/log/puppetlabs/pxp-agent/pxp-agent.log
- On the primary server:
/var/log/puppetlabs/orchestration-services/orchestration-services.log
- On compilers:
/var/log/puppetlabs/puppetserver/pcp-broker.log
If you can’t resolve the issue after using these steps, open a Support ticket and attach the logs listed above.
How can we improve this article?
2 comments
Log location is /var/log/puppetlabs/pxp-agent/pxp-agent.log on target node
Ryan, thanks for that correction, I'll take care of it.
Please sign in to leave a comment.
Related articles