Resolution: Won't Do
Affects Version/s: PUP 5.5.1
Fix Version/s: None
Template:PUP Bug Template customfield_10700 250482
Method Found:Needs Assessment
QA Risk Assessment:Needs Assessment
OS: Centos7 1804
Install: Puppet agent 5.5.1
Puppet Server: 5.3.1-1
systemd Config: ExecStart=/opt/puppetlabs/puppet/bin/puppet agent $PUPPET_EXTRA_OPTS --no-daemonize
Process: /opt/puppetlabs/puppet/bin/ruby /opt/puppetlabs/puppet/bin/puppet agent --no-daemonize
After a network failure puppet agent try to connect to the puppet server (and it can not do it since there is a network problem) getting errror (it is expected):
Unable to fetch my node definition, but the agent run will continue:
Failed to open TCP connection to puppet5:8140 (getaddrinfo: Name or service noi known)
After network is back puppet runs every 30 mins but it still give the same error even though the network service is re-establish
I can telnet puppet 8140.
IF I run manually "puppet agent -t" it is successful but the systemd process is still failing.
If the systemd service runs normally it fails.
The ONLY option to make it work is to perform a "systemctl restart puppet"
I perform a tcpdump on port 8140 but nothig is coming out of the systemd process, so it seems as it is stuck with its failure.
Expected Way of working:
After network recovery the pupept agent should connect to the server. It works fine if doing manually and only solution is to restart the process which then works fine.