I have a small home network with a number of (virtual) Ubuntu machines to monitor with nagios and recently switched to Nagios 4. Since then, nrpe appears to be unstable.
My network does not use SSL and I have no intention to introduce it at the moment.
As a consequence, I configured all nrpe services on the machines to monitor to start with the "-n" option to disable ssl and configured all check_nrpe calls from the nagios machine to also use the "-n" option.
What confuses me is that this seems to work for a while (service statuses appear good in nagios), but then services start to trigger "CHECK_NRPE: Receive header underflow - only -1 bytes received (4 expected)." and services go to UNKNOWN... and then after some time, services seem to recover and be fine again. This is moving back and forth with no pattern.
When the check via nagios fail, they also fail when running these from the command line on the nagios machine.
Sometimes all machines are affected for a while, sometimes only one machine. Most of the time, rebooting the machine resolves the issue, at least for a while.
Does anyone have any idea what is causing this and how to resolve it. At this point, nagios is pretty useless.