This article applies as of PRTG 22

Our SNMP and SSH sensors occasionally show the Down status with timeout and authentication errors without obvious reasons. What could be causing these issues, and how can I avoid SNMP and SSH sensor instabilities?

Sporadic timeouts and authentication errors with SNMP and SSH sensors

SNMP sensors and SSH sensors that switch between the sensor states Up and Down from time to time, showing timeout and authentication errors, indicate a connection instability. This instability can be a result of the configuration of your monitored devices. If you are sure that credentials, passwords, or private keys on the target devices remained unchanged all the time, check other connections that the affected servers might establish.

Look for configurations that devices on which SNMP and SSH sensor are failing have in common. For example, a customer reported that the devices on which the sensors showed the errors all sent syslogs to an external server. First, timeouts happened only sporadically. But after switching off the external server without disabling the sending of syslogs, the sensors permanently showed the Down status.

Turning off the syslog receiver server also resulted in strange behavior of other services on the sender servers. For example, some services needed a lot of time to spawn child services. This also affected SSH. Because this always happened when a new connection was created, it showed up as a timeout or authentication error in PRTG.

In this case, disabling the sending of syslogs to external servers resolved the timeout and authentication issues of SNMP and SSH sensors. If you encounter such issues as well, keep in mind when investigating the errors that these can be potential side effects of applications like sending syslog messages from the monitored server.