I'm using SNMP to get temperature and voltage values from your UPS.

Now the management module run into some issue and kept reporting the last value over and over. The value itself was within the limits, so no alert was raised. Usually these values are fluctuating and it is obvious when you look at the chart that something is wrong.

Example Chart

We are currently not using "Unusual Detection", would this recognize such a change? Or is there another way to report this?

Thanks

Article Comments

Hello Stephan,

Depending on your settings of the Unusual Detection, it is likely that PRTG will identify such behavior so I recommend to have a closer look at the settings under Setup | System Administration | Monitoring.

However, we should concentrate on the sensor's behavior nonetheless. Did you already try to re-create the sensor? Also restarting the targeted machine might help to exclude that SNMP values are hanging somewhere.

Best,
Sebastian


Jul, 2020 - Permalink

Hi Sebastian,

thanks for your reply.

The underlying problem was loss of communication between the UPS and the UPS management module. Therefore the module kept sending the last value. After restarting the module it is now reporting an error (as expected, UPS is faulty) and the sensor went down. So this should not be a sensor issue.

I have now enabled unusual detection with recommended settings and enabled logging. Is there any way to test this?

Thanks Stephan


Jul, 2020 - Permalink

Hi,

I'm afraid there is no way user friendly way in simulating a suspicious behavior that triggers the unusual detection in PRTG.

The only thing I could think of would be to run a SNMP simulator that simulates the UPS system and to distort values in the simulator.

Best,
Sebastian


Jul, 2020 - Permalink