Right now it seems its possible to enable/disable unusual sensor detection either:

- Globally (for all of PRTG) - On a device-by-device basis

Neither of these options work particularly well (at least for us - but I would have to imagine most Paessler customers).

We want/need unusual sensor detection enabled globally, and we want/need it enabled for all devices but we would like to disable it for specific sensors.

Right now we have 205 sensors in an "unusual" state. As you might guess its practically impossible to find legitimate unusual sensors in all the noise.

The majority of unusuals are sensors monitoring windows services. They are detecting unusual sensor execution time. Frankly we couldn't care less how long it takes PRTG to check a service status and I can't imagine who would.

Why the windows service "sensor execution time" even triggers an unusual is beyond me. What we care about is if a service is started or stopped. We'd like to disable unusual sensor status for all windows service sensors.

The next most frequent unusual sensor is bandwidth related. We have sensors monitoring network bandwidth usage on individual network cards in VMs and Physical Servers as well as on our Network Equipment (firewalls, network switches etc).

The thing is that traffic can fluctuate on some of these ports frequently. So it would be nice if we could disable unusuals on a port by port (sensor by sensor) basis so that we have fewer unusuals showing up.

Article Comments

Hi Brad,

Thanks for sharing your detailed feature request. I just created an internal ticket, please bear with me that I cannot tell you if or when this is going to make it into PRTG.

Best regards, Felix


Jun, 2018 - Permalink

Hi everyone,

This feature request has been converted into an official feature request: https://helpdesk.paessler.com/en/support/solutions/articles/81141-enable-disable-unusual-detection-on-sensor-level

Please read our new guide on how we want to handle requests in the future. Thanks!

Kind regards,
Stephan Linke, Tech Support Team


Aug, 2018 - Permalink

This one is a year old now and is the third highest requested feature (behind AD integration and performance) - any movement here. Unusual detection is super useful for some sensors and has given us early warnings about problems - but they are buried in the noise...

I'd also like a way to reset the unusual status (or at least acknowledge for X hours, days, weeks) - because as useful as it is, it can be super distracting to have poinless detections (response time is unusually LOW for this weekday - YAY thats a good thing!) - or it taking weeks to stop getting alerts after you make a change.

Yes I know you can clone a sensor - but until you can "clone and replace" - I have to go through all my sensor factories and maps and re-do everything.

Please fix this. Pretty please....


Jun, 2019 - Permalink

It's indeed one of the features requested more often and we also opened internal development tickets for that. However, they've been prioritized lower than other things that require our attention. We still have it on our radar, but I can't give you an ETA on it, unfortunately :(


Jun, 2019 - Permalink