Hi, i think that is a feature request.

Notification Triggers "When sensor state is DOWN for at least 120 seconds, perform..." I would like to have the possibility to replace the "120 seconds" by a variable. Perfect would be the double value of the interval from the individuell sensor. So maybe in the root path stands that: "When sensor state is DOWN for at least '2x Sensor Intervall Scan Time", perform..." So it doesn't matter if one sensor has 60s interval or 5 minutes for scanning his state. So, it is possible for every sensor to recover in his own second interval time, without a notification is send.

thanks! best regards Matthias

Article Comments

Hello,

in order to delay the down status scaled by the scanning interval, there is an option (by default inherited from the parent object) that for X (with X up to 5) scan the sensor switches to warning first, instead of down. You find this in the same settings section where the scanning interval is defined.

This does not work with channel limits, these change the sensor status instantly. (Otherwise, with umpteen channels possible per sensor, this could cause a considerable impact on performance.) Also, with channel limits, the trend is usually roughly linear on short time scales, meaning there is a high chance the sensor will stay down anyway, so why not changing the status instantly.

But the aforementioned option does work with scan errors, allowing to reduce noise.


Jan, 2022 - Permalink

Hi, i guess you didn't understand exactly what i mean. I don't care about the status, but the notification, which is send after a permanent delay. That delay I have to set in the notification trigger settings, e.g. 120s. That value should be a parameter, which reads out the scan interval of the sensor and multiplied it with factor two. Based on that, the notification for "down" will be send always after the double of the scan interval time. Regardless of whether the interval is 120s or 5min.


Jan, 2022 - Permalink

Hello,

as the notification triggers on the status change, delaying the status change to "down" based on scans and hence the scanning interval is the closest thing possible with PRTG.

You can create feature requests this way, with something similar already implemented I am not sure about chances of implementation anytime soon.


Jan, 2022 - Permalink