Go to All Forums

Network Devices - Alert on Data Collection issues to montior

Hello,

I have found that when there are data collection issues on a network device, the monitor itself remains Green and does not alert.

You can find these issues under the on-premise Poller summary.

However, we don't constantly monitor the on-premise poller summary for each poller.

Is there a way to have the network device monitor itself change to trouble if there are data collection issues - IE, no data collected after an hour, or something like that?

That would be very apparent and could be actioned on quickly. If the monitor remains green, it gives the false appearance that the device is up and healthy. 

 

Like (1) Reply
Replies (2)

Hi Dennis,

Thank you for bringing this up. If data collection stops but the device monitor remains up, the issue is likely related to the On-Premise Poller’s data collection process rather than the device itself. As you mentioned, such issues are currently visible only under the On-Premise Poller Summary page.

To make these issues easier to track, we recommend configuring a threshold profile using the Alert if there is an issue with the data collection from monitors attribute in the On-Premise Poller's Threshold and Availability Profile. This will help you receive alerts whenever the On-Premise Poller encounters data collection problems.

We understand your request for the device monitor to automatically switch to Trouble when no data is collected for a specific duration. We’ll explore the feasibility of adding this as a built-in feature.

Best regards,
Rama
Like (0) Reply

Hi Rama,

 

We find the alert on the poller itself is problematic.

Currently, if a monitored port on a switch is down because it's just disconnected, the poller reports that as "Some interfaces are not polling as expected."

 

Of course, there are no metrics because the there is no responding device. It's not an issue with the interface, the switch, or anything else. That is what happens when no device is plugged in.

Therefore, that alert is not helpful or desired for us as "Some interfaces are not polling as expected." even when we would consider nothing wrong.

It has the effect of making the alert to you indicated as not helpful, because it would alert on switches that really don't have collection issues.

I would not able to find a way to muta that specific alert for the poller either.

Like (0) Reply

Was this post helpful?