Each time we pause sensors on a server for maintenance and then resume them, the WMI memory sensor goes into a Down state and triggers an alert email to be sent out. The Down state usually only lasts about 10-20 seconds and resolves its self but the email gets triggered regardless. The notification settings for Down state are set to 120 seconds, so this does not make sense.

Is there a way to avoid this false alert being sent out other than to disable sending alerts?

Thanks, Jon

Article Comments

What is the scanning interval set to for this WMI interval? You could try adjusting this to see if the scanning interval is too high and is scanning before the service can be contacted.

Also, you should not unpause these sensors until the machine is fully up. If these sensors are dependent on another sensor like Ping, you can delay the start-up of the WMI sensor by following the information here Delay WMI


Apr, 2013 - Permalink

Thanks for the reply but it does not appear that the custom PTF sensor mentioned in that post is still available.

Currently the scanning interval is set to every 10 minutes. I attempted to set a dependency on another sensor and set a 30 second delay but this did not help.

During my testing, there were times when the sensor never went into warning but the alert message was still sent. That is what makes no sense.

The alert message consists of two summarized notifications, one stating that the sensor was in error state with 0% memory available and the other (2 seconds later) says that memory is OK.

This only happens on the WMI Memory sensor and happens on all of our servers.

Thanks


Apr, 2013 - Permalink

It seems like this may be a larger problem than I thought.. Could you send a support ticket to support@paessler.com along with your logs?

Upload Logs


Apr, 2013 - Permalink