Showing results for 
Search instead for 
Did you mean: 
Create Post

Schedule silencing of alerts while still monitoring

Schedule silencing of alerts while still monitoring

It would be extremely helpful to schedule when alerts for a node can be silenced; such as for periodic reboots, CPU intensive jobs, maintenance windows, etc.  There may be a 4 hour maintenance window that a node could be rebooted in or all resources maxes out.  It is a known event and we do not want the email notifications, but still want the node to be monitored to obtain statistical data on it.

While kludgey, the unmanage scheduling utility comes close to doing this, but it completely stops polling the node, which then causes us to lose all information on the node during that time, such as when the reboot actually took place, or how long resources were used for.

Unmaged Data loss.png

It would be nice if the scheduling utility had the option to just mute alerts (like under a node's management section), or the alerting section of the NPM website had a scheduling section.

Mute alerts.png

I found this post ( which indicates that it was implemented in version 12.1, but it's no clear how to use it or where it was implemented.  I submitted Case #1355911 to get additional instructions and was told this was only implemented in the older alert manager (win32 app), but this is not available in the newer versions of NPM with the alerting built into the web interface.  I was then directed to this forum.

Level 13

I am on NPM 12.1 think your looking for this? its within the 'schedule'


Level 8

Yes, sortof.  There is no way to make this a recurring event, for example - weekly reboots, nightly intensive jobs, etc.  That option will only mute the alerts for one given period.

The unmanage utility allows one to make this a recurring "event", but completely stops polling, the node, which is NOT want we want.

I am on NPM 12.2

Level 12

You could do this with the alert options in Time of Day:


Or you could also set up custom properties to mute interfaces and alerts in the alert logic, set a script to change those fields at certain times. This behavior while not necessarily the easiest to implement seems like it is possible to implement currently.

Here is a good thread that goes into some detail bit old but I think it could be useful: TIPS & TRICKS: Stop the madness! Avoiding alerts but continuing to pull statistics.

Level 8

Thanks for the ideas.  Unfortunately disabling the alert (eg node down) during a specified window would mean disabling it for all nodes/devices, when I only want to silence a handful of servers every week, and all at different times.  The alternative would be to create an alert for each server, which would then get complicated and messy.

I'm leery of implementing anything too complicated (like the scripts and custom fields) because this also needs to be easy for our junior admins to modify.  For example, when they schedule patch groups to deploy updates, which the days change on every month.

As i_like_eggs​ pointed out, the scheduling is already in NPM, it just needs a "recurring" option.

for those who like pictures. there should be some form of "muting" in this tool


Otherwise bring the Scheduling tool into the web-funcionality

Level 8

This Screen makes the system completely unmanage status from that point of time it would not track any information of the system. it is not providing what is running background of the machine. 

Level 8

HerrDoktor​ this is unmanaging the node. when I manage the node sam is not collecting any data from the node. there is not recurring scheduling of muting alerts.

Yes, I am fully aware of this. I was pointing out that there is a tool that COULD possibly do the job, but it lacks the Option to MUTE. My post is NOT a solution to your FR, just a visualization for those who like pictures

Community Manager
Community Manager
Status changed to: Open for Voting