Your data. Anywhere you go.

New Relic for iOS or Android


Download on the App Store    Android App on Google play


New Relic Insights App for iOS


Download on the App Store


Learn more

Close icon

Feature Idea: Windows Service Monitoring

windowsservice
windows
feature-idea
rfb

#1

Vote Now! | Windows ServiceMonitoring

  • I want this, too
  • I have more info to share (reply below)
  • I have a solution for this

0 voters

We take feature ideas seriously and our product managers review every one when plotting their roadmaps. However, there is no guarantee this feature will be implemented. This post ensures the idea is put on the table and discussed though. So please vote and share your extra details with our team


Windows service monitoring?
#2

Is there a way to monitor specific Windows services currently? (alerts when a windows service stops etc)


#3

Hey @tgohl

Unfortunately at this point we do not have an availability monitor for specific processes in our server monitor. Our PM team is actively monitoring these forums for feature requests such as these.


#4

I would second the request. Would be a very useful feature.

I already ensure our windows services are running using puppet, but running doesn’t always mean responsive.

You could link to an online log monitor that alerts - like papertrail or sumo (other online logging tools are available). We currently trigger alerts from errors in the logs in this way as a method of finding when there are problems even when the service is running.

-Martin


#5

Hey Martin! I’ve gone ahead and logged your feature request for our product management team. Thank you for sharing your ideas and use case on this!


#6

+1 for this feature.


#7

I’ve added your +1, @stuart_taylor :slight_smile:


#8

I would vote this feature up as well! Good suggestion and great product so far.


#9

I’ve added your vote, @TechieJamie :slight_smile:


#10

Please include my vote too!


#11

Please add my vote too


#12

I took care of those votes @eric_hyh and @Trevor_Dearham! Although it is worth noting that with our new Alerts Beta, you can set an alert for any metric in New Relic. So, you could create an alert that goes off if a specific process metric gets too low, like in instances where it might have stopped!


#13

@pedwards At the moment we would have to build a plugin or using something like the New Relic Windows Permon Plugin to monitor the state of a Windows service and then alert on that, where New Relic is already pulling data about process memory and CPU usage and so could extent that to check for the state of services. The problem with the plugin route is that it, like custom dashboards, are separate entities, so I can’t see all my data in one location. I assume that the point of the new Alerts is so that custom dashboards and plugins can be dropped in favour of data being reported to Insights and still be alerted on?


#14

Please add my vote too.


#15

@Trevor_Dearham the new alerting system isn’t necessarily meant to replace custom dashboards and plugins. The aim is to provide lots more flexibility, especially in terms of which metrics can generate an alert. You could potentially alert on a process metric recorded by the Windows Server Monitor, e.g.:

ProcessSamples/{user name}/{process name}

You can set the policy so that conditions are violated when a specific value (average works well for the above) for the metric goes below a threshold.

@boris_netzer I’ll add your +1 to the request on Windows service monitoring for our product team to review.


#16

@QuietWyatt I understand that Alerts in and of itself isn’t intended to replace custom dashboards and plugin, but it is hopefully paving the road to having 3rd party metrics in the same dashboard as the rest of the metrics which Insights now allows.

In terms of the process metric, what is the meaning of the 6 values, as I assume that CPU and memory usage is in there?


#17

@Trevor_Dearham - We will need to do a little digging on our side to figure out that last question. I believe “An Average” maps to CPU % but I’m not sure what the memory metric maps to. There should also be the number of process instances as well.


#18

@Trevor_Dearham I don’t see a way to get per-process CPU and memory consumption metrics for use with the Alerts Beta at this point. I suggested the ProcessSamples/{user name}/{process name} metric since it allows you to alert on the metric recording itself. When the average for that metric drops below 1, you know that the process isn’t running and you can trigger an alert on this. We don’t yet have a way to single out CPU or memory on individual processes.


#19

@Tjd Thank you. It would be helpful to know what the other two values are, as in two case I’d like to monitor based on the number of instances of a process.

@QuietWyatt Thank you. I won’t need a per process value, as the check I want to perform is about the presence of a process or based on the number of instance, e.g. alert if below this number of instances or alert if above this number of instances.


#20

@Trevor_Dearham I’ve been playing with this metric feature in the alerting beta and haven’t been able to single out instance count. All values seem to refer to CPU usage. Average seems to correlate closely to the value you’d see in the normalized 0-100% CPU usage from Server Monitoring.

I’ll let our product team know that you’re interested in singling out instance count using metric-based alerting.