Im a new Netdata Business customer. I’ve been using the community version for some time, but wasn’t able to get any exposure to the Windows agent until this past Friday.
One thing I would love to do is have Netdata monitor the status of an executable and send an alert if the process dies.
I can see the process in the list of top CPU/memory consumers, so Netdata definitely sees it. I’m not sure how I can send an alert if it dies.
Even if doing this requires an integration (ideally free) to get it working…that’s OK. But I would prefer to not have to install anything else on an already resource constrained server (VPC).
The way you can alert on this is through the number of processes in such a group:
template: kafka-process-died
on: app.processes
chart labels: app_group=kafka
lookup: max -1m average unaligned of processes
units: process
every: 1m
crit: $processes < $this || $processes == nan || $last_collected < $now - 5
to: root
This would alert if there are no processes left or when the number of processes drops below the average over the last minute. In this case the app group being filtered on is Kafka.
Hi Constantine, how would I write the rule for the new Windows agent? I want to monitor specific processes if they are stopping. I also want to know a little more about how to trigger the alerts in the to: section. How does that work with Netdata cloud alerting?