I'm collecting metrics from a few servers. I'd like to somehow monitor that servers are running and reporting data. I'm migrating from Sensu, where the concept of a server is an explicit thing and when that thing stops sending data, it automatically triggers an event. I'm struggling to do something similar with opentelemetry/signoz. I see that the infra overview has a status, what is that based on, can I alert on that somehow?
s
Srikanth Chekuri
02/03/2025, 2:51 PM
To alert on, you take a metric from the server and see if it is still being reported by the host by comparing it to the same with some time shift.
s
Sascha Grossenbacher
02/03/2025, 3:06 PM
thanks, didn't know about the time shift thing, interesting, set it up, looks good, will need to test it. I guess once the server hasn't been reporting for the configured time shift, it will no longer report anything for that server and the alert would resolve?
s
Srikanth Chekuri
02/03/2025, 3:07 PM
yes
s
Sascha Grossenbacher
02/03/2025, 3:09 PM
makes sense, need a way to resolve the alert when I remove a server. and I can adjust the time shift to have it alert for a longer time.