Hi Team, I'm having a particular issue with a micro-service (springboot) that was instrumented (automatic instrumentation) and some metrics are not being seen. If I check /actuator/prometheus I see:

http_server_requests_seconds_count{exception="None",method="PUT",outcome="SUCCESS",status="200",uri="/api/v1/payment-orders/{id}/cancel",} 12.0
http_server_requests_seconds_sum{exception="None",method="PUT",outcome="SUCCESS",status="200",uri="/api/v1/payment-orders/{id}/cancel",} 0.919633034

But I I try to run PromQL

sum(rate(http_server_requests_seconds_count{uri=~"{{.uri}}"}[1m]))

I don't get any data. I've instrumented other services the same way and I'm getting that metric in particular. Also I'd like to mention that I'm getting traces from that service

What is the metrics collection interval? The reason could be your

[1m]

in the query. If they are collected at same or higher interval then you will not see any data.

I've tried with even 1 day

Make sure your

uri

is non null value that returns the results.

@Srikanth Chekuri will the actuator metrics be automatically captured? Or do we need to enable via receiver in opentelemetry-collector-metrics?

No, they are not automatically captured, but OP mentioned some metrics are not being seen, not that nothing is working. And it is more likely the

[1m]

selection in the promQL. If the metrics are collected every 60s or higher, then promql won’t return data. It has to be something higher. You can verify this by running

sum(rate(signoz_calls_total{}[1m]))

vs

sum(rate(signoz_calls_total{}[2m]))

in testbed.

yeah...I remember @Jose Infazon can you check by changing

[1m]

to

[2m]

in promql?