Hi
We are using custom metrics with Fly, and we noticed we stated having some data gaps. We thought it started happening when we splitted our app in two processes, web and worker, but after specifying the process in the metrics endpoint, the gaps are still there.
What’s the best way to debug this issue?
halfer
August 20, 2025, 5:38pm
2
Could this be a duplicate of the below post?
We’re experiencing a critical metrics collection issue with our service running on Fly. User-specific metrics have stopped appearing in our grafana monitoring dashboards (both internal and external) as a result.
Context
We are using https://api.fly.io/prometheus/ORG/ with proper authorization setup
CPU usage of the fly machines is normal, no performance issues
We seem to have partial metrics coming in
Suspected Cause
We suspect this might be a metrics cardinality limit issue either at:
Fly…
Yeah, based on the description I’d say it’s the same issue. Thanks @halfer . Do you know by any chance how to mitigate it?
system
Closed
August 28, 2025, 1:29pm
4
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.