Robert Pająk
Robert Pająk
@open-telemetry/technical-committee, yes it is still applicable. What is more concurrent reads can also be unsafe if the log record was synchronously modified in a processor. That is why in Go...
@dashpole, are you able to take this?
The idea is to add `otel_scope_[attribute]` and `otel_scheme_url` labels to metrics. I will do my best to help with this issue.
A prototype in Go: https://github.com/open-telemetry/opentelemetry-go/pull/5947
Based on the prototype, I think this could be split into 4 sub-issues: 1. Handle scheme URL as identifying 2. Handle scope attributes as identifying 3. ~~Define a temporary env...
@open-telemetry/prometheus-interoperability, @open-telemetry/collector-maintainers: PTAL
> (Unrelated, but what is the distinction between sub-issues 1 and 2 above?) @jade-guiton-dd I have forgotten to delete "and scope attributes"` from sub-issue 1 😅 Fixed 😉 I will...
@jade-guiton-dd, I think that Prometheus receiver in collector-contrib is also crucial (not only exporter).
> did you mean a single prototype where the draft PR modifying the SDK exporter is integrated into the Collector to test that it fixes our issue when producing scope...
I updated the OTel Go's exporter implementation here: https://github.com/open-telemetry/opentelemetry-go/pull/5947. PTAL