Prometheus fails to scrape internal metrics since 0.118

[vinzent]

Since we've upgraded to 0.118 (from 0.112) we cannot scrape metrics anymore with our internal cluster monitoring (OpenShift Monitoring, Prometheus).

Seems this is related to the change in #1573 which binds metrics to localhost only.

Reading https://opentelemetry.io/docs/collector/internal-telemetry/#configure-internal-metrics :

Alternatively, you can expose the Prometheus endpoint to one specific or all network interfaces when needed. For containerized environments, you might want to expose this port on a public interface.

If I change it manually to 0.0.0.0 scraping with the cluster-internal Prometheus works again.

I think this should be configurable to be able to allow access to metrics for Prometheus scraping.

[atoulme]

Please open a support ticket to get help. Issues are not monitored and only kept around for archival.