[Collector-chart] Self Monitoring not Setting Job name

[Syndlex]

I updated from 0.116.0 to 0.117.12, and now the job name used to scrape its own configuration is not working anymore.

This is my setup:

- job_name: telemetry/internal_service_name
  scrape_interval: 20s
  static_configs:
    - targets:
        - ${env:MY_POD_IP}:8888

I know that the Prometheus Scraper was updated, so I think the problem is there. The job name is always shown as job="otelcol-contrib".

Also, service.name is now exported in the labels as: otelcol-contrib;internal_service_name.

Does anyone have an idea of what changed?

I tried using just - job_name: internal_service_name and also overwriting it with relabel_configs, but neither worked.

[TylerHelmuth]

@Syndlex can you confirm that you still have metrics being emitted from the collector at ${env:MY_POD_IP}:8888? Does your collector have service.telemetry.metrics.address set?

[Syndlex]

@TylerHelmuth service.telemetry.metrics.address yes I still get metrics. But the job label is broken.

[TylerHelmuth]

The chart didn't change anything, maybe this is a bug in the prometheus receiver?

[Syndlex]

yes. I think its this issue: open-telemetry/opentelemetry-collector-contrib#38097