Monitoring documentation incorrectly claims the number of running jobs can be queried
The documentation here: https://docs.gitlab.com/runner/monitoring/#configuration-of-the-metrics-http-server
Says:
The exposed information includes:
- Runner business logic metrics (e.g., the number of currently running jobs)
This is wrong: the Prometheus endpoint does not expose a metric giving the number of currently running jobs. I just wasted time setting it up because I wanted to get this one value, and it turns out not to exist...
In general it is very useful to be able to find out when jobs are running, so you can time system restarts and so on in a way that doesn't interrupt jobs.