JobSink controller logs Failed to export to Prometheus: "job-sink_client_latency" is not a valid metric name

[mgencur]

Describe the bug
The job-sink Pod repeatedly reports this:

2024/11/01 15:52:27 Failed to export to Prometheus: "job-sink_client_latency" is not a valid metric name
2024/11/01 15:52:34 Failed to export to Prometheus: "job-sink_client_results" is not a valid metric name
2024/11/01 15:52:34 Failed to export to Prometheus: "job-sink_client_latency" is not a valid metric name

Expected behavior
JobSink controller reports metrics properly.

To Reproduce
Deploy any JobSink resource.

Knative release version
1.15

Additional context
Add any other context about the problem here such as proposed priority

[pierDipi]

We need to remove the dash from the component name

eventing/cmd/jobsink/main.go

Line 68 in f53d038

const component = "job-sink"

/good-first-issue

[knative-prow]

@pierDipi:
This request has been marked as suitable for new contributors.

Please ensure the request meets the requirements listed here.

If this request no longer meets these requirements, the label can be removed
by commenting with the /remove-good-first-issue command.

Details

In response to this:

We need to remove the dash from the component name

eventing/cmd/jobsink/main.go

Line 68 in f53d038

const component = "job-sink"

/good-first-issue

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[yatesliang]

Is it okay to rename the component job_sink?
To obey the Prometheus naming rules: https://prometheus.io/docs/concepts/data_model/#metric-names-and-labels

[pierDipi]

@yatesliang yes!