Support kubernetes_role argument for prometheus.operator.servicemonitors

[yoyosir]

PR Description

This PR adds support for endpointslice role in prometheus.operator.servicemonitors. Before this PR, the role defaults to endpoints. So there was no way to have servicemonitor to scrape an endpointslice in alloy.

Example config in alloy after this PR.

prometheus.operator.servicemonitors "servicemonitors_endpointslice" {
  forward_to = [prometheus.relabel.servicemonitors_filter.receiver]
  selector {
    match_expression {
      key      = "kubernetes-scrape-role"
      operator = "In"
      values   = ["endpointslice"]
    }
  }
  clustering {
      enabled = true
  }

  kubernetes_role = "endpointslice"
}

prometheus.operator.servicemonitors "servicemonitors" {
  forward_to = [prometheus.relabel.servicemonitors_filter.receiver]
  selector {
    match_expression {
      key      = "kubernetes-scrape-role"
      operator = "NotIn"
      values   = ["endpointslice"]
    }
  }

  clustering {
      enabled = true
  }

}

Which issue(s) this PR fixes

Fixes #1714

Notes to the Reviewer

PR Checklist

• CHANGELOG.md updated
• Documentation added
• Tests updated
• Config converters updated

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
1 out of 2 committers have signed the CLA.

✅ yoyosir
❌ Zheng Yong

---

Zheng Yong seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[thampiotr]

Thanks! I believe that your GitHub user is different to the one used to create Git commits and because of that the CLA signing is not passing. Could you try making sure you use the email associated with your GitHub account? This can be done without exposing your actual email address.

[thampiotr]

We will need to document this in user-facing docs:

• https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.podmonitors.md
• https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.servicemonitors.md
• https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.probes.md

[yoyosir]

Thanks for the review @thampiotr ! Just realized my previous changelog and docs were not updated. Updated those. I'll change the other monitors that use this argument.

One question: Since my previous commit email cannot be changed, should I close this and open up a new PR instead?

[yoyosir]

Actually one more question. I noticed for probes and podmonitors it seems more proper to have the role fixed to ingress and pod respectively. I'll leave the decision to you on whether I should enable the role override for
https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.podmonitors.md
https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.probes.md

[yoyosir]

@thampiotr I created a new one #2023 seems the CLA is passing there.

[thampiotr]

Actually one more question. I noticed for probes and podmonitors it seems more proper to have the role fixed to ingress and pod respectively. I'll leave the decision to you on whether I should enable the role override for https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.podmonitors.md https://github.com/grafana/alloy/blob/main/docs/sources/reference/components/prometheus/prometheus.operator.probes.md

Let's leave them as they are and only change servicemonitors for now.

[thampiotr]

@thampiotr I created a new one #2023 seems the CLA is passing there.

Cool, let's move conversation there.