Prometheus CI deploy keeps failing with startup timeout error

[Transcendental-Programmer]

Hi,

I saw everyone running into the ci check fail : one of my PRs (krkn-chaos/krkn-lib#163) fails when installing Prometheus, but then i noticed @tsebastiani PR krkn-chaos/krkn-lib#160 worked just fine—and neither of us changed the CI setup.

What’s going on?
When the Helm chart tries to install kube-prometheus-stack, it spits out:

Run # nodePort mapping 30000 -> http://localhost:9090
namespace/monitoring created
Error: INSTALLATION FAILED: 1 error occurred:
	* Prometheus.monitoring.coreos.com "kind-prometheus-kube-prome-prometheus" is invalid: spec.maximumStartupDurationSeconds: Invalid value: 0: spec.maximumStartupDurationSeconds in body should be greater than or equal to 60


Error: Process completed with exit code 1.

In short: the CRD now insists that maximumStartupDurationSeconds must be at least 60 seconds, but by default it’s 0, so it errors out.

Why did his PR pass but mine fail?

• maybe the Helm chart got updated between the two CI runs
• maybe our CI runners updated Kubernetes or the operator CRD
• either way, the default value (0) is now invalid under the new rules

How to fix

figured out we need to change code in .github/workflows/tests.yml after line 38

adding

  --set prometheus.prometheusSpec.maximumStartupDurationSeconds=300

References

prometheus-community/helm-charts#4585

Let me know if I’ve missed anything or i can go ahead and update the workflow. Thanks!