Helm chart metrics broken?

[dhess]

Hi, thanks for this very useful project.

Some of the changes in #133 seem to be related to making metrics more secure. I believe the Helm chart needs to be updated to support the new configuration options, because since deploying v0.7.0, our metrics are now broken:

cloudflare-zero-trust-operator 2025/01/07 23:23:28 http: TLS handshake error from 127.0.0.1:47230: client sent an HTTP request to an HTTPS server

We use a ServiceMonitor similar to the one you've defined here: https://github.com/BojanZelic/cloudflare-zero-trust-operator/blob/main/config/prometheus/monitor.yaml

It has worked fine up until the v0.7.0 upgrade, and now repeatedly gives the error message shown above. I assume that either --metrics-secure=false needs to be set, or port 8443 needs to be used for metrics and the kube-rbac-proxy config changed to match.

[BojanZelic]

This should be fixed by #147

we removed the kube-rbac-proxy component since it's not needed anymore; metrics scraping should happen over 8443

Helm chart v0.7.1 should have the fix

[dhess]

Thanks! Can you publish the new Helm chart, please?

[BojanZelic]

There's a new chart published https://github.com/zelic-io/charts/releases/tag/cloudflare-zero-trust-operator-0.7.1

[dhess]

Confirmed fixed, thank you!