Alvaro Aleman

Removing good first issue label, as it is not quite clear if we want this and if so, how it would look. /remove-good-first-issue

> Got it. We should definitely fix the behavior, returning an error when using InNamespace predicate on non-namespaced resources makes sense to me. sounds good to me

Yeah I agree. Doing this for CRDs (recognized by not being recognized by client-go scheme) seems fine. For everything else, it seems a lot of work and might fall out...

> he k8s code to help enforce requirements on the PolicyRule (see PolicyRuleBuilder) but importing this requires a bunch of replaces added into go.mod. Is this viable/okay to proceed with?...

Hey @howardjohn , The reason this logic exists is that the `DelegatingLogger` ends up consuming a huge amount of memory if no Logger gets set: https://github.com/kubernetes-sigs/controller-runtime/issues/1122 The current logic is...

Deduplicating the rules with a nested map of apiGroup -> resource -> resourceNames -> verbs should be relatively easy. One caveat is that maps are unordered. Then again this is...

What exactly should it have triggered in https://github.com/kubernetes-sigs/cluster-api/pull/6362 but didn't? /cc @shinigambit

/label tide/merge-method-squash

> My read of current pod-utils code is that it assumes base_ref to always be present / non-empty (ref: #20665 (comment)) Yeah, that is correct (for periodics, for presubmits/postubmits we...

Can you post the yaml of such a pod, please?