namespace defaults to in-cluster namespace when run with explicit `KUBECONFIG`

[hojerst]

When a KUBECONFIG is set explicitly to a external cluster, the client still defaults to the namespace mounted in /run/secrets/kubernetes.io/serviceaccount. This means the same kubeconfig setting/file will behave differently depending if it is run inside or outside of a cluster.

IMHO this should be changed to default to the default namespace as if run outside of the cluster. Defaulting to the serviceaccount's namespace doesn't really make sense here as the referenced cluster might not even have this namespace and it leads to inconsistent and unexpected behaviour.

[porridge]

It's funny how the logic was here for years, and I got down to tracking down this behaviour to this logic where the library explicitly avoids using the implicit default namespace today and noticed your issue just as I was about to file my own, just hours after yours :-)

To provide an example of where this happens is when one runs KUBECONFIG=/some/other/clusters/config operator-sdk scorecard in a pod in a namespace other than default, and the provided kubeconfig does not explicitly provide a namespace (which is typical I think).

[hojerst]

(sorry for the off-topic, just scroll on - but this needs to be said): whats even more funny is the fact that my whole endeavor finding this issue was intially caused by wrong kubeconfigs in kapp inside kuttl runs - an issue that you seem to have reported and fixed, but I was running an older version of kuttl with the bug present. If I would have updated, I would never run into this. Small small world...

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[porridge]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[porridge]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle rotten
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[porridge]

/remove-lifecycle rotten