cleanup docs post GCS infra migration

[neolit123]

see this and linked tracked issues:

• fix: use dl.k8s.io, not kubernetes-release bucket #2872
• fix: use dl.k8s.io, not kubernetes-release bucket #2872 (comment)
• Deprecate and migrate away from gs://kubernetes-release k8s.io#2396 (when this is closed we can update k/kubeadm)

the k/kubeadm docs contain some outdated examples. the PR above fixes some of them, but there is remaining work.

• remove / change all gs:// entries in docs
• remove /cluster/* script examples in docs
  docs: cleanup reference to /cluster/ scripts #2878

note we cannot change gs://k8s-release-dev/ci/
yet:
#2872 (comment)

so this issue tracks future work.
but the /cluster change can be PRed now.

[pacoxu]

/assign @rjsadow

[chendave]

Just a record, here is still reference to gcs bucket,

kubeadm/docs/testing-pre-releases.md

Lines 117 to 119 in 3d02f6c

	- Kubernetes release labels for GA, alpha/beta versions like e.g. `stable`, `stable-1`, `stable-1.10`, `latest`,
	`latest-1`, `latest-1.10`. See <https://console.cloud.google.com/storage/browser/kubernetes-release/release/>
	for the full list of labels.

Is there any alternative to replace it? https://console.cloud.google.com/storage/browser/kubernetes-release/release/

When we have any other options, we can update the link above.

[chendave]

remove / change all gs:// entries in docs

@neolit123 what do you mean for this todo item? I searched in the code base, seems this has been addressed unless the gs://k8s-release-dev/ci/ (I understood this will stay for an while).

[neolit123]

Is there any alternative to replace it?

@rjsadow do you know if we can list these labels with curl?

[neolit123]

remove / change all gs:// entries in docs

@neolit123 what do you mean for this todo item? I searched in the code base, seems this has been addressed unless the gs://k8s-release-dev/ci/ (I understood this will stay for an while).

we can't change it for now, afaik. it was discussed earlier as planned.

[rjsadow]

I don't believe there's a direct 1-for-1 replacement with curl in this case since directory browsing isn't available through the proxy. It might be worth reaching out to sig-release and seeing if we could add an index file or some other mechanism into the release bucket that provides a list of all latest and stable versions.

[neolit123]

i think we can just remove this sentence, as the prev one already includes the most relevant labels

See https://console.cloud.google.com/storage/browser/kubernetes-release/release/
for the full list of labels.

[rjsadow]

Let me take a crack at rewording it. The problem with removing it and referencing the release page is that only includes GA versions. If someone wants to test with a pre-release version they really should reference these files to get the tags to use.

i.e.

❯ curl -L https://dl.k8s.io/release/latest-1.txt
v1.28.0-alpha.2

Which is the tag that should be used for the images. I don't know any better way to point someone to pre-release image tags.

[neolit123]

Which is the tag that should be used for the images. I don't know any better way to point someone to pre-release image tags.

i don't know either for pre built ci images. in kubeadm e2e tests we build images from tars for a particular k/k commit. those tars are for sure available in the current ci bucket.

[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

[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