stop using the kubelet flag --pod-infra-container-image in kubeadm

[neolit123]

this is a pending TODO from:

• handle kubelet skew with and without dockershim support #2626

but the flag has proven problematic as it was also used to pin the pause image and prevent the kubelet GC from collecting it.
logic was added in kubelet, CRI and container runtimes to support this (but has a bug in containerd).

tracking issue in k/k;

• Clean up dockershim flags in the kubelet kubernetes#106893

some related PRs:

• kubelet: mark the deprecated pod-infra-container-image flag for removal in 1.35 kubernetes#122739
• [WIP] Remove --pod-infra-container-image from kubelet kubernetes#123144
• pkg/kubelet: allow sandbox image pinning from CRI kubernetes#118544

according to this commented we can only remove the flag completely when containerd version 2.0+ becomes widely used:

• kubelet: mark the deprecated pod-infra-container-image flag for removal in 1.35 kubernetes#122739 (comment)

bug outlined here:

• bug fix: make sure cri image is pinned when it is pulled outside cri containerd/containerd#9732

---

pending AIs are:

• remove the flag from the kubelet in release X
  TODO
• remove the flag when kubeadm deploys kubelet X, for other kubelets down to X-3 we must continue to pass the flag
  TODO
• remove the flag when kubeadm upgrades to kubelet X?
  TODO

[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 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

[SataQiu]

/remove-lifecycle stale

[AnmolxSingh]

Hi @SataQiu!
Is this issue still open?