kubeadm: clean up exited containers and network checkpoints

[yujuhong]

No description provided.

[k8s-reviewable]

This change is 

[yujuhong]

Not very important, but just some extra cleanup.

By the way, the cleanup logic heavily depends on kubelet's internal implementation. I'd suggest moving away from that in the future if possible (yes, I know it's not easy...).

/cc @jbeda @mikedanese

[mikedanese]

/lgtm

[k8s-github-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: mikedanese, yujuhong

Needs approval from an approver in each of these OWNERS Files:

• cmd/kubeadm/OWNERS [mikedanese]

You can indicate your approval by writing /approve in a comment
You can cancel your approval by writing /approve cancel in a comment

[k8s-github-robot]

Automatic merge from submit-queue (batch tested with PRs 43508, 43836)

[luxas]

@yujuhong Cool, thanks. I noticed this issue yesterday during KubeCon when the kubelet started reporting unready status some mins after init after not being able to delete old networks that don't exist anymore. Not sure if that's a bug or feature.

However, for kubeadm this patch is critical for v1.6.1, can you make a cherrypick ASAP, please?

[yujuhong]

kubelet started reporting unready status some mins after init after not being able to delete old networks that don't exist anymore

I don't think this should happen. kubelet shouldn't become unready just because some pods/containers can't be cleanup. If you can reproduce it, feel free to file a bug.

I won't say this PR is critical, but it's always good to make sure you have a clean slate after running reset. I'll make a cherrypick PR.

[k8s-cherrypick-bot]

Commit found in the "release-1.6" branch appears to be this PR. Removing the "cherrypick-candidate" label. If this is an error find help to get your PR picked.