dockershim/network: pass ipRange CNI capabilities

[squeed]

What this PR does / why we need it:
Updates the dynamic (capability args) passed from Kubernetes to the CNI plugin. This means CNI plugin authors can offer more features and / or reduce their dependency on the APIServer.

Currently, we only pass the portMappings capability. CNI now supports bandwidth for bandwidth limiting and ipRanges for preferred IP blocks. This PR adds support for these two new capabilities.

Bandwidth limits are provided - as implemented in kubenet - via the pod annotations kubernetes.io/ingress-bandwidth and kubernetes.io/egress-bandwidth.

The ipRanges field simply passes the PodCIDR. This does mean that we need to change the NodeReady algorithm. Previously, we would only set NodeNotReady on missing PodCIDR when using Kubenet. Now, if the CNI configuration includes the ipRanges capability, we need to do the same.

Which issue(s) this PR fixes:
Fixes #64393

Release note:

The dockershim now sets the "bandwidth" and "ipRanges" CNI capabilities (dynamic parameters). Plugin authors and administrators can now take advantage of this by updating their CNI configuration file. For more information, see the [CNI docs](https://github.com/containernetworking/cni/blob/master/CONVENTIONS.md#dynamic-plugin-specific-fields-capabilities--runtime-configuration)

[squeed]

/sig network
/sig node

[kragniz]

/ok-to-test

[squeed]

/retest

[squeed]

CI is all green :-)

[squeed]

@kubernetes/sig-network-pr-reviews this is ready for review.

[k8s-ci-robot]

@squeed: Reiterating the mentions to trigger a notification:
@kubernetes/sig-network-pr-reviews

In response to this:

@kubernetes/sig-network-pr-reviews this is ready for review.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[krmayankk]

Will this allow scheduler to consider ip available on a node when making scheduling decisions

[squeed]

@krmayankk no, this doesn't change anything with the scheduler. It just makes it easier for CNI plugins to get the PodCIDR. Right now, CNI plugins have to contact the apiserver; this removes that round-trip.

[squeed]

It seems that bandwidth shaping has also been added in #63194 - I'll rework this PR.

[m1093782566]

It seems that bandwidth shaping has also been added in #63194 - I'll rework this PR.

Thanks @squeed :)

[squeed]

Still waiting for #63194 to be merged; CI is green, it's just pending review.

[squeed]

OK, blocker has been merged; this PR is ready for review and merge again.

[squeed]

Flake: #65466

/retest

[m1093782566]

/lgtm

[squeed]

/retest

[squeed]

Paging @dcbw - can I get an approval?

[dcbw]

Not all CNI plugins will actually want --allocate-node-cidrs, it really depends on what happens to be in /etc/cni/net.d or what happens to get written there when the net plugin's daemonset starts. So I don't think we can assume CNI==allocate-node-cidrs. Maybe have another environment variable for local-up-cluster.sh that turns this option on if NET_PLUGIN=cni?

[squeed]

Update PR to remove change to hack/local-up-cluster.sh per advice from @dcbw.

This is ready for final review and merge (whenever the flaky tests pass).

[squeed]

/retest

Has CI been green in months?

[dcbw]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: dcbw, m1093782566, squeed

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• pkg/kubelet/dockershim/network/OWNERS [dcbw]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[k8s-github-robot]

/test all [submit-queue is verifying that this PR is safe to merge]

[k8s-ci-robot]

@squeed: The following test failed, say /retest to rerun them all:

Test name	Commit	Details	Rerun command
pull-kubernetes-bazel-test	5d9ec20	link	/test pull-kubernetes-bazel-test

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

[k8s-github-robot]

Automatic merge from submit-queue (batch tested with PRs 64445, 67459, 67434). If you want to cherry-pick this change to another branch, please follow the instructions here.