Feat: Add Multi-Tenant Manager to CCM by priyapande · Pull Request #935 · kubernetes/cloud-provider-gcp

priyapande · 2025-12-05T22:51:30Z

Add support for a tenant controller manager that launches node controllers separately for each tenant that are scoped to it's tenant authentication & filtered nodes view.

k8s-ci-robot · 2025-12-05T22:51:40Z

This issue is currently awaiting triage.

If the repository mantainers determine this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Details

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-sigs/prow repository.

k8s-ci-robot · 2025-12-05T22:51:41Z

Welcome @priyapande!

It looks like this is your first PR to kubernetes/cloud-provider-gcp 🎉. Please refer to our pull request process documentation to help your PR have a smooth ride to approval.

You will be prompted by a bot to use commands during the review process. Do not be afraid to follow the prompts! It is okay to experiment. Here is the bot commands documentation.

You can also check if kubernetes/cloud-provider-gcp has its own contribution guidelines.

You may want to refer to our testing guide if you run into trouble with your tests not passing.

If you are having difficulty getting your pull request seen, please follow the recommended escalation practices. Also, for tips and tricks in the contribution process you may want to read the Kubernetes contributor cheat sheet. We want to make sure your contribution gets all the attention it needs!

Thank you, and welcome to Kubernetes. 😃

k8s-ci-robot · 2025-12-05T22:51:42Z

Hi @priyapande. Thanks for your PR.

I'm waiting for a github.com member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Details

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-sigs/prow repository.

pkg/apis/providerconfig/register.go

bowei · 2025-12-08T17:53:08Z

/assign

k8s-ci-robot · 2025-12-10T00:04:39Z

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: priyapande
Once this PR has been reviewed and has the lgtm label, please ask for approval from bowei. For more information see the Code Review Process.

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

Details

Needs approval from an approver in each of these files:

OWNERS

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

bowei · 2025-12-10T00:11:43Z

Can we split off the vendor change in a separate commit? Right now it's 300 files, but I'm assuming most of it is in vendor/

cmd/cloud-controller-manager/main.go

pkg/controller/nodemanager/cloud.go

pkg/controller/nodemanager/filteredfactory.go

pkg/controller/nodemanager/filteredinformer.go

aojea · 2025-12-10T15:01:25Z

pkg/controller/nodemanager/starter.go

+	klog.Infof("[%s] Creating OSS Cloud Node Controller...", pcKey)
+	nodeController, err := node.NewCloudNodeController(
+		filteredFactory.Core().V1().Nodes(),
+		m.kubeClient,
+		scopedCloud,
+		m.config.ComponentConfig.NodeStatusUpdateFrequency.Duration,
+		m.config.ComponentConfig.NodeController.ConcurrentNodeSyncs,
+	)
+	if err != nil {


there has to be two independent launchers, and users should opt-in via the standards flags to enable one or other controllers, check this for prior art https://github.com/kubernetes/cloud-provider-gcp/pull/895/files

The litmus test is that the addition of this new controller/folder will have zero impact, and the behavior is governed by the flags

cmd/cloud-controller-manager/main.go

aojea · 2025-12-10T21:46:34Z

We met offline, to summarize our discussion, try to follow the same approached used in #895

leverage the initializers behavior to have two different nodecontrollers, with the new one disabled by default, this will allow users to use the existing flags semantics to enable or disable one or other controller


--controllers strings     Default: "*"
--
  | A list of controllers to enable. '*' enables all on-by-default controllers, 'foo' enables the controller named 'foo', '-foo' disables the controller named 'foo'.All controllers: bootstrap-signer-controller, certificatesigningrequest-a

avoid mixing code as much as possible:
prefix files on shared folders eg. cmd/gkemultinodecontroller.go
prefix folders that are not shared with oss code eg. /pkg/controller/gkenodemanager
do not export libraries , if some library needs to be exported within the project just nest it under internal eg./pkg/internal/informers
do not commit generated libraries or CRDs or any other type of API , instead use an external repository and revendor it here , see use the GKE networking CRDs from GoogleCloudPlatform platform repo #725

go.mod

aojea · 2026-02-06T09:41:04Z

cmd/cloud-controller-manager/gketenantcontrollermanager.go

+		klog.Errorf("Failed to create network client: %v", err)
+		return nil, false, err
+	}
+	networkInformerFactory := networkinformers.NewSharedInformerFactory(networkClient, 12*time.Hour)


are we doing this resync every 12 hours on purpose? this has created some internal issues in other. controllers and nobody was able to explain me the reason to add this .. if we want to periodically go over the elements in the cache then this should be much lower, if we do not need that then this should be 0

That's a valid point, changed it to zero as the controller can rely on the event stream to process the required changes and a full walkthrough of cache isn't necessary. Thanks

cmd/cloud-controller-manager/gketenantcontrollermanager.go

aojea · 2026-02-06T09:46:10Z

pkg/controller/nodeipam/starter.go

+
+// StartNodeIpamController starts the NodeIPAM controller.
+// It returns the controller interface, a boolean indicating if it started, and an error if any.
+func StartNodeIpamController(


it feels like this nodeipam starts should not be mixed in the same commit or even the same PR, if something goes wrong we have to revert the whole thing

moved refactoring & vendor logic to separate PR - #966

k8s-ci-robot added do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. needs-triage Indicates an issue or PR lacks a `triage/foo` label and requires one. labels Dec 5, 2025

k8s-ci-robot added needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. labels Dec 5, 2025

k8s-ci-robot added the needs-ok-to-test Indicates a PR that requires an org member to verify it is safe to test. label Dec 5, 2025

k8s-ci-robot requested review from bridgetkromhout and cici37 December 5, 2025 22:51

k8s-ci-robot added sig/network Categorizes an issue or PR as relevant to SIG Network. size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files. labels Dec 5, 2025

aojea reviewed Dec 6, 2025

View reviewed changes

pkg/apis/providerconfig/register.go Outdated Show resolved Hide resolved

k8s-ci-robot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Dec 6, 2025

k8s-ci-robot assigned bowei Dec 8, 2025

priyapande force-pushed the ccm-poc branch from 12b46be to fe0b6b9 Compare December 10, 2025 00:04

k8s-ci-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Dec 10, 2025