CSI Drivers in Windows as HostProcess Pods

[mauriciopoppe]

Enhancement Description

• One-line enhancement description (can be used as a release note): CSI Drivers in Windows as HostProcess Pods
• Kubernetes Enhancement Proposal:
• Discussion Link: https://docs.google.com/document/d/1_WtQUnnlXE_M-YzYVrPZbjEgSxMvY9EYFwl61MC2i5I/edit#bookmark=id.m6yzw8n3okss
• Primary contact (assignee): @mauriciopoppe
• Responsible SIGs: sig-storage
• Enhancement target (which target equals to which milestone): This enhancement is for an out of tree component, as such, its development is not tied to a Kubernetes release, @msau42 suggested creating a KEP to track the change because it's a big refactor that all CSI Drivers need to do and the design superseeds the CSI Proxy design added in https://github.com/kubernetes/enhancements/blob/master/keps/sig-windows/1122-windows-csi-support/README.md.
• Alpha
  • KEP (k/enhancements) update PR(s): KEP-3636 CSI Drivers in Windows as HostProcess Pods #3641
  • Code (k/k) update PR(s): (This is an out of tree component and doesn't need changes in k/k)
    • CSI Proxy tracking issue: Run Windows node CSI Drivers as HostProcess containers kubernetes-csi/csi-proxy#217
  • Docs (k/website) update PR(s):
    • Updating [Windows Storage](https://kubernetes.io/docs/concepts/storage/windows-storage/ storage) docs: TBD
    • Blogpost: TBD

[mauriciopoppe]

/sig storage

[mauriciopoppe]

/sig windows

[marosset]

/cc

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs 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 or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[alexander-ding]

/remove-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

[mauriciopoppe]

/remove-lifecycle stale

[jayunit100]

what ever happened to this one?

[mauriciopoppe]

hey @jayunit100, I captured the current state of the integration at https://docs.google.com/document/d/1_WtQUnnlXE_M-YzYVrPZbjEgSxMvY9EYFwl61MC2i5I/edit#bookmark=id.ww7xe8kiazif.

From a development point of view I think we're done (thanks to @alexander-ding), there were a few problems about the integration with GKE:

• security: GKE autopilot and the privilege level of the DaemonSet
• node skew: on k8s upgrades we bump the control plane components first including DaemonSets, a concern we had is that we could end up in an scenario where the CP is upgraded to 1.X and the new DaemonSet deployed but the nodes are still 1.X-2. This was a problem before because the kubelet would be too old, I think that now if the scenario is CP 1.27, nodepools 1.25 then it works so this problem is not there anymore.

The node skew scenario is no longer a problem, the security scenario is a problem specific to GKE but I believe any CSI Driver which makes the component privileged would have this problem, I think I need to expand more on the suggestions.

Another problem is defining the graduation criteria, devs from AWS were trying to integrate this work, I think Azuredisk is already running the CSI Driver as a privileged container but without using the new version of CSI Proxy. Having more CSI Drivers using the library would help me define the graduation criteria e.g. "Graduation Criteria: At least 2 CSI Drivers are using the new version of CSI Proxy".

[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

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages 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:

• Reopen this issue with /reopen
• Mark this issue as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

[k8s-ci-robot]

@k8s-triage-robot: Closing this issue, marking it as "Not Planned".

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages 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:

• Reopen this issue with /reopen
• Mark this issue as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

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.

[mauriciopoppe]

/reopen

[k8s-ci-robot]

@mauriciopoppe: Reopened this issue.

In response to this:

/reopen

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.

[mauriciopoppe]

/remove-lifecycle rotten

[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