node-feature-discovery icon indicating copy to clipboard operation
node-feature-discovery copied to clipboard

Does NDF publishes the node features to the kubelet?

Open AlenK123 opened this issue 3 years ago • 6 comments

So pods can request a pci device or other hardware (similar to device plugin api)

AlenK123 avatar Jul 22 '22 10:07 AlenK123

So pods can request a pci device or other hardware (similar to device plugin api)

NFD does not register any device plugin resources. It can label nodes based on which PCI devices are available. These labels can then be used by device plugins daemonsets to get their pods only to those nodes that have the right devices.

mythi avatar Jul 22 '22 10:07 mythi

Are there any plans in the future for NDF to publish resources to the kubelet?

AlenK123 avatar Jul 22 '22 17:07 AlenK123

Device plugin resources? I don't think so (vendors have their device plugins implemented for this). Is there a particular device or resource you would want to use?

mythi avatar Jul 23 '22 06:07 mythi

We added experimental "Extended Resources" support to NFD some time ago: https://kubernetes-sigs.github.io/node-feature-discovery/v0.7/get-started/features.html#extended-resources

Just to be clear, it is by no means a replacement for the usage of device plugins. The device plugin API does more than just expose a feature/resource as an extended resource. NFD will not do any allocate/deallocate calls or health checks.

zvonkok avatar Jul 25 '22 08:07 zvonkok

It seems to me that NDF and device plugin capabilities go hand in hand, why not combine them?

AlenK123 avatar Aug 01 '22 17:08 AlenK123

Device plugins are much more than just updating the node status (advertising the extended resource(s)). We don't want to have vendor-specific device management logic in NFD and I don't think the vendors want it either 🙂

As Zvonko mentioned, NFD has this experimental feature of advertising extended resources but there is no way to add any device plugin functionality.

marquiz avatar Aug 08 '22 07:08 marquiz

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

k8s-triage-robot avatar Nov 06 '22 07:11 k8s-triage-robot

The Kubernetes project currently lacks enough active 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 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 rotten

k8s-triage-robot avatar Dec 06 '22 08:12 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-triage-robot avatar Jan 05 '23 09:01 k8s-triage-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.

k8s-ci-robot avatar Jan 05 '23 09:01 k8s-ci-robot