New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Issue with k8s.io/docs/setup/independent/create-cluster-kubeadm/#pod-network #6682

Closed
BastienL opened this Issue Dec 15, 2017 · 6 comments

Comments

Projects
None yet
5 participants
@BastienL
Copy link

BastienL commented Dec 15, 2017

This is a...

  • Feature Request
  • Bug Report

Problem:

The guide for flannel as a pod network recommends to apply https://raw.githubusercontent.com/coreos/flannel/v0.9.1/Documentation/kube-flannel.yml. However, the spec for the kube-flannel container in the DaemonSet do not set resources requests and limits. As a result, the kube-flannel-ds pods will have a QoS class of BestEffort.

This can dangerous on low-cpu or low-ram nodes, as that would allow kubelet to remove that pod if there's too much pressure on the resources of the node. Without a kube-flannel-ds pod running on a node, all the pods on that node become unreachable.

Proposed Solution:

Set requests and limits resources on the kube-flannel container, like on https://github.com/coreos/flannel/blob/master/Documentation/kube-flannel.yml#L127-L133.

With this change, the kube-flannel-ds pod gets a QoS class of Guaranteed, preventing it to be killed in the node runs out of resources.

Page to Update:
https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm/#pod-network

@tengqm

This comment has been minimized.

Copy link
Contributor

tengqm commented Jan 15, 2018

@BastienL Have you filed an issue to flannel project? This doesn't sound like a k8s doc problem.

@BastienL

This comment has been minimized.

Copy link

BastienL commented Jan 15, 2018

@tengqm I think the k8s documentation should link to https://github.com/coreos/flannel/blob/master/Documentation/kube-flannel.yml#L127-L133 instead of https://raw.githubusercontent.com/coreos/flannel/v0.9.1/Documentation/kube-flannel.yml. Not sure it's a flannel issue then. But I'll let them know too.

@tengqm

This comment has been minimized.

Copy link
Contributor

tengqm commented Jan 16, 2018

ya, I agree when possible we should use links that are general and stable

@fejta-bot

This comment has been minimized.

Copy link

fejta-bot commented Apr 26, 2018

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

@fejta-bot

This comment has been minimized.

Copy link

fejta-bot commented May 26, 2018

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten
/remove-lifecycle stale

@fejta-bot

This comment has been minimized.

Copy link

fejta-bot commented Jun 25, 2018

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment