cc/ @Random-Liu We ran into the similar issue with node problem detector.

XRef kubernetes/node-problem-detector#13 for similar issue we encountered in node problem detector.

For the record, I have been relying on this behaviour for months. I have kubelet running on my master nodes with --register-schedulable=false. That way, normal pods won't schedule, but the fluentd DaemonSet does.

Yes, I need the same thing. I'm not proposing changing that behavior, just adding some sort of label for --register-schedulable=false machines that can be easily used in a DaemonSet selector. (Or documentation of it if such a label already exists.)

I believe it's tracked here #22205

I believe that #22205 is about having the scheduler respect affinity contraints for daemonset pods.

In this case, the scheduler already knows how to use node selectors... I'm just seeking to have kubelet automatically label nodes that are marked --register-schedulable=false so that I can use that label in the daemonset's node selector.

(Unless there's already a label/mechanism to do this?)

/sig apps

/sig scheduling

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.

Prevent issues from auto-closing with an /lifecycle frozen comment.

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

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

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