-
Notifications
You must be signed in to change notification settings - Fork 347
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
param --enable-worker not working #4272
Comments
You mean If you want to run a worker-only node (that needs to join an existing cluster using a join token), have a look at the |
I also encountered the same problem. I asked to install a host, which is both a management node and a worker node。 k0s install controller --single --enable-worker --enable worker When I run the application, the pod is always pending. |
@hztsm This seems like a separate problem. Would you mind to file another issue and provide logs? |
@hztsm Please provide the output of Without the logs, it is only possible to guess at several possible common reasons. Pod cannot be scheduled on tainted node(this should not be your problem, but it was mine, so I'll just post the logs and resolution for this case) The describe logs would look something like this (leaving out irrelevant parts)
This is because your node is tainted.
Run this to remove the taint (notice the
Or start k0s next time with |
AND, when I do
so what works for me is |
That is somewhat surprising. There shouldn't be any differences concerning CNI between |
Yes, here's few GHA runs for the various scenarios
|
That's in the logs in the original issue report. Shouldn't this be correctly set to true? |
In the original issue, k0s wasn't started with the |
I'm sooo oblivious 😬 That's why v1.30.0 is not starting with --single: So, to use --single, you might want to wait until v1.30.1, which will ship in the next week, I think, or provide a custom config to v1.30.0 which changes the kube-router metrics port, or use v1.29.4. |
Just got here, using Based on the discussion above I think that this is the intended behavior, since Noticed that there is also the flag Excerpt from the help of
Think it could help to tweak the help text of |
Realized that no Pods were scheduled due to taints on the Node object. Just using "--enable-worker" is not enough. See: k0sproject/k0s#4272
Before creating an issue, make sure you've checked the following:
Platform
Version
k0sproject/k0s:v1.29.3-k0s.0
Sysinfo
`k0s sysinfo`
What happened?
No response
Steps to reproduce
Version: v1.29.3+k0s.0
Process ID: 8
Role: controller
Workloads: true
SingleNode: false
Kube-api probing successful: true
Kube-api probing last error:
NAME STATUS ROLES AGE VERSION LABELS
k0s Ready control-plane 4m42s v1.29.3+k0s beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k0s,kubernetes.io/os=linux,node-role.kubernetes.io/control-plane=true,node.k0sproject.io/role=control-plane
Expected behavior
Version: v1.29.3+k0s.0
Process ID: 8
Role: controller +worker
Workloads: true
SingleNode: false
Kube-api probing successful: true
Kube-api probing last error:
Actual behavior
No response
Screenshots and logs
No response
Additional context
No response
The text was updated successfully, but these errors were encountered: