Skip to content
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

Need a way to show masters #33533

Closed
justinsb opened this issue Sep 27, 2016 · 2 comments
Closed

Need a way to show masters #33533

justinsb opened this issue Sep 27, 2016 · 2 comments

Comments

@justinsb
Copy link
Member

justinsb commented Sep 27, 2016

If we are going to replace unschedulable with a taint for the master, we need a way to make the role (cf #33530) visible. Otherwise there is no way to know which node is the master:

> kubectl get nodes
NAME                             STATUS    AGE
ip-172-20-110-215.ec2.internal   Ready     3m
ip-172-20-110-216.ec2.internal   Ready     3m
ip-172-20-117-246.ec2.internal   Ready     12m

I propose we add some form of "smart" presentation of the labels, or maybe just the kubernetes.io/role label, as the --show-labels output is overwhelming:

kubectl get nodes --show-labels
NAME                             STATUS    AGE       LABELS
ip-172-20-110-215.ec2.internal   Ready     5m        beta.kubernetes.io/arch=amd64,beta.kubernetes.io/instance-type=t2.medium,beta.kubernetes.io/os=linux,failure-domain.beta.kubernetes.io/region=us-east-1,failure-domain.beta.kubernetes.io/zone=us-east-1c,kubernetes.io/hostname=ip-172-20-110-215.ec2.internal
ip-172-20-110-216.ec2.internal   Ready     5m        beta.kubernetes.io/arch=amd64,beta.kubernetes.io/instance-type=t2.medium,beta.kubernetes.io/os=linux,failure-domain.beta.kubernetes.io/region=us-east-1,failure-domain.beta.kubernetes.io/zone=us-east-1c,kubernetes.io/hostname=ip-172-20-110-216.ec2.internal
ip-172-20-117-246.ec2.internal   Ready     14m       beta.kubernetes.io/arch=amd64,beta.kubernetes.io/instance-type=m3.large,beta.kubernetes.io/os=linux,failure-domain.beta.kubernetes.io/region=us-east-1,failure-domain.beta.kubernetes.io/zone=us-east-1c,kubernetes.io/hostname=ip-172-20-117-246,kubernetes.io/role=master

cc @davidopp @mikedanese

@errordeveloper
Copy link
Member

Yes, this would be great. Role label in kubectl get nodes -o wide would be very convenient, I think.

justinsb added a commit to justinsb/kubernetes that referenced this issue Oct 31, 2016
We are moving towards marking master nodes as tainted, and not
necessarily unschedulable.  Further now we encourage users to taint
nodes, marking them unschedulable.

Thus the reliance on "Unschedulable" is not really a great indicator for
the master.

Instead, recognize the existing node 'role' markers, and surface them
where Unschedulable is (in the status).

We recognize:

 * a kubernetes.io/role label
 * a kubeadm.alpha.kubernetes.io/role label
 * a taint with Key 'dedicated'

Fix kubernetes#33533
@justinsb
Copy link
Member Author

PR in #35901

justinsb added a commit to justinsb/kubernetes that referenced this issue Nov 4, 2016
We are moving towards marking master nodes as tainted, and not
necessarily unschedulable.  Further now we encourage users to cordon
nodes, marking them unschedulable.

Thus the reliance on "Unschedulable" is not really a great indicator for
the master.

So, recognize the existing node 'role' markers, and surface them
where Unschedulable is (in the status).

We recognize:

 * a kubernetes.io/role label
 * a kubeadm.alpha.kubernetes.io/role label

Fix kubernetes#33533
justinsb added a commit to justinsb/kubernetes that referenced this issue Nov 7, 2016
We are moving towards marking master nodes as tainted, and not
necessarily unschedulable.  Further now we encourage users to cordon
nodes, marking them unschedulable.

Thus the reliance on "Unschedulable" is not really a great indicator for
the master.

So, recognize the existing node 'role' markers, and surface them
where Unschedulable is (in the status).

We recognize:

 * a kubernetes.io/role label
 * a kubeadm.alpha.kubernetes.io/role label

Fix kubernetes#33533
k8s-github-robot pushed a commit that referenced this issue Nov 10, 2016
Automatic merge from submit-queue

kubectl: show node label if defined

We are moving towards marking master nodes as tainted, and not
necessarily unschedulable.  Further now we encourage users to taint
nodes, marking them unschedulable.

Thus the reliance on "Unschedulable" is not really a great indicator for
the master.

Instead, recognize the existing node 'role' markers, and surface them
where Unschedulable is (in the status).

We recognize:

 * a kubernetes.io/role label
 * a kubeadm.alpha.kubernetes.io/role label
~a taint with Key 'dedicated'~

Fix #33533
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants