-
Notifications
You must be signed in to change notification settings - Fork 1.9k
bug 1738456 setting the control plane as unschedulable #17016
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
Conversation
|
The preview will be available shortly at: |
wking
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me. Needs to go into GCP UPI docs too, but maybe we don't have those yet.
|
@adellape, please pick up this change in your draft of the GCP UPI docs. |
|
@openshift/team-documentation PTAL |
e36cb06 to
4bf86c1
Compare
c9af188 to
d275527
Compare
|
Just merged #17043 which is likely the conflict culprit here. |
06a5b9b to
abb54d2
Compare
abb54d2 to
b045591
Compare
| Because you create your own compute machines later in the installation process, | ||
| you can safely ignore this warning. | ||
|
|
||
| ifdef::aws,gcp[] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@jianlinliu, we didn't have a step to remove these manifest files for vSphere or bare metal.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
After I took another look at the files that I'd need to change to apply this step to vSphere and bare metal, I realized that it would be easier to do it on this PR. For these methods, the 4.1 steps (eg vSphere) skip generating the manifests and just have you generate the Ignition configs. With this change, you need to generate the manifests too (proposed 4.2 change).
The way the conditionals are set up, the 4.2 docs still won't make you remove the control plane or compute manifests.
| link:https://github.com/kubernetes/kubernetes/issues/65618[Kubernetes limitation], | ||
| router Pods running on control plane machines will not be reachable by the | ||
| ingress load balancer. | ||
| . Modify the `manifests/cluster-scheduler-02-config.yml` Kubernetes manifest file to prevent Pods from being scheduled on the control plane machines: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@jianlinliu, in the original GCP PR, this was marked as an optional step. Will you please double-confirm that we always need to change this setting on GCP?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I checked upstream doc:
https://github.com/openshift/installer/blob/master/docs/user/aws/install_upi.md#make-control-plane-nodes-unschedulable
https://github.com/openshift/installer/blob/master/docs/user/gcp/install_upi.md#make-control-plane-nodes-unschedulable
Did not see where indicate it as an optional step. Could you point me the original GCP PR?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Per my test result, I do not think it is a optional step when we are guiding user set worker replica to 0. Now I think everywhere keep consistent now, it is a MUST (from comments in bugzilla).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@adellape, FYI ^
|
Jianlin approved this change on the bug, so I'm going to merge. Thanks! |
|
/cherrypick enterprise-4.2 |
|
/cherrypick enterprise-4.3 |
|
@kalexand-rh: new pull request created: #17364 DetailsIn response to this:
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. |
|
@kalexand-rh: new pull request created: #17365 DetailsIn response to this:
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. |
https://bugzilla.redhat.com/show_bug.cgi?id=1738456
@wking, are you ok with the way I translated this change?