-
Notifications
You must be signed in to change notification settings - Fork 2.7k
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
Ingress: Add support for is-default-class on IngressClass #23719
Conversation
f460e7e
to
0663468
Compare
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.
CI workflow LGTM, uses pull_request
trigger so no specific risk in case of attack via malicious PR (forks do not have access to upstream secrets).
Helm changes are trivial, LGTM.
- name: Create minikube cluster with multiple nodes | ||
uses: medyagh/setup-minikube@ab221dee176f8eabd8deddf849b5bf1d6244a6e8 # v0.0.11 | ||
with: | ||
minikube-version: ${{ env.minikube_version }} | ||
network-plugin: cni | ||
cni: false | ||
wait: false | ||
start-args: --nodes 3 --docker-opt "default-ulimit=nofile=102400:102400" |
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.
Is there any specific reason to use Minikube here instead of Kind? Not a problem at all but I'm wondering if the pipeline would not run faster on a Kind-based cluster.
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.
Happy to move it to kind (I was wondering the same thing as you...), but it builds upon the previous conformance tests which all use minikube, I only modified this one to use the default behavior
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.
Your call, if not worth the trouble it's OK as is, and consistency with the other conformance workflows is also good. Probably @sayboras just needed Minikube for the other workflows.
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'll keep it minikube for now and will probably file an issue to consider kind (unless @sayboras has a reason)
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.
Last time I can't get kind + metallb (required for LB service) to work from host (i.e. github runner), so I just use minikube with handy minikube tunnel command.
Recently, I noticed that we have labs prepared with Kind + MetalLB, so it's worth to give it a crack again.
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.
as discussed offline, we can tackle this separately, we might even use LB IPAM feature if possible.
@meyskens can you please fill me in on motivation for this? I understand the change, but just looking for a bit more context. |
This PR came out of #23181 (which was passed onto be by my team as good first issue). |
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 found a couple of typos, but looks good otherwise from my side. Thanks!
0663468
to
32d730c
Compare
32d730c
to
6a76d5d
Compare
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.
Thanks!
56ce9a1
to
651c6c1
Compare
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.
One minor thing, then LGTM!
651c6c1
to
f37de60
Compare
This adds support for the ingressclass.kubernetes.io/is-default-class flag on the IngressClass. If set Cilium ingress wil pick up all ingress objects where no class is set also. This also adds exposure to enable it in Helm under ingressController.default Fixes: cilium#23181 Signed-off-by: Maartje Eyskens <maartje.eyskens@isovalent.com>
f37de60
to
9302b3f
Compare
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.
Nice!
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.
Thanks a lot for taking up this task, LGTM ✔️
/test |
Suite-k8s-1.16.K8sUpdates Tests upgrade and downgrade from a Cilium stable image to master failure due to the below error. I don't think it's related to the changes here (only on operator and Ingress). Re-run to confirm the flake.
|
/test-1.16-4.19 |
Reviews are in, all tests passed (ci-verifier test is not required for this change). Mark this ready to merge. |
@sayboras Do we already have a flake issue for that? |
Good point, confirmed that its a flake as re-run was good. Here is it #23845 |
Thanks Tam! |
This adds support for the ingressclass.kubernetes.io/is-default-class flag on the IngressClass. If set Cilium ingress wil pick up all ingress objects where no class is set also.
This also adds exposure to enable it in Helm under ingressController.default
Fixes: #23181
Signed-off-by: Maartje Eyskens maartje.eyskens@isovalent.com
Please ensure your pull request adheres to the following guidelines:
description and a
Fixes: #XXX
line if the commit addresses a particularGitHub issue.
Fixes: <commit-id>
tag, thenplease add the commit author[s] as reviewer[s] to this issue.
Fixes: #issue-number