Closed johngmyers closed 2 years ago
IMO, anything related to LBC should not block anything. If you want, we can revert the commit in 1.23.
We can't ship a broken addon. But we can revert the upgrade. We are still waiting for a few other addons so there is time if upstream moves fast.
We don't want to ship a broken addon, but this is already broken. It doesn't work at all for k8s 1.22+ and we still ship it.
We can also make ec2:DeleteSecurityGroup
to unconditional action.
Not working at all for k8s 1.22+ is not "broken". It still worked for k8s 1.21 and earlier. With this issue it stopped working for k8s 1.21 and earlier.
I don't think we should open up the RBAC. We could change the condition to use their tag, but the controller would still be failing to tag the SG with the admin-provided tags.
I still see it as "broken". I don't mind reverting the update. I don't mind even removing it from the kOps addons.
@olemarkus mentioned that the new version of LBC has been merged so can we close this?
This can probably be closed, but I can't test it right now.
/kind bug
1. What
kops
version are you running? The commandkops version
, will display this information.master branch. Broken code appears to be in release-1.23 branch as well
2. What Kubernetes version are you running?
kubectl version
will print the version if a cluster is running or provide the Kubernetes version specified as akops
flag.1.21.7
3. What cloud provider are you using?
AWS
4. What commands did you run? What is the simplest way to reproduce this issue?
Examine the kops-aws-aws-load-balancer-controller e2e test.
Alternatively:
5. What happened after the commands executed?
The ALB controller is unable to delete the "k8s-traffic-" ingress for the ALB due to lack of IAM permissions. This is because the security group is missing the
KubernetesCluster
tag.6. What did you expect to happen?
Security group is tagged with the tags that kOps passed AWS LBC through the
default-tags
flag. LBC is able to delete the security groups it creates.7. Please provide your cluster manifest. Execute
kops get --name my.example.com -o yaml
to display your cluster manifest. You may want to remove your cluster name and other sensitive information.8. Please run the commands with most verbose logging by adding the
-v 10
flag. Paste the logs into this report, or in a gist and provide the gist link here.9. Anything else do we need to know?
Introduced by #12920.
Filed upstream as kubernetes-sigs/aws-load-balancer-controller#2424