-
Notifications
You must be signed in to change notification settings - Fork 196
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
Bug: ASO + CAPZ integration broken #4274
Comments
cc @nojnhuh, are you aware of this? |
Can you share the full ASO deployment and webhook service YAML as rendered in CAPZ? |
I haven't seen this before locally or in our tests. @rbjorklin How are you installing CAPZ and ASO?
We are still bundling ASO v2.8.0 with CAPZ: https://github.com/kubernetes-sigs/cluster-api-provider-azure/blob/v1.16.1/config/aso/kustomization.yaml#L5. Where are you seeing evidence that it's actually ASO v2.9? |
In terms of changes in this space, I added 2 new labels here in 2.9.0: #3184, but tests all pass and the existing |
You're absolutely right and I can't read. It is indeed version See the manifests below:
|
CAPZ generally assumes it and ASO are being installed in the capz-system namespace. How are you installing it in the management-aks-aso namespace? Is the KubeadmControlPlane controller also running in that namespace? |
This is what I do: ❯ kind create cluster \
--name management-cluster \
--kubeconfig ~/.kube/configs/kind-management-cluster.yaml
❯ clusterctl init \
--kubeconfig ~/.kube/configs/kind-management-cluster.yaml \
--core cluster-api:v1.7.5 \
--bootstrap kubeadm:v1.7.5 \
--control-plane kubeadm:v1.7.5 \
--infrastructure azure:v1.16.1 \
--wait-providers \
--target-namespace management-aks-aso I just tried removing |
Yes: ❯ k -n management-aks-aso get pods
NAME READY STATUS RESTARTS AGE
azureserviceoperator-controller-manager-db79d7d5d-46xkc 1/1 Running 0 93s
capi-controller-manager-7cdf496cd5-rmv8t 1/1 Running 0 95s
capi-kubeadm-bootstrap-controller-manager-5b99b66467-hpsz5 1/1 Running 0 95s
capi-kubeadm-control-plane-controller-manager-544d667cc9-rnsht 1/1 Running 0 95s
capz-controller-manager-7b77f56677-j66km 1/1 Running 0 93s |
That's what I would recommend since that's what we do in all of our tests. It is theoretically possible to have multiple CAPI setups in the same workload cluster, but I'm not sure that use case is supported by CAPI. |
Thanks for your time and assistance! Turns out this is not a bug then and I just need to RTFM: |
I've filed #4277 to make the ASO label selector more robust anyway, as I've never liked that we have such a generic one. That doesn't mean CAPZ will work how you tried to install it, but it should mean ASO is a bit more resilient to these sorts of topologies if they do come up. |
Version of Azure Service Operator:
2.9.02.8.0 that comes with version 1.16.1 of cluster-api-provider-azure.Describe the bug
The
capz-controller-manager
pod is throwing errors when attempting to connect toazureserviceoperator-webhook-service
.To Reproduce
Steps to reproduce the behavior:
Deploy an AKS-ASO cluster with cluster-api.
Expected behavior
No TLS errors are thrown.
Screenshots
Additional context
Detailed information, click to expand.
E0916 22:22:15.403491 1 controller.go:329] "Reconciler error" err="failed to reconcile resources: failed to apply resource: conversion webhook for containerservice.azure.com/v1api20231001storage, Kind=ManagedCluster failed: Post \"https://azureserviceoperator-webhook-service.management-aks-aso.svc:443/convert?timeout=30s\": tls: failed to verify certificate: x509: certificate is valid for capi-kubeadm-control-plane-webhook-service.management-aks-aso.svc, capi-kubeadm-control-plane-webhook-service.management-aks-aso.svc.cluster.local, not azureserviceoperator-webhook-service.management-aks-aso.svc" controller="azureasomanagedcontrolplane" controllerGroup="infrastructure.cluster.x-k8s.io" controllerKind="AzureASOManagedControlPlane" AzureASOManagedControlPlane="management-aks-aso/management-aks-aso" namespace="management-aks-aso" name="management-aks-aso" reconcileID="3cff6256-7a50-4f65-8d6c-eb55d942c048"
Suggested fix:
Editing the service in place and changing this to say:
app.kubernetes.io/name: azure-service-operator
fixes the problem.The text was updated successfully, but these errors were encountered: