AKS legacy API tests sometimes hit HTTP 409 on azure-cni DaemonSet

[nojnhuh]

Which jobs are flaky:

https://storage.googleapis.com/k8s-triage/index.html?pr=1&text=daemonsets.apps%20%22azure-cni%22%3A%20the%20object%20has%20been%20modified%3B&job=cluster-api-provider-azure

Which tests are flaky:

Testgrid link:

https://prow.k8s.io/view/gs/kubernetes-ci-logs/logs/periodic-cluster-api-provider-azure-e2e-aks-main/1845928655998423040

Reason for failure (if possible):

It's not totally clear to me if installing CNI explicitly on an AKS cluster is necessary. I think we started doing that for BYO node pools in #4052, but if our templates already specify Azure CNI, does it really make a difference? Can we simply remove these from the tests?

cluster-api-provider-azure/test/e2e/azure_test.go

Line 690 in 68b882c

withAzureCNIv1Manifest(e2eConfig.GetVariable(AzureCNIv1Manifest)),

cluster-api-provider-azure/test/e2e/azure_test.go

Line 769 in 68b882c

withAzureCNIv1Manifest(e2eConfig.GetVariable(AzureCNIv1Manifest)),

cluster-api-provider-azure/test/e2e/azure_test.go

Line 898 in 68b882c

withAzureCNIv1Manifest(e2eConfig.GetVariable(AzureCNIv1Manifest)),

If not, we should probably constrain this DaemonSet to only the BYO nodes.

We shouldn't dig too deep into this until we have the BYO node tests running again: #5153

Anything else we need to know:

• links to go.k8s.io/triage appreciated
• links to specific failures in spyglass appreciated

/kind flake

[One or more /area label. See https://github.com/kubernetes-sigs/cluster-api/labels?q=area for the list of labels]