配置Kubernetes联邦集群放置策略时遇kubectl get clusters无资源问题

Let’s walk through the most likely issues that could be causing this error after you’ve set up your placement policies and configured the admission controller:

1. Verify your kubectl context is targeting the Federation API Server #

It’s easy to accidentally stay connected to your local cluster’s context instead of the federation one. Here’s how to check and fix this:

• List all available contexts:

  kubectl config get-contexts
  

• Switch to your federation context (replace <federation-context> with your actual context name):

  kubectl config use-context <federation-context>
  

Even if you think you’re in the right context, double-check—this is the most common culprit for missing federation resources.

2. Ensure the Federation API Server’s admission controller changes took effect #

You edited the apiserver deployment to add the SchedulingPolicy admission plugin, but we need to confirm those changes are working correctly:

• First, check if the apiserver pod restarted successfully:

  kubectl -n federation-system get pods | grep apiserver
  

  Look for a Running status and a recent restart time (indicating the deployment picked up your changes).
• Next, inspect the apiserver logs for errors related to the admission plugin or config file:

  kubectl -n federation-system logs <apiserver-pod-name>
  

  Search for keywords like SchedulingPolicy, admission-control-config-file, or error—you might find issues like an incorrect config file path, invalid YAML in the config, or a missing plugin dependency.
• Confirm your --admission-control-config-file parameter points to the correct path where your scheduling-policy-admission.yaml is mounted in the pod. If you used a ConfigMap to mount the file, verify the ConfigMap exists and the mount path matches:

  kubectl -n federation-system get configmaps
  

3. Check if federated clusters are properly registered #

If your clusters weren’t successfully registered with the Federation API Server, kubectl get clusters will return nothing. Try these steps:

• Re-register your cluster (follow your federation setup steps for registration) to rule out registration failures.
• Verify your kubectl version is compatible with the Federation API Server version—mismatched versions can lead to unrecognized resources. Run kubectl version to cross-check.

4. Confirm you’re querying the correct API group #

Federation resource groups can vary slightly based on your federation version. Try querying with the full resource name to avoid ambiguity:

kubectl get clusters.federation.k8s.io

Or if you’re using an older federation version:

kubectl get clusters.core.federation.k8s.io

You can also list all supported resources to confirm the clusters resource exists in your current context:

kubectl api-resources | grep cluster

Start with the context check first—it’s the quickest fix. If that doesn’t work, move on to verifying the apiserver logs and cluster registration.

内容的提问来源于stack exchange，提问作者nir