Troubleshoot common issues in AKS Edge Essentials

Artikkel
07/22/2024

This article describes how to find solutions for issues you encounter when using AKS Edge Essentials. Known issues and errors are organized by functional area. You can use the links provided in this article to find solutions and workarounds to resolve them.

Open a support request

See the Get support article for information about how to use the Azure portal to get support or open a support request for AKS Edge Essentials.

Deployment issues

Untrusted publisher issue

Error message: "Do you want to run software from this untrusted publisher? ....."

Workaround: Update your PowerShell execution policy to RemoteSigned:

# Get the execution policy on the system
Get-ExecutionPolicy
# Set the execution policy for this process only
if ((Get-ExecutionPolicy) -ne "RemoteSigned") { Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope Process -Force }

Low disk space causes previously cached container images to be deleted

When the node runs out of disk space, some of the preloaded images are garbage collected by the containerd runtime. In this case, first free up some disk space, and then run the following command to pull the cached images again:

    Invoke-AksEdgeNodeCommand -NodeType Linux -command "sudo /path/to/script/reimport-ci.sh

This script checks for the missing images and reimports them as needed.

Azure Arc connectivity

Issue: Disconnect-AksEdgeArc doesn't remove the pods from the cluster.

Workaround: If the pods aren't cleaned up, run the following commands to manually clean up the existing Azure Arc-related resources before trying to reconnect again:
```
kubectl delete ns azure-arc
kubectl delete clusterrolebinding azure-arc-operator
kubectl delete secret sh.helm.release.v1.azure-arc.v1
```
Issue: Azure Arc connectivity doesn't work in a proxy environment.

Workaround: You can enable system-wide proxy settings by following Internet options > Connections > LAN Settings.

Offline deployments

Failed to get nodeagent certificate: Not Found

Check the network adapter configuration. During deployment, AKS Edge Essentials needs an adapter that's enabled and has the correct IP address, subnet, and default gateway. These values are automatically populated in a DHCP environment. If you're setting manually, ensure all three are set. In many cases, the default gateway isn't set, which results in this error.

Kubernetes

Kube-vip pod continuously restarts

In some scenarios, the kube-vip pod loops and restarts continuously.

Validation

To confirm that the scenario you're encountering is the same issue documented in this article, check that the kube-vip pod in the kube-system namespace has a high number of restarts by running the following command:

kubectl get pods –n kube-system

Cause

There are a few different reasons why the kube-vip pod might be constantly restarting. These causes include:

Using an OS disk that is not backed with an SSD disk, or a premium SSD disk when using an Azure VM. You can review the hardware requirements here.
Disk latency is too high. If the disk latency is greater than 10 ms, it can result in request timeouts, leader loss, and potential cluster instability. You can review the hardware requirements for etcd here.

Resolution

To mitigate this issue, review your underlying storage infrastructure to ensure that it meets the performance requirements for etcd and AKS Edge Essentials. Also, consider using premium SSD-backed storage or optimizing your storage configuration for performance.

Next steps

AKS Edge Essentials overview

Del via