Windows Pods IPs Unreachable

[vincent-guillemette-fr]

Hello Calico,

Following tutorials found on MicroK8s, I set up a cluster including :

• 1 Linux Control Plane,
• 1 Windows Worker.

Linux workloads work like a charm, but when I deploy pods on the Windows node, its IP cannot be reached at all.

Expected Behavior

To be able to ping Windows Pods from:

• Linux Host,
• Windows Host,
• Linux Pod.

Current Behavior

Can only ping Pod from itself.

Would you be so kind to help me troubleshoot this behavior ?

[coutinhop]

@vincent-guillemette-fr could you please post details about your setup? Calico version, k8s version, Linux distro and version, Windows version, etc...

Taking a (rather wild, from the limited info) guess at what might be wrong, this could be this Windows containers bug microsoft/Windows-Containers#516

Please take a look at this known issue #9019 (Windows build 20348.2582 and up seems to break networking), there is a fix on this comment: #9019 (comment)

[vincent-guillemette-fr]

@coutinhop
Thank you for helping us.
I'm afraid I'm a bit confused about all the components of the cluster and the information relevant to each one of them.

It's resolved, but we applied two fixes at a time, so I don't know which one, if not both, solved the connectivity issue:

• The bug you referred to : Container networking on Kubernetes broken after Server 2022 July 2024 / KB5040437 (OS Build 20348.2582) update microsoft/Windows-Containers#516,
• The Network Adapter switch mentioned here: https://ranchermanager.docs.rancher.com/how-to-guides/new-user-guides/kubernetes-clusters-in-rancher-setup/use-windows-clusters#rancher-on-vmware-vsphere-with-esxi-67u2-and-above.

[vincent-guillemette-fr]

We rolled back the network adapter switch and can then confirm we were affected by the bug you mentioned.
Thank you again !