Solution to the problem when deploying k3s nodes in VMware Cloud Director.

[AitorMSuarez]

Hello,
I just want to report a problem we've solved when deploying k3s in a VMWARE CLOUD DIRECTOR environment.

We had an issue where when adding nodes, they were successfully added to the master node, but the pods deployed on the nodes were not accessible through the ingress, and when deploying something like Nginx, it would give us a gateway timeout.

In the end, after a month, the solution was to change the network interface when creating the VMs. Instead of choosing the default VMXNET3 network interface, we selected E1000E, and it worked on the first try without any further actions needed.

Best regards, and I hope this helps others.

[brandond]

This sounds like a known issue with hardware checksum offload on VMware. It affects all CNIs that use vxlan encapsulation, not just flannel. See projectcalico/calico#3145 (comment)

The bug is in the vmxnet driver, iirc.

[github-actions]

This repository uses a bot to automatically label issues which have not had any activity (commit/comment/label) for 45 days. This helps us manage the community issues better. If the issue is still relevant, please add a comment to the issue so the bot can remove the label and we know it is still valid. If it is no longer relevant (or possibly fixed in the latest release), the bot will automatically close the issue in 14 days. Thank you for your contributions.