Windows Host networking corrupted during cluster join phase

Hi all,
We have a Rancher 2.6 cluster deployed on OCI (Oracle Cloud Infrastructure) and we’re trying to provision an RKE cluster with Windows support, according to Rancher Docs: Launching Kubernetes on Windows Clusters (VXLAN option)
We have successfully deployed the linux stuff (control plane + worker node) and the cluster is up and running.
The problem arise when we try to join the first Windows node (Windows server 2019 1809); Rancher agent starts and begin to download the required modules; everyting seems to works fine but after a while the Windows node stops responding and, when we connect to the node via console, we find the network interface not configured.

The Windows node is running Docker EE, installed as described in Configure Docker in Windows | Microsoft Docs

below the status of docker containers on Windows node:

If we try to do a manual restart of the Kubelet we get the following error:

Error response from daemon: container 4ea080ae52689fdb30a9d2ae8dfd844c5605fbaf7781702516dd73c212e0b716 encountered an error during hcsshim::System::Start: failure in a Windows system call: The virtual machine or container exited unexpectedly. (0xc0370106)
Error: failed to start containers: kubelet

Anyone had a similar problem?

Thank you.
Stefano