Frequent "error querying ethtool ioctl link state" in talos logs

[jfroy]

Bug Report

Description

In v1.8.0-beta.0, I am now seeing this somewhat regular talos log line:

[talos] error querying ethtool ioctl link state {"component": "controller-runtime", "controller": "network.LinkStatusController", "link": "<LINK>", "error": "no such device"}

This appears to be logged shortly following this kernel log line:

eth0: renamed from tmp<random>

which I believe comes from cilium's normal operations.

The output of talosctl get linkstatus looks OK, so this may be a race when an interface gets renamed that resolves itself.

Environment

• Talos version:

  Client:
          Tag:         v1.7.6
          SHA:         ae67123a
          Built:       
          Go version:  go1.22.5
          OS/Arch:     linux/amd64
  Server:
          NODE:        192.168.1.13
          Tag:         v1.8.0-beta.0
          SHA:         b92c2e49
          Built:       
          Go version:  go1.22.7
          OS/Arch:     linux/amd64
          Enabled:     RBAC
  

• Kubernetes version:

  Client Version: v1.31.0
  Kustomize Version: v5.4.2
  Server Version: v1.31.0
  

• Platform: baremetal
• Kernel: custom 6.10.6 configuration

I've attached a support file.

[smira]

These are harmless, but we should silence them

[Syntax3rror404]

eth0: renamed from tmp

Could be the CNI which is normal.