Time Query Error

[DizzieNight]

Bug Report

Description

Out of nowhere my nodes start getting the following error

It's generally only 1 or 2 workers at a time but I have no idea why it is happening. The workers become completely unresponsive.

Logs

Environment

• Talos version: 1.8.3
• Kubernetes version: 1.31.2
• Platform: Baremetal/VM

[smira]

The error sounds like a firewall issue. As Talos doesn't block outbound connections, check your CNI, specifically if you're using something more advanced like Cilium.

[DizzieNight]

I am using cilium, sorry should have mentioned that. I'm not too familiar with it. If you are then would you have any ideas where I can look to find the problem?

[smira]

I'm not a Cilium expert, but operation not permitted is usually returned by the firewall. So my guess is that Cilium blocks these connections. No idea why, could be mostly anything, including overlapping pod subnet with your host network, or some network policies, or simply a bug.