I am also experiencing this problem. I have a CM5 running questing with
6.17.0-1004-raspi, and every so often networking locks up as per the
description. It was fine with plucky and its 6.14 kernel.

Mine is also a Kubernetes node (using RKE2), running Calico, with both
Calico and kube-proxy using the nftables back-end - so I think that
mostly rules out bpf? If I keep the node running with a minimal workload
(and the node cordoned) it's stable and doesn't seem to fail; if I let
it run CronJobs then it reliably fails after a few days.

The workaround setting the CPU scaling governor to "performance" did
_not_ work for me. I tried this yesterday and networking still failed
overnight, though I changed the setting after I had one of the RCU stall
messages. I had no additional kernel log output until networking died
and libceph noticed the mons disappearing.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2133877

Title:
  Complete network hang on Raspberry Pi 5 with kernel 6.17 under load -
  possibly related to CPU frequency scaling

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-raspi/+bug/2133877/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to