Follow-up, 16 days on:
- On 2026-10-02 I moved the affected guest (daringly) from
7.0.0-1017-azure to noble's 6.8.0-1070-azure. Everything else stayed the
same: the install, the workload and the host. It has had no oopses
since. Three more Ubuntu 26.04 guests on the same host have also run
6.8.0-1070-azure since 2026-10-11/12, with no crashes. The last guest
crash in the Hyper-V log was on a 7.0 kernel, on 2026-10-04.
- Host confounder: the host has logged no WHEA errors since 2026-10-03,
and its last crash was on 2026-10-06. All the 7.0 oopses in this report
fall inside the window when the host was unstable, so I can't fully rule
the host out for those.
- I haven't run 7.0 since 2026-10-07. The second VM ("kernel-lab", where
apport-collect ran) is still available, with a reproducer that crashed
7.0 within minutes, most reliably during depmod. If it helps, I can boot
a specific test kernel, a mainline build or a patch there and report
whether it crashes. I can also try to bisect between 6.8 and 7.0.
Please let me know which you'd find most useful.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2169338
Title:
Kernel oops / GPF in path_is_under() from complete_walk() on openat
— 7.0.0-34/-38-generic and 7.0.0-1017-azure, Hyper-V Gen2 guest
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2169338/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
