I can reproduce this on Ubuntu 24.04.4 LTS with the HWE kernel. Adding
my data point.

**System:**
- Ubuntu 24.04.4 LTS (Noble Numbat), fresh install
- Kernel: 6.17.0-14-generic (linux-generic-hwe-24.04, from noble-updates/main)
- CPU: Intel Core i7-6700 (Skylake, 4C/8T)
- Chipset: Intel Z170 (100 Series/C230 Series)
- GPU: NVIDIA GeForce GTX 1080 (GP104), driver 580.126.09 proprietary
- RAM: 64GB DDR4
- Ethernet: Intel I219-V (wired, default route)
- USB WiFi: mt76x2u (MT7612U) — was unplugged at the time of one of the freezes
- Desktop: GNOME (X11 session via gdm-x-session)

**Symptoms — identical to this bug:**
Complete hard system freeze. The journal log stops abruptly with no error, no 
kernel panic, no OOM message. SSH from another machine is unresponsive. Magic 
SysRq keys do not work. The only recovery is a hard power-button hold.

Last journal lines before freeze (completely normal, then nothing):

  <timestamp> systemd[1]: Finished sysstat-collect.service
  <+11s>     ghostty[5252]: warning: failed to activate the on-screen keyboard

No /var/crash/ report is generated. journalctl -b -1 -p 0..3 shows only
unrelated ntfs3 MFT warnings.

**Timeline:**
- Ubuntu 24.04.4 was freshly installed with linux-generic-hwe-24.04, which 
pulled in kernel 6.17.0-14. The system has only ever booted this kernel on this 
install.
- Hard freezes began within a day of installation and have recurred roughly 
every 1-2 days since.
- The same hardware ran stable for years under previous OS installs with older 
kernels.

**Additional observations:**
1. This is NOT a GNOME/Wayland issue in my case — I am running X11, not Wayland.
2. This is NOT a GNOME extensions issue — I have standard Ubuntu extensions 
only.
3. Thermals are not a factor: CPU ~50°C, GPU ~45°C at idle.
4. A separate freeze (on a previous boot) was preceded by mt76x2u USB WiFi 
driver errors (page_pool_release_retry() stalled pool shutdown), but the most 
recent freeze had the mt76x2u adapter physically removed ~2.5 hours before the 
freeze occurred.
5. No kernel boot parameters (pcie_aspm, intel_idle.max_cstate, etc.) were set.

**What I plan to test:**
- Booting kernel 6.14.0-37-generic or 6.8 (available in repos) to confirm this 
is a 6.17 regression
- Adding pcie_aspm=off as a kernel parameter to rule out PCIe power state 
transitions

This bug appears to affect 24.04 LTS users as well (not just 25.10),
since the HWE kernel metapackage now pulls in 6.17. Given that the HWE
kernel is offered via standard apt updates, this could affect a large
number of 24.04 users.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2127687

Title:
   Unreliable Hard System Freeze/Kernel Lockup on Browser Notification
  (Ubuntu 25.10 / Kernel 6.17)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2127687/+subscriptions


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

Reply via email to