## Update (2026-09-20): failure rate has jumped — 5 hangs in 8 days, 0
in the previous 4 weeks; two more silent freezes on plain s2idle suspend
entry

Two more hangs on 2026-09-19, both on a plain suspend (no hibernate
involved in the failing step), same signature as before: journal ends
right after "PM: suspend entry (s2idle)", keyboard backlight stays lit,
no reaction to any key/power short-press, hard power-off required. Next
boot: dirty EFI partition, systemd-journald "corrupted or uncleanly shut
down", orphan inode cleanup.

- 10:12:39 — second suspend request issued 6 s after a very short first suspend 
(woken immediately by AC plug-in). It was a suspend-then-hibernate request 
(default 2h delay); the machine froze on s2idle entry itself, not at the RTC 
wake.
- 20:42:03 — suspend requested 16 s after a successful 1h42 suspend/resume. 
Machine stayed frozen ~9.5 h until a forced reboot.

## Statistics (journal analysis, 450 suspend/hibernate entry/exit events
since 2026-08-15)

- Suspend/hibernate entries issued < 60 s after a resume: 104, of which 2 
failed (both on 2026-09-19).
- All other entries: 123, of which 3 failed (2026-09-12 hibernate, 2026-09-14 
suspend, 2026-09-16 hibernate).
- So a quick re-suspend after a resume is NOT the trigger (~2% failure rate in 
both groups).
- However, there were **zero failures from 2026-08-15 to 2026-09-11, then 5 
failures between 2026-09-12 and 2026-09-19** (roughly one per day). They happen 
on 7.0.0-27 and 7.0.0-31, before and after the BIOS update to 1.29.0. No 
firmware, graphics stack or systemd package was updated before the first 
failure of the series (packages updated 08-09/09: libc, perl, gnupg, curl...; 
13/09: kernel 7.0.0-31, wireless-regdb).

I cannot find what changed around 2026-09-12. The failures leave no
usable trace (kernel is frozen before it can log; last minutes of
journal are lost on power-off), so I cannot narrow the cause further
from logs alone. Happy to test patches/parameters or run a debug kernel
if useful (netconsole/serial capture is possible if requested).

## Hardware/software (unchanged)

Dell Inspiron 14 5425, AMD Ryzen 7 5825U, AMD Barcelo iGPU (amdgpu),
BIOS 1.29.0, kernel 7.0.0-31-generic, Ubuntu, KDE Plasma 6.5, Wayland.
`HibernateMode=shutdown` workaround applied since 2026-09-16.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2161312

Title:
  System hangs on suspend (s2idle) with kernel 7.0.0-28-generic, works
  fine on 7.0.0-27

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2161312/+subscriptions


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

Reply via email to