Additional log findings from journalctl After digging into the journalctl logs from my failed shutdown sessions, I found the following suspicious kernel messages right before the power-off sequence stalled: Plaintext
kernel: workqueue: inode_switch_wbs_work_fn hogged CPU for >10000us 4 times, consider switching to WQ_UNBOUND kernel: workqueue: inode_switch_wbs_work_fn hogged CPU for >10000us 5 times, consider switching to WQ_UNBOUND It seems that the inode_switch_wbs_work_fn kernel thread (which manages writeback I/O operations prior to unmounting filesystems) hogs the CPU and causes a deadlock right as systemd reaches poweroff.target. Dropping these log entries here in case they help narrow down the root cause of this regression. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2161289 Title: Kernel 7.0.0-28-generic: system hangs on shutdown/reboot on ThinkPad T14 Gen 2 AMD To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2161289/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
