Failing to suspend - udev-worker refusing to freeze

qherring · 2023-10-26 20:32:17

seth wrote:

Try to install and boot the lts kernel.

after the last boot issue I had a couple days ago I tried to install and boot LTS just to have another fallback. When I booted into it I was having an issue where nothing would happen after I entered my password to decrypt the disk.

Did not feel like more troubleshooting in that moment so I removed LTS. I will try and reinstall it again now wish me luck.

Last edited by qherring (2023-10-26 20:33:44)

qherring · 2023-10-26 20:55:25

I just did:

pacman -Syy
pacman -S linux-lts
mkinitcpio -p linux
grub-mkconfig -o /boot/grub/grub.cfg

Just booted into linux without issue.
I have not given linux-lts a try yet - need to take care of a few things today before more troubleshooting.

I feel like there is some issue with my grub menu not getting updated properly on updating the kernel. Before, when I was reinstalling linux via chroot it was telling me it generated a fallback image successfully:

==> Creating zstd-compressed initcpio image: '/boot/initramfs-linux-fallback.img'
==> Image generation successful

However there was no fallback option listed in grub upon reboot.

seth · 2023-10-26 21:27:15

pacman -Syy
pacman -S linux-lts

Is bad for two reasons:
First and foremost you're promoting partial updates beyond the kernel, second the double "y" can corrupt your package database (and puts undue stress on the mirror)

So the solution was to re-install the kernel and/or regenerate the initramfs - does the file in /boot/initramfs-linux-fallback.img exist to begin with?

Frequent corruptions of the boot partition/directory are worrysome - do you maybe run into https://bbs.archlinux.org/viewtopic.php?id=289336 ?
(Pay attention to the update messages and notably errors at this stage!)

qherring · 2023-10-26 23:14:15

seth wrote:

pacman -Syy
pacman -S linux-lts
Is bad for two reasons:
First and foremost you're promoting partial updates beyond the kernel, second the double "y" can corrupt your package database (and puts undue stress on the mirror)

got it. so just -Sy should do or always -Syu in this scenario?

seth wrote:

pacman -Syy
pacman -S linux-lts
does the file in /boot/initramfs-linux-fallback.img exist to begin with?

it exists now, not sure about before

seth wrote:

do you maybe run into https://bbs.archlinux.org/viewtopic.php?id=289336 ?
(Pay attention to the update messages and notably errors at this stage!)

no I dont seem to have such errors on mkinitcpio

Last edited by qherring (2023-10-26 23:34:20)

seth · 2023-10-27 06:03:45

got it. so just -Sy should do or always -Syu in this scenario?

"pacman -S foo" or "pacman -Syu foo"

no I dont seem to have such errors on mkinitcpio

I meant in general - right now everything works again, resp. back to "only"

Im getting pretty frustrated here. The inability to suspend and restart has not been resolved still. Even with all three parameters set (pcie_aspm=off, iommu=soft, and nvme_core.default_ps_max_latency_us=0)

?

Keep them still, since apparently there're ongoing issues w/ your disk and since notably pcie_aspm=off prevented the frequent nvme errors this might be relevant itr.

Can you suspend/resume w/ the LTS kernel?

qherring · 2023-10-30 02:05:10

seth wrote:

Can you suspend/resume w/ the LTS kernel?

Just tried LTS again. It asks for my password to decrypt as usual and when I enter my password it does nothing. I know it is not a keyboard input issue since I can enter the wrong password and it will tell me so. Also when I input the correct password my cpu fan runs (this is typical of when I enter my password correctly at this stage on linux kernel) indicating that it is processing. I would not say it freezes since the terminal cursor continues blinking, but nothing ever happens after this. At this point I can only reboot with the physical power button.

Last edited by qherring (2023-10-30 02:06:52)

seth · 2023-10-30 15:44:57

After my system is running for a few days I cannot put it to sleep nor reboot properly (I have to hard reset with power button). Appears some issue with udev-worker being unresponsive. My system activity tell me the unresponsive process (udev-worker) is in "disk sleep" and I cannot kill it no matter what I try.

Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdd: Worker [488008] processing SEQNUM=488611 killed
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdc: Worker [487982] processing SEQNUM=488612 killed
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sde: Worker [488016] processing SEQNUM=488613 killed
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdd: Worker [488008] terminated by signal 9 (KILL).
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdc: Worker [487982] terminated by signal 9 (KILL).
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sde: Worker [488016] terminated by signal 9 (KILL).
Oct 03 11:53:20 mundofundo kernel: task:(udev-worker)   state:D stack:0     pid:398595 ppid:1      flags:0x00004006
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sdb: Worker [488014] processing SEQNUM=487482 killed
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sdd: Worker [488005] processing SEQNUM=487489 killed
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sde: Worker [487972] processing SEQNUM=487494 killed
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sdc: Worker [487938] processing SEQNUM=487484 killed
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sdb: Worker [488014] terminated by signal 9 (KILL).
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sdd: Worker [488005] terminated by signal 9 (KILL).
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sde: Worker [487972] terminated by signal 9 (KILL).
Oct 03 11:49:52 mundofundo systemd-udevd[458348]: sdc: Worker [487938] terminated by signal 9 (KILL).
Oct 03 11:50:53 mundofundo systemd-udevd[458348]: sdd: Worker [488008] processing SEQNUM=488611 is taking a long time
Oct 03 11:50:53 mundofundo systemd-udevd[458348]: sdc: Worker [487982] processing SEQNUM=488612 is taking a long time
Oct 03 11:50:53 mundofundo systemd-udevd[458348]: sde: Worker [488016] processing SEQNUM=488613 is taking a long time
Oct 03 11:52:58 mundofundo kernel: task:(udev-worker)   state:D stack:0     pid:398595 ppid:1      flags:0x00004006
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdd: Worker [488008] processing SEQNUM=488611 killed
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdc: Worker [487982] processing SEQNUM=488612 killed
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sde: Worker [488016] processing SEQNUM=488613 killed
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdd: Worker [488008] terminated by signal 9 (KILL).
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sdc: Worker [487982] terminated by signal 9 (KILL).
Oct 03 11:52:58 mundofundo systemd-udevd[458348]: sde: Worker [488016] terminated by signal 9 (KILL).
Oct 03 11:53:20 mundofundo kernel: task:(udev-worker)   state:D stack:0     pid:398595 ppid:1      flags:0x00004006

Do you have a clean (of nvme errors) and complete system journal, notably to illustrate the role of the sata devices that show up in the original journal (and whether sda isn't affected and maybe why9?

qherring · 2023-11-04 17:22:39

Four days ago I ran an update. so far no suspend issues since this.

linux-firmware was updated, perhaps this did it.

I will keep the system running to verify if this indeed solved it.

Arch Linux

#26 2023-10-26 20:32:17

Re: Failing to suspend - udev-worker refusing to freeze

#27 2023-10-26 20:55:25

Re: Failing to suspend - udev-worker refusing to freeze

#28 2023-10-26 21:27:15

Re: Failing to suspend - udev-worker refusing to freeze

#29 2023-10-26 23:14:15

Re: Failing to suspend - udev-worker refusing to freeze

#30 2023-10-27 06:03:45

Re: Failing to suspend - udev-worker refusing to freeze

#31 2023-10-30 02:05:10

Re: Failing to suspend - udev-worker refusing to freeze

#32 2023-10-30 15:44:57

Re: Failing to suspend - udev-worker refusing to freeze

#33 2023-11-04 17:22:39

Re: Failing to suspend - udev-worker refusing to freeze

Board footer