65
submitted 2 days ago* (last edited 1 day ago) by [email protected] to c/[email protected]

My Fedora 40 system won't boot with any kernel over 6.8.10.

I've had this problem for about a month now and have been putting it off, as I haven't really been using my computer recently, but it needs to be addressed.

I'm running a Fedora 40 KDE 6 distro, which was upgraded from Fedora 39. It's been working pretty well for the half year I've been using it but recently I've encountered an issue. About a month ago, after upgrading my system to kernel 6.8.10, my system started to hang while in boot.

usb 1-10 device descriptor read/64, error -71 [ OK ] Started plymouth-start.service - Show Plymouth Boot Screen [ OK ] Started systemd-ask-password-plymoūquests to Plymouth Directory Watch [ OK ] Reached target paths.target - Path Units. [ OK ] Found device dev-disk-by\x2duuid-dūsung SSD 970 EVO Plus 500GB fedora. [ OK ] Reached target initrd-root-device.target - Initrd Root Device. [ *** ] Job dev-mapper-cl\x2dswap.device/start running (xxs / no limit)

Fortunately Fedora saves the previous 2 kernels versions and I was able to boot into my computer using Kernel 6.8.9. When I went to check what was wrong I noticed that with kernel 6.8.10 it always hangs at this part of the boot process.

[ *** ] Job dev-mapper-cl\x2dswap.device/start running (xxs / no limit)

I understood that this has something to do with swap but I'm unsure what the issue is exactly. My Fedora install doesn't have a swap partition, it uses zram. I'm unsure if that's an issue.

I've reinstalled kernel 6.8.10 and it didn't fix the issue. I've also upgraded my machine to use kernel 6.8.11 and see if that would fix anything, but it did nothing, and upgrading to 6.8.12 probably won't fix anything either. I've installed a dnf plugin called versionlock meant to pin certain kernels as to not delete them and have already pinned kernel 6.8.9, but I'd still rather avoid upgrading.

What I really don't understand is what changed. Why does my system boot successfully in kernel 6.8.9 but fails in 6.8.10. I've read that others have had a similar experience with the 6.8.10 kernel on fedora, albeit for different reasons.

If someone can point me to the answers that would be great, but an explanation as to how to read that log and steps I could take to identify and troubleshoot would be just as welcomed.

UPDATE: In my /proc/cmdline I had an argument known called resume=/dev/mapper/cl-swap which was trying to find a swap partition that didn't exist.

I used the command cat /proc/cmdline to view my boot arguments and then ran the command sudo grubby --remove-args=<resume=UUID=xxx> --update-kernel=ALL where <resume=UUID=xxx was replaced with resume=/dev/mapper/cl-swap.

Where I got my answer

Thank you [email protected] for linking the thread.

top 20 comments
sorted by: hot top controversial new old
[-] [email protected] 6 points 1 day ago
[-] [email protected] 2 points 1 day ago

Thanks for linking the thread, I spent a whole day looking through fedoraproject threads, don't know how I missed this one. Again thanks.

[-] [email protected] 24 points 2 days ago

Boot to the previous kernel and run updates until you get a 6.9, or go download and install the rpms yourself.

They pushed a bad patch with 6.8.10 I think? They had to roll it back and push another real quick, but some caching issue still delivered it to a bunch of people. You should on the 6.9 line now anyway as 6.8 is EOL.

[-] [email protected] 2 points 1 day ago

Sorry for responding so late I was busy at the moment.

I didn't fully understand what you meant by " run updates until you get a 6.9"

Also what exactly should I be downloading and installing. The new patched kernel?

[-] [email protected] 3 points 1 day ago

When you run 'dnf update ', you want it to say it's going to install kernel 6.9.X.

[-] [email protected] 2 points 1 day ago

Hello I did as you advised and my system installed the kernel 6.9.4

Nothing really changed. Still hung at that same exact spot. Honestly the only thing I'd like right now is to know what Job dev-mapper-cl\x2dswap.device/start means

Still thank you for replying

[-] [email protected] 2 points 1 day ago* (last edited 1 day ago)

See if it's the swap, disable zram and/or add a swap file.

Doesn't have to be a swap partition, you can create a file, format it as swap and assign it in /etc/fstab.

Btw when you say you don't have swap do you mean you don't have regular swap file/partition (because you have zram swap) or you don't have swap at all?

[-] [email protected] 1 points 1 day ago

That is a systems job for finding and mounting swap space on your machine. Is it failing?

[-] [email protected] -2 points 2 days ago
[-] [email protected] 3 points 1 day ago

dev-mapper swap...

Any resume argument in your boot commands? Try removing them and see what happens

[-] [email protected] 4 points 1 day ago

Fedora's curremt kernel version is 6.9 so test with that

[-] [email protected] 3 points 1 day ago
[-] [email protected] 4 points 2 days ago

Unless you have some kind of knockoff SSD, that ūsung SSD looks like something is corrupted to me. usb 1-10 device descriptor read/64, error -71 might be unrelated.

This could be a problem with RAM defects or overclocking. If your computer is overclocked, try setting it to stock configuration. Also run a memtest to check if your RAM sticks aren't going bad. I don't know what might've changed between 6.8.9 and 6.8.10 to cause this, but it could just be a coincidence (i.e. the kernel defaulting to a different RAM page that suffers from corruption for whatever reason).

These messages are actually part of the systemd startup sequence, so the kernel has already loaded at this point. This means the problem may not be the kernel, but the initramfs installed/generated for your computer. You can try regenerating your initramfs on Fedora by running dracut --regenerate-all as root. Before you do that, you may also want to double check your /etc/fstab to make sure nothing accidentally added a swap device for some reason.

[-] [email protected] 4 points 2 days ago
[-] possiblylinux127 2 points 1 day ago

It looks like it is booting fine to me. If you get to systemd the kernel is at least mostly working

[-] [email protected] 1 points 2 days ago
[-] [email protected] 6 points 2 days ago

This is a specific case.

[-] [email protected] 1 points 2 days ago

@Doctor_Rex What happens when you boot the recovery/rescue kernel, or add the flag nomodeset to the current kernel? I feel it *is* booting, just not showing anything (so a GUI issue), sadly I've seen that before.

this post was submitted on 02 Jul 2024
65 points (98.5% liked)

Linux

45525 readers
1282 users here now

From Wikipedia, the free encyclopedia

Linux is a family of open source Unix-like operating systems based on the Linux kernel, an operating system kernel first released on September 17, 1991 by Linus Torvalds. Linux is typically packaged in a Linux distribution (or distro for short).

Distributions include the Linux kernel and supporting system software and libraries, many of which are provided by the GNU Project. Many Linux distributions use the word "Linux" in their name, but the Free Software Foundation uses the name GNU/Linux to emphasize the importance of GNU software, causing some controversy.

Rules

Related Communities

Community icon by Alpár-Etele Méder, licensed under CC BY 3.0

founded 5 years ago
MODERATORS