Help! 
I have an HP Zbook Firefly G10 Mobile Workstation with Tumbleweed installed for some time. No problems until today. The first thing I noticed is it couldn’t open Firefox. Not responding. Then I opened dolphin and it worked, but with a delay. It gave an error that showed the path of the file I was opening and said “not writable”.
The system was up to date. So I decided to power it down and reboot. On power up it went into emergency mode. It asked for root password and suggested the command: journalctl to look at the log files. I can see those now and there are several warning messages and errors, but I don’t know how to interpret them or what to do from here. I don’t know how to make that log file visible here other than taking pic’s with my phone. I could also just type the warnings and error lines. There aren’t too many.
Any help appreciated.
See if you can boot from a read-only snapshot from Grub. If that works, go to a tty from the login screen and do
sudo snapper rollback
If I’m doing what you mention correctly, none of the snapshots work. I went to the “read-only” snapshot option and clicked on the first option. But when I tried to boot, I got the same error message. I tried every snapshot that was there and none worked.
I noticed the first message I get with the failed boot is "HP PID log size 0 is invalid.
Not sure if that helps or matters. Any other ideas?
We would need to see more, post a picture from your phone ?
These are in reverse order of the actual output. There are 5 more screenshots I will post below. But these 5 show most of the warnings and errors.
Here’s your issue
The IO error is pretty worrying.
Any ideas how to move forward?
Q1: Is there any essential data on the disk? If so:
Q2: Do you have a proper backup of your homedir?
To explain, it only processes the initrd, but it cannot access the root disk file system because there’s a wrong superblock. btrfs cannot even fix it itself and that is not a good sign. You could boot from a Live USB and see if you can get it mounted.
I’ll give that a shot. I do have a back up from a few days ago, but it would be nice not to have to install a fresh version.
From the Live USB, consult the openSUSE btrfs wiki pages, don’t just run any commands on it. Some can be really destructive
I’m able to use rescue on the Live USB and get to the tty1 mode. Is there a way to mount from here?
I tried to boot from the USB, but it said no bootable media
I have no idea what you mean here. This says both you can, and you can’t.
BTW, an install USB would work too, as long as you can boot from it
When I booted from the USB, there were 3 options. I can’t recall one, but the other two were “Rescue” and “Boot”. I can get the exact wording if it helps. Anyways, I first tried the boot option, but that failed. Then tried rescue and it worked, but I’m just in the command line ttyl mode.
Does that make more sense? Thanks for being patient
in tty1 after logging into root, it suggests: try mount-rootfs-and-do-chroot
When I do that it says: mount-rootsf-and-do-chroot [OPTIONS] DEVICE
Mount root filesystem found on DEVICE to temporary location, taking care of special mount points like /dev, /sys, and /proc
Options:
-r, --readonly Mount DEVICE read-only
-h, --help Show these usage instructions
Is this an option for moving forward?
Once the installer has properly started, hit Ctrl+Alt+F2 . You’ll see a root prompt. Now try to mount:
mount /dev/nvme0n1p2 /mnt
If that does work cd into /mnt, but my bet is it won’t. There are a couple of options:
- btrfs got corrupted beyond repair, would need a reinstall
- the nvme is broken
- the motherboard controller is broken.
Yeah, but it’s better to test mounting first. What you describe works like this on your system:
mount /dev/nvme0n1p2 /mnt
mount --bind /dev /mnt/dev
mount --bind /sys /mnt/sys
mount --bind /proc /mnt/proc
chroot /mnt
mount -a
