Skip to main content ↑ ↓
Topic solved
This topic has been marked as solved and requires no further attention.
Topic: Artix no longer boots after SSD and (likely) kernel upgrade (Read 333 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Artix no longer boots after SSD and (likely) kernel upgrade

I've got an issue (if not more than one) that I'm running into with a server I installed Artix on, and I haven't been able to figure it out yet.

A few important details:
  • I'm using btrfs
  • rEFInd
  • I upgraded the system with a new SSD, using DD to copy bits from old to new, then expanding the partition/volumes to fit the larger new drive.

Both the old and the new drive are still in the system, and while they initially had identical UUIDs, I did eventually change both the UUID and PARTUUID

I've made a disk image backup of my root partition using dd, copying from the data partition into a disk image, so I shouldn't lose much at this point if I try troubleshooting in ways that are potentially destructive.

I recognize that I should probably be running Ubuntu or Debian for a server like this (and I'll probably switch sometime in the future), but for now, I'm trying to get back to an operating/functional state.

So, if I boot what I assume is the correct boot item from my UEFI, then select either Artix boot options in rEFInd, it throws me into an emergency shell identical to the one shown here:

Code: [Select]
ERROR: Root device mounted successfully, but /sbin/init does not exist.
sh: can't access tty: job control turned off
Bailing out, you are on your own now. Good luck.
[rootfs /]# _

When in this shell, I'm able to see /sbin/init, but it's a (sym)link referring to busybox, not the actual init file...but I also don't know if busybox is meant to be "init" at this stage

I know this is all over the place, but I'm trying to provide the information I get from approaching this from different angles, and it's a lot.

Please feel free to ask any questions needed for clarification.

I'm pretty desperate here, as there are services both I and my family rely on that are down because of this. Thanks.[/list]

Re: Artix no longer boots after SSD and (likely) kernel upgrade

Reply #1 –
This thread seems to address a similar problem, but reinstalling base in the chrooted environment did not fix the issue.
https://forum.artixlinux.org/index.php/topic,2444.15.html

When chrooted, I've first mounted my partitions using these commands:

Code: [Select]
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=256,subvol=/@ UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=257,subvol=/@home UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/home
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=258,subvol=/@swap UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/swap
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=260,subvol=/@srv UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/srv
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=261,subvol=/@log UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/var/log
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=262,subvol=/@tmp UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/tmp
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=263,subvol=/@abs UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/var/abs
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=264,subvol=/@pkg UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/var/cache/pacman/pkg
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=5,subvol=/ UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c       /mnt/btrfs
mount -o rw,noatime,nodiratime,compress=zstd:3,ssd,discard=async,space_cache=v2,autodefrag,commit=120,subvolid=265,subvol=/@snapshots UUID=f5377f96-2d42-45ba-95d3-dd16455dfb1c /mnt/.snapshots

mount -o rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,utf8,errors=remount-ro UUID=76BD-D757 /mnt/boot

Unfortunately, the result is the same even after this.
 

Re: Artix no longer boots after SSD and (likely) kernel upgrade

Reply #2 –
After a number of troubleshooting steps, I chrooted into the system via live-usb and reinstalled linux and linux-firmware, which finally fixed one of my problems.

The initial issue I was facing was that something went wrong with docker/Portainer, so I rebooted the system. After a reboot, the only network interface visible was lo0. This meant that the built in NIC e1000e was not being started/used.

This resulted in Docker not starting (I guess), and obviously no network connection.

When I reinstalled the linux and linux-firmware in chroot and rebooted, everything just came back up as it should.

I need to clean up my boot menu and probably move to Ubuntu/Debian for this server, but for the moment, everything appears to be back to normal.

TL;DR: reinstalled linux and linux-firmware via artix-chroot, all worked after that.