What caused corruption of my install?

[etcetera]

It is an Ubuntu 22.04 install but I do not think it is an Ubuntu issue. So putting it here for a more generic approach.
I triple boot Win10, Ubuntu 22.04 and RHEL 8.6.

I was moving some of the SSDs around, swapped the primary boot SSD, which is Win10 with something else and that is when it basically started. I tried to boot without the Win10 SSD and got the following (not an exact result but almost identical).
Basically what changed the disk ID and how? Does Windows have something to do with it? Trying to understand so it does not happen again.

Possible to recover this or just faster to reinstall? Thankfully it's not my primary install but just a clone I use in place of primary, so I could reclone the primary from another install.

Gave up waiting for root device. Common problems:
— Boot args (cat /proc/cmdline)
— Check rootdelay= (did the system wait long enough?)
— Check root= (did the system wait for the right device?)
— Missing modules (cat /proc/modules; ls /dev)
ALERT! /dev/disk/by-uuid/aa91b9fe-1e27-xxxx-xxxx-7xxxx7d4exxxxx5 does not exist.
Dropping to a shell!

BusyBox v.1.13.3 (Ubuntu 1:1.13.3-1ubuntu11) built-in shell (ash)
Enter 'help' for list of built-in commands.

(initramfs)

 

[APTI]

no idea what system you are on. I triple boot fedora 40, win 7 and win 10. each on it's own ssd. refer to the article I wrote about multiple boot, it may apply to you and help. https://www.linux.org/threads/dual-and-multi-boot-systems.47320/

 

[etcetera]

MSI Titan GT80
Each OS is on its own disk. They never intersected.
I have taken disks and transplanted them elsewhere.
The multiboot thing worked for years until this corruption happened. I usually did Win10, Win10 and Redhat or Oracle Linux or such, Buntu is a relatively new introduction. I don't fully get what and how changed this Ubuntu disk. I am not really concerned as it's not even close to being my primary jnstall, just a back to a backup and I am going to reinstall it but the real focus of the question is not how to fix it but the root cause of the malfunction.

 

[APTI]

windows put out recent updates that screws GRUB. Best option is to not use GRUB. The article I wrote and referenced above tells you how to do that. The fight between windoze and linux never stops and they are like little children fighting. I send them to separate corners using the device in the article and they place nice after that.

 

[Condobloke]

Seeing your problem appears to coincide with the Ubuntu install, pay some attention to the ubiquity installer.

Perhaps install Linux Mint or similar for a quick and dirty comparison?

 

[etcetera]

I do not use grub. I use the built-in F12 function in Dell to select the boot entry. Hit F12 during bootup and it will display all the bootable SSDs.

 

[etcetera]

The problem was that the box for no apparent reason switched from AHCI mode to RAID in BIOS (UEFI). I never went into it or changed it recently. Both were trying to boot but never made it.
I should have realized it was not a Linux thing as the Windows install also stopped working, and that was the common denominator.

 

[Trenix25]

I do use grub to boot up. I can boot the main system on a hard drive, a maintenance system on a flash drive, or Windows 10 on another hard drive inside the computer. The retailer installed it so I just left it there. I don't use it. You should always specify your file systems with a UUID instead of something like /dev/sda1. Make sure that all of them are actually unique. I found that I have to tell grub to load the Linux kernel in such a way as to cause the root file system to be mounted read only at first, or I also get that (initramfs) prompt when the root file system refuses to mount. The linux command loads the Linux kernel when using grub in grub.cfg. You can specify which file system to use as root and whether to mount it read only or read/write by using linux /path/to/kernel/image root=UUID=YourRootFilesystemUUIDHere ro quiet to tell the kernel to mount your root file system using a specific UUID and to initially mount it as read only. The word quiet suppresses a lot of extra debugging messages. You will then need to remount root to read/write if you mount it as read only at first. I could never get my root file system to mount if I tried to mount it as read/write at first. It always had to be mounted as read only, which is helpful considering I want to check it first with e2fsck without having to boot to the maintenance system first. I normally remount root as read/write while still in single usere mode before mounting any of the other file systems. I wrote my own boot up script that runs in single user mode before switching to multi-user mode by moving to the graphical target. I had to do a little tinkering to get everything to start up just right. It's my computer after all so it should work the way I want it to. Isn't that kind of the point of Linux, to manage my own computer, instead of letting other people do so?

Signed,

Matthew Campbell