Rebuilt system won't boot after being dormant for 2 years

… why don’t you try to boot the system like this:

at the grub prompt (which you get to see when you hit ESC)
you press e for edit
and then append
init=/bin/bash
to the kernel command line

then boot using F10 or CTRL-x
it’s all written at the bottom of that same page

You will be in a very limited system which you cannot even properly shut down.
But you will be root and can change any users password …

if you do:
ls /home
you will recognize your username

change password with:
passwd username
or even root’s with just:
passwd

You may have to remount / read-write - I was surprised that I did not have to do that.

https://www.linuxtechi.com/boot-arch-linux-single-user-mode-rescue-mode/

Even if he got to that limited system, he would still have a 2 year old Manjaro install. Trying to update that would be a disaster. It would likely try and update grub, and might leave him with a system that doesnt boot into anything.

1 Like

That is how.

There is little difference between a 2 year old system and a year and a half year old system. Since you are unsure of when you last stopped using the system that year and a half might be a year and three fourths. Even a year and a half year old system is likely not to have the drivers that your new hardware needs. Since there is almost 8 months between the 19.0 release and the release of the CPU.
You are also discounting the fact that trying to update anything older than a month old install can lead to issues. A year and a half? Its likely a huge waste of time.

The kernel command line is ‘Boot: Manjaro.x86_64 kde’? So the edit would result in
‘Boot: Manjaro.x86_64 kde init=/bin/bash’?

EDIT: Note that when I press E for edit, I get a screen which says “GNU GRUB version 2.04~manjaro” at the top, and below there, in a blue outline box, there is a rather lengthy entry which starts with

“setparams 'Boot: Manjaro.x86_64 kde ’ 'x86_64 lang=en_US” and continues for severaal more lines. This is when attempting to boot to the live dvd, which gives me a menu where I can boot to the dvd or opt to “Detect EFI bootloaders”. The latter option displays two bootloaders, which I assume reflects the live dvd and the old system drive.

No.
What you have here in front of you is not the kernel command line.
I do not know what it is that you see or how you got that …

Have a look at the link I provided.
Find out how to display the grub menu.
From there select edit (E) …

The kernel command line you should see looks something like this:
linux /boot/vmlinuz-5.15-x86_64 root=UUID=3832c08b-4825-4f25-8db4-f47b5e9585de rw quiet udev.log_priority=3

yes - this is the right place - the second to last line …

10-4. Thanks for the further explanation. FYI I was editing my comment when you posted, so if you take another look you’ll see how I got that haha. I will look at that link now.

I saw that and have expanded my post as well.

Ah, yes, I see that you saw my edit. Great. it helps so much to know you are on the same page with someone haha. Looking at your link now.

you shouldn’t need to, now that you know you are in the right place - but just didn’t identify the correct line among all that text

Currently running: 5.4.18-1-MANJARO (linux 54)

so you managed to enter tty?
you can run a system update, but be aware that it probably wont work:
sudo pacman-mirrors --fasttrack 5 && sudo pacman -Syyu

Thanks, brahma. I did run up against the login, though, so there’s that. Just starting from scratch may be the way I have to go, but just in the trying I’m learning a lot. I really appreciate all your input.

yes, just do a fresh install

Ok, so, the command line was very short. It was something like “/boot/vmlinuz-${2}”, but I didn’t write it down I’m afraid. Anyway, not sure how much the syntax matters, but I added the command you suggested and I got this result:

ERROR: Root device mounted successfully, but bin/bash does not exist.
Bailing out, you are on your own. Good luck.

sh: can’t access tty; job control turned off
[rootfs ]#

Quite a deviation from what the page you linked says I should see. I don’t know what the “fs” part of [rootfs ] is, but commands don’t seem to be doing anything. I am going to try again and try to be sure I am doing it correctly.

of course the syntax matters - it has to be this - and nothing else

… but you (most likely) didn’t add it in the correct place

you append it, with a space in between, to the end of the line that is the kernel command line
which is the one starting with “linux”
and it is the second to last line in what you see …

Yours should look very similar to the one I posted - which is from my current system.

the alternative - which would be that /bin/bash doesn’t exist - is not plausible at all
… you could try /bin/sh instead - which should default to bash

This is all just to enable you to reset your password, you know?
To then enable you to boot into emergency mode
or into single user mode
which is where from you then can attempt to update your system.

a chroot from the live system on the CD would probably work as well - if the CD can get you to a TTY


as of now:
you are in the rootfs - which is
the initial RAM disk (or: initrd)
the filesystem on your harddisk is not mounted yet

I have not just written about it - I have literally done this procedure like 10 times by now (in the last hour) :wink:
reset my passwords twice …
I know that it works …

You should edit the title of this thread to reflect what the problem is. Perhaps " New system build won’t boot on system that hasn’t been updated in a year and a half"

what you can do also :
on windows , create a USB with Ventoy
download iso manjaro and drag& copy to USB
then try to boot with USB manjaro as live iso

  • you can do a chroot to see what you need ( list package installed , maybe save /home/< user > )
  • if you try to update , it will break
  • also check version minima kernel that support 5700G ( arch + cpu + apu/gpu )
1 Like

Weirdly, that isn’t the case. The command line I get is:

linux /boot/vmlinuz-$2

That’s all there is. Nevertheless, I must have typed something incorrectly (perhaps I typed initrd instead of init or omitted a “/”), because when I ried it again I got something similar to what the linuxtechi page is saying I should see. But instead of the first three lines that start with “Starting version. . . .” and “dev/sda1:. . . .” and “dev/sda1:. . . .”, respectively, it’s just a single line that says:

[ 0.0035651] do-IRQ: 14.55 No irq handler for vector

but from this point it matches, with the next line reading:

bash: cannot set terminal process group (-1): Inappropriate ioctl for device

“ls /home” did nothing, though I was able to reset the password. Perhaps i need a command that directs it to look at the system drive rather than the live dvd drive (which would never have had a username set for it)? That way at least I know whether the info being returned actually correlates to that disk. Running

fdisk -l

tells me that the linux system drive is being read as /dev/sda1 through /dev/sda4. The kernel appears to be /dev/sda2, with a size of 15.6G and ‘type’ indicated as “Linux swap”. sda3 and sda4 are indicated as “Linux filesystem”. I don’t know why I created the exttra partitions.

I have indeed created a bootable usb. I was hoping there might be a not-too-complicated way to recover my system without having to go about a fresh install. This is beginning to look like it is not doable.

I was doing some reading about using chroot for system recovery and it looks like it involves a degree of difficulty which would be too time consuming for my limited knowledge of linux commands.

We will see, but I’m about at the point where I’m going to just have to throw in the towel and just do a new install. Thank you for the suggestion, though.