Fresh install of Manjaro failing with Installer/KDE/GParted

@Ben I thought of that too - but since the disk had two partition - listed from lsblk and one of those was ext4 - I assumed the disk was not locked.

@BrightAnima Now the disk has been cleared.

The message is expected - as we have deliberately destroyed all signatures and file system information.

This means the disk can be used in the installer.

I will say that your issue is rare and unexpected.

btrfs is a great file system which btrfs stores data completely different for other file systems, as it does not use partitions but instead sub-volumes which are dynamic data structures. Free space is thus a completely different term with btrfs - this has often created confusion.

btrfs is great for data security, if data is somewhat unchanging - like documents,pictures,movies etc. - but it is not so great for systems where data changes all the time; on such systems one must be vigilant so the file system is not overloaded.

2 Likes

Having checked on Windows, Hibernation was indeed here because the file was present in the system folder, so I turned it off.

I’m gonna head back to the LE now to try again.

Well, it still shows me the same error of the installer trying to create a partition, but fails

Strange - what is the output from

lsblk -f

It is pretty much the same as the output from earlier, with the only change being the NVMe having absolutely nothing now

And looking on GParted, the disk is unallocated and the only warning I have is ‘unrecognized disk label’

That is expected - it means the device is writeable - there is something weird with the disk though.

Just out of curiosity, would it help if I say that the installer (or KDE) doesn’t find any valid partition tables on any of the of internals?

I also tried running the command the installer tried and it gives me the shared libraries error and cannot read file data I/O

That will point to a system firmware setting - but I have no idea which.

I do remember though - I have seen many topic where

with

behaves very strange when used with Linux kernel

It’s very odd, because the first time I got it to run (on btrfs), it was running just fine. But now it doesn’t work. :thinking:

Did you change any firmware related between the attempts?

Because that is the only thing that would prevent recognising the disks.

Pure speculation

Speculating if the board expose the disk as vmd or something similar - I vaguely recall something about editing the grub kernel cmdline and adding vmd.

I cannot recall the exact wording but that may be worth researching.

I checked on Github (GitHub - manjaro-plasma/download: Download Manjaro Plasma Edition - Developer Builds · GitHub) but there is no ISO with Linux 7.0 - which has some big improvements for AMD based systems.

Fixed the server issue on https://manjaro.dk/iso

Not that I know of, I was mainly starting to learn how to use Manjaro for the first time.

Although I’ll look around see if that ends up being the case with the cmd setting.

Aight, gotcha, I’ll hop on Windows to download it and burn it on the micro SD using Rufus (Or Ventoy if it’s preferred).

While I download and set up it all, since I saw that Linux 7 recently got released, when do you think Manjaro will support it as a stable release?

The Linux 7.0 kernel is in unstable and testing and soon in stable branch too.

The ISO is always provided with LTS kernel - currently 6.18 - but as 7.0 is in the official repo, you can easily change kernel.

Something like

sudo mhwd-kernel -i linux70

As for the release ISO - it will change when kernel.org release the next LTS.

Well, I can troubleshoot some more with the ISO you gave me, and if it still fails, at least I know the stable 7.0 is getting close to release, so I won’t mind staying on Windows until then.

But overall, I’ve gone on a wild ride since Monday night and I’m glad that I could have help on top of my own research!

And I think that I won’t be able to go any further, as after trying both Ventoy and Rufus to get the unstable ISO to boot up, it doesn’t get past the black screen when i used the open source drivers. I can still try with the proprietary drivers, but i’m not sure if it’s gonna work either.

Many users - embracing Linux for the first time - has the impression that a computer is a computer - it is not.

My years - providing computer support for a living - I have learned, there is no such thing as a computer is a computer.

They are as different as human beings and if you compare to a woman - they are to be equally delicate.

Have fun …

Even though I couldn’t get a fix on the ISO I already had, I still massively thank you for guiding me through this strange ordeal of mine, hats to you!

Okay, I have very good news! After reinstalling AtlasOS (custom Windows) on the NVMe it has now allowed me to see the partitions again in Manjaro’s LE @linux-aarhus

I can now install Manjaro in full ext4 on the NVMe!

I personally don’t know what specific setting AtlasOS changed for that to happen, but I’m sure as hell not gonna complain!

Slight update: ‘partially helped’ I managed to install Manjaro, but I have no background and I don’t have access to the home menu, only the Manjaro Hello window with the applications button. I’m gonna try reloading in the LE to get rid of my regular windows installation

Mod edit: Consecutive posts merged. :wink:

Okay, I’m still working on it as I got into the main Manjaro installation with a complete black screen. Though I did find a post explaining the issue, the only problem is that the commands that I tried ended up not working.

With the desktop and firefox packages, it finds them and installs them, but the moment it’s done, it gets me the following error:

warning: Public keyring not found; have you run 'pacman-key --init'?
error: GPGME error: Invalid Crypto engine

So I obviously go through the process of getting said key, only to find that not only it’s not in the system, but it can’t update the trust database to get said key.

I also tried the startx command and it return me an error 1.

Do you have any insight for this one by any chance?

x11 won’t be there by default on new .ISOs. You’ll need to use startplasma-wayland, or maybe dbus-run-session startplasma-wayland. :wink:

So the first one says that it couldn’t start D-bus, so instinctively I use the second command and I have a slew of errors. I’m gonna try to get a clear photo of it for you.

I hope the text can be seen clearly since I don’t have any other way to share huge wall of texts for errors :sweat_smile:


Update Apr 24:

I continue troubleshooting with what I can find on this forum and a few others, where one thing lead me to another, and a third, but now, I think I’m stuck in a sort of loop because, the one thing that I need is the keyring that’s missing, but having tried various options, I don’t really see any exit of this so far.

In short, whenever I try to get something downloaded from the servers, it fails because it didn’t find the public keyring, but to get that keyring, I need to get it from the servers.

The main culprits I found are:

(when using pacman-key --init)

gpg: error while loading shared libraries: libgcrypt.so.20: cannot open shared object file: No such file or directory

chmod: cannot access '/etc/pacman.d/gnupg//trustdb.gpg' : No such file or directory 

ERROR: Trust database could not be updated.

(when trying to download anything from the server)

warning: Public keyring not found; have you run 'pacman-key --init'?
error: GPGME error: Invalid Crypto engine 

I also forgot to mention, removing and resynchronizing the gnupg folder doesn’t change the Trust database issue