Very unstable system?!

you mentioned two disk, one with win and one with manjaro, but in the

there is windows and also manjaro installed on the same disk?


what file system are you using? is it encrypted?


fast boot disabled in bios - you may not have this option;
secure boot disabled in bios;
AHCI enabled in bios - if you have RAID, and switch to AHCI, you will not be able to boot into windows;

how did you flash the iso? did you use Ventoy?

From live session you do not have access to any of the linux data you have, you can’t mount those partitions from a File Manager either?

might be helpful:

Windows is on the other one.

For linux I’m using f2fs with no encryption.

Those are disabled.

If you are referring to this:

It is likely but, I don’t really remember. It’s more probable there was something wrong with my copy of a old manjaro iso I have been using back then. I’ve just remembered that I had to quickly compile some c++ code on that system, and it did not compile properly. It was not sequential: it randomly jumped from line to line, always in the same wrong order (sometimes it tried to disallocate memory even before it was allocated)

Now I am using “Win32DiskImager” on windows.

I was not able to mount either vfat with grub nor f2fs with manjaro. File manager also can’t mount them. Other partitions work fine.

Thanks! I will try this when I’m back.

EDIT:
This tool seems to be really amazing, but sadly I had little to no use out of it.
After running “fix linux system” (or something similar, there was only one option that seemd like it could help)
I believe it just runs fsck.

Something worth mentioning: this system couldn’t mount my linux system partition.
After running almost any of it’s repair tools, there is a prompt to select partition to run it on. Next to this system partition there is: “unable to mount” message.

use Ventoy, and download the latest iso from manjaro.org


also check this link regarding f2fs - which also could be a reason for your issues:

I really do not know how I didn’t have found this tool before. It makes life so much easier.

I’ve got that problem before. I did solve it by physically removing drive containing windows and connecting it again after installation.

yes ventoy is very good…
did you run fsck on your system?:

sudo fsck.f2fs -f /dev/nvme1n1p2
sudo fsck.f2fs -f /dev/nvme1n1p5

Yes

The output now says:

fsck from util-linux 2.38.1
Info: Force to fix corruption
Info: MKFS version
  "Linux version 5.15.60-1-MANJARO (builduser@fv-az462-678) (gcc (GCC) 12.1.1 20220730, GNU ld (GNU Binutils) 2.39) #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022"
Info: FSCK version
  from "Linux version 5.15.60-1-MANJARO (builduser@fv-az462-678) (gcc (GCC) 12.1.1 20220730, GNU ld (GNU Binutils) 2.39) #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022"
    to "Linux version 5.15.60-1-MANJARO (builduser@fv-az462-678) (gcc (GCC) 12.1.1 20220730, GNU ld (GNU Binutils) 2.39) #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022"
Info: superblock features = 0 : 
Info: superblock encrypt level = 0, salt = 00000000000000000000000000000000
Info: Segments per section = 1
Info: Sections per zone = 1
Info: total FS sectors = 209715200 (102400 MB)
Info: CKPT version = 5ee4987f
Info: Checked valid nat_bits in checkpoint
Info: checkpoint state = 281 :  allow_nocrc nat_bits unmount
[ASSERT] (sanity_check_nid: 404)  --> nid[0x3] ino is 0

[FSCK] Max image size: 244 MB, Free space: 100666 MB
[FSCK] Unreachable nat entries                        [Ok..] [0x0]
[FSCK] SIT valid block bitmap checking                [Ok..]
[FSCK] Hard link checking for regular file            [Ok..] [0x0]
[FSCK] valid_block_count matching with CP             [Ok..] [0x0]
[FSCK] valid_node_count matching with CP (de lookup)  [Ok..] [0x0]
[FSCK] valid_node_count matching with CP (nat lookup) [Ok..] [0x0]
[FSCK] valid_inode_count matched with CP              [Ok..] [0x0]
[FSCK] free segment_count matched with CP             [Ok..] [0xc71b]
[FSCK] next block offset is free                      [Ok..]
[FSCK] fixing SIT types
[FSCK] other corrupted bugs                           [Fail]
Info: Duplicate valid checkpoint to mirror position 512 -> 1024
Info: Write valid nat_bits in checkpoint

Done: 0.000000 secs

Does this mean that for some reason all data was erased?

if its the 100gb nvme1n1p2, then it looks like its really empty …
and you cant access the f2fs partitions?
and if you check in kde partition manager, did it shows that there are some files?

Only system one now, I couldn’t access vfat /boot/efi pratition before, but I have erased it few times trying to reinstall grub. Other partitions work fine.

Partition manager says nothing about files, but I don’t really believe what it says, there was never anything about flags or mount points, just blank spae.

with files i meant, if it shows how much is used, or if it shows the partition as empty…
since you cant access it, you could try reinstall again, but this time dont use f2fs

I can’t force it to work. It just shows:

Size:         100.00 GiB
Available:    --
Used:         --

I will try, but before that I will create bit to bit copy of that partition using dd. Maybe I will find a way to restore damaged part of that partition.

I don’t think f2fs is a reason of this failure. I’ve been using it for years now without an issue. Only recently after I’ve reinstalled windows (this time on GPT, before it was on MBR) strange stuff started to happen.

I suggest updating BIOS before reinstalling system - Asus PRIME X470-PRO - BIOS & FIRMWARE

Really good suggestion. Last time I did it was something close to one year ago.

I’ve just finished cloning partition and I don’t really understand the output.
I’ve used clonezilla partclone with dd mode and -rescue flag:

Summary of image checking:
==========================
GPT 1st partition table file found!
GPT 2nd partition table file found!
GPT partition table file for this disk saved by gdisk was found: nvme1n1
This is not an image for whole disk. Skip checking swap partition info...
MBR file for this disk was found: nvme1n1
Image was saved by dd or partclone.dd. No need or no way to check the file system integrity: nvme1n1p2
The image of this partition is restorable: nvme1n1p2

PS. Next time you can run this command directly:
/usr/bin/ocs-sr -q1 -c -j2 -rescue -z1p -i 102400 -sfsck -senc -p choose saveparts 2022-11-11-12-img_MANJARO_FAILURE_BACKUP nvme1n1p2

I don’t really understand why there is MBR file, probably it is normal, but is it?

No it isn’t. Guess you didn’t recreate the gpt table. Windows like to have hybrid mod MBR+GPT:

In my experience mixing them is a bad idea on Linux.

So I have reinstalled the system, after updating the BIOS. This time I am using ext4 with disabled journaling. Overall fast write speed that f2fs offers would mostly be useful only when installing big updates (so not that useful)

For now I didn’t have any signs of unstable behavior. I will mark this as solution if after two months there will be no marks of instability.

From now I will do weekly system backups and I would recommend that to anyone.
When using such amazing tools as ventoy and clonezilla it is much less time consuming than trying to recreate all of data that could be lost,
take my word for it :)

And of course thanks to everyone here for time and afford!

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.