I used partition manager on Galileo thumb drive to move my main EOS partition to the left, and the log shows it moved successfully. But when it did a check on the new partition location it says error. Not much else on the log. The contents appear to be there, but under sudo parted -l it has no preset file system type, so i cannot mount it in order to reinstall bootctl or kernels.
What steps can I take to set the file system type back to ext4 and verify the partition move doesn’t have any other errors?
On a similar note, I have the EFI partition to the right of this EOS partition, then free unallocated space to the right of that. But it seems I cannot move the EFI partition even when booted through a thumb drive ISO. Is there any way to move it so that I can combine all unused partitions back to the main EOS partition?
I guess you could do it using the included partition manager but I am not that familiar with this one to tell you the exact steps. Most probably, marking the partition, right-clicking on it or looking into the menus will reveal the option.
Edit:
Here you go:
Partition → Check
Checks a partition: This command checks the selected partition and its file system for errors and tries to correct any problems it finds. During this process the file system on the partition will also be resized to fill the partition completely.
This command is only enabled if a partition is selected that can be checked and is not currently mounted.
I’m not very familiar with these steps so i’ll include the konsole notes. the partition in question is P4
I saw a similar post by freebird that said
"My first attempt at ‘fixing’ would be:
arch-chroot into the system from a live boot (as described in our wiki)
Run sudo blkid to get a list of the current identifiers, especially the ones referring to your ‘lost’ system
Edit (with sudo) your /etc/fstab, so that the NOW correct UUIDs are used to refer to the partitions. It might be well to save a backup copy before modding
restart"
I’m not sure how to edit my fstab with the correct UID, in terms of how the command would look.
[liveuser@eos-2023.11.17 ~]$ sudo blkid
/dev/nvme0n1p9: UUID="49A3-9670" BLOCK_SIZE="512" TYPE="vfat" PARTUUID="ff347e10-282a-bd40-8b5e-74477f6aac16"
/dev/nvme0n1p3: UUID="6DEF-B0F1" BLOCK_SIZE="512" TYPE="exfat" PARTLABEL="Storage" PARTUUID="fd1eed63-48ea-4fd2-895a-bd1561cc6163"
/dev/nvme0n1p4: PTTYPE="dos" PARTLABEL="EOS Plasma" PARTUUID="666c0ee0-43c5-8c41-a3c7-b9250670b0f6"
/dev/sdb2: SEC_TYPE="msdos" LABEL_FATBOOT="VTOYEFI" LABEL="VTOYEFI" UUID="83A6-E98E" BLOCK_SIZE="512" TYPE="vfat" PARTLABEL="VTOYEFI" PARTUUID="b58e8847-c669-3595-2d2c-b35666a2f64a"
/dev/sdb1: LABEL="Ventoy" UUID="1400-C738" BLOCK_SIZE="512" TYPE="exfat" PARTLABEL="Ventoy" PARTUUID="da74ad5d-fe88-1d29-8984-ecc5513246eb"
/dev/loop0: BLOCK_SIZE="1048576" TYPE="squashfs"
/dev/mapper/ventoy: BLOCK_SIZE="2048" UUID="2023-11-17-11-34-04-00" LABEL="EOS_202311" TYPE="iso9660" PTUUID="b64c04a4" PTTYPE="dos"
/dev/sda2: LABEL="Backups" UUID="F65E-F209" BLOCK_SIZE="512" TYPE="exfat" PARTUUID="6bcf4ce5-02"
/dev/sda1: LABEL="Snapshot EOS KDE" UUID="72dec046-8f03-4f2d-9871-c0ddaac58236" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="6bcf4ce5-01"
[liveuser@eos-2023.11.17 ~]$ sudo fsck -V /dev/nvme0n1p4
fsck from util-linux 2.39.2
[/usr/bin/fsck.ext2 (1) -- /dev/nvme0n1p4] fsck.ext2 /dev/nvme0n1p4
e2fsck 1.47.0 (5-Feb-2023)
ext2fs_open2: Bad magic number in super-block
fsck.ext2: Superblock invalid, trying backup blocks...
fsck.ext2: Bad magic number in super-block while trying to open /dev/nvme0n1p4
The superblock could not be read or does not describe a valid ext2/ext3/ext4
filesystem. If the device is valid and it really contains an ext2/ext3/ext4
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
e2fsck -b 8193 <device>
or
e2fsck -b 32768 <device>
/dev/nvme0n1p4 contains `DOS/MBR boot sector' data
[liveuser@eos-2023.11.17 ~]$ lsblk -lf
NAME FSTYPE FSVER LABEL UUID FSAVAIL FSUSE% MOUNTPOINTS
loop0 squashfs 4.0 0 100% /run/archiso/bootmnt
nvme0n1
nvme0n1p3
exfat 1.0 6DEF-B0F1
nvme0n1p4
nvme0n1p9
vfat FAT32 (EFI) 49A3-9670
[liveuser@eos-2023.11.17 ~]$
Glancing through the output of the fsck command, I am afraid your partition has gotten corrupted somehow. I would rather not give you any suggestion to not make the matter worse.
I step aside and let other forum members with more experience in dealing with these sort of the problems to assist you.
Thank you. I do have a Timeshift backup on a thumb drive, if need be. All my data is backed up on clouds as well, the only thing I’d really lose if there’s no saving this is my OS/desktop setup/settings and app settings. Would Timeshift be able to restore those things onto a fresh Plasma install?
Do you think the partition move would have gone any differently had it be GParted instead of the partition manager on Galileo’s ISO?
I wouldn’t know. There is always a certain risk for data loss or corruption in partition management and I would say especially when moving them around.
Your app settings are stored in dot files and dot folders under your home directory. For example if using Firefox then in ~/.mozilla. Most of the apps have their settings in ~/.config.
Also, if you have done some tweaks system-wide, they are generally done under /etc. So if you have a backup of that directory, you are pretty much covered.
In the unfortunate event that there will be no way of rescuing the partition and if you have to reinstall, I personally wouldn’t do a full restore of everything from a snapshot. I would restore the settings step by step on a per case basis.
So I wiped my entire drive since I need to work and couldn’t keep messing with that partition. But when I do a time shift restore, I get errors once it restarts and you’re right, it’s in the uuid. Is there a way to change the uuid in whatever time shift restored? Unfortunately I’m stuck in this window after time shift. Is there a way to change the uuid in whatever time shift restored? @kresimir
Unfortunately I’m stuck in this window after time shift. I don’t think there’s a way to edit files in time shift backup, to do a restore with corrected uuid? Or maybe I can change my new uuid to match the old one?
Do you mean the fstab info in the time shift backup or in this boot up screen after the restore was made? If the latter, how do I find the new UUID to edit? If I try to edit fstab it says permission denied but I’m in root, is there a specific command that would override this? Sorry, you’re talking to a newbie lol.
Below is my current UUID when viewing fstab, are we talking both root and EFI or just EFI that needs to be changed?
So my original system when I made the timeshift backup was not encrypted, just ext4 with fat32 EFI, this new partition is encrypted ext4 luks with vfat. That may be why it’s not able to mount? I’d rather keep it encrypted but if it’s not possible to keep the recovery then I’ll repartition without the encryption if that’s the compromise needed.
Thank you for helping out, I’m learning a lot as I read other threads!
There are 12gb in that time shift restore file, probably 10k files or more and for me to figure out exactly where all my customizations are is a bit daunting. You’re right though that all that clutter is extra baggage.
You said the apps are in the / and the system tweaks are in/etc. Would this mean I should only have those two folders selected after another clean install? I’d still run into this issue where it doesn’t recognize the system format. Would I need to install ext4 unencrypted in order to recover these settings?
But I don’t think I quite said: “the apps are in the /”.
When you install a package, its folders and files gets installed in many different locations.
The applications settings, as said before are stored under the home directory.
Here is what I would do on a fresh install:
I install a package/application. I restore a previous settings for it. One at the time.
If there is something under /etc that I would need, then I would take it from the backup and put it in the right location.
It takes some time, but then you would have much more control in the process.
I get where you’re coming from, the etc folder is only 7mb so I don’t see why not just copy the whole thing? I have no idea where the layouts would be, the macros or startups etc.
I remember most of my apps being in the /home/.local folder which is obviously gone.
Just for knowledge are there any other folders in this backup list that are worth looking through for apps or otherwise? Didn’t see much poking around.
I guess you could do that if you wish. I wouldn’t do that personally. As mentioned already, if there is something specific I need, /etc/default/grub as an example, I would just grab it from the backup.
Also the binaries for the apps you install should be under /usr/bin but you will find mostly binaries from the base system installation. Also some apps could get installed under /opt.
At the end, here is my unsolicited opinion for a backup:
I wouldn’t care for snapshots of the system files/folders. I backup everything under home plus /etc and also perhaps a list of explicitly installed packages (pacman -Qqe). Using GNOME, I would perhaps backup my desktop settings ( dconf dump / > desktop-full-backup)
Sounds good, thank you! Where can I find my actual system settings and preferences in /etc? I honestly don’t know what most of these files and folders represent, so it’s not really possible for me to piece this folder together. I need to learn a lot more haha. Ultimately it’s not so bad going through all the settings again
I am not sure you will find them there.
They should be under ~/.config, ~/.local(/share) and perhaps other locations under home directory. Maybe KDE users could give a more detailed answer.
Read about Filesystem Hierarchy Standard. Scrape the Web.
