friendica.eskimo.com

Welcome to Friendica.Eskimo.Com

Home of Censorship Free Hosting

 Eskimo North Linux Shells, E-mail, Virtual Machines, Web Hosting.

E-mail, Web Hosting, Linux Shell Accounts terminal or full remote desktops.

Sign Up For A Free Trial Here

Please tell your friends about federated social media site that speaks several fediverse protocols thus serving as a hub uniting them, hubzilla.eskimo.com, also check out friendica.eskimo.com, federated macroblogging social media site, mastodon.eskimo.com a federated microblogging site, and yacy.eskimo.com an uncensored federated search engine. All Free!
evasync
evasync via Linux • lemmy (AP)

cant mount home on boot

Im giving a go fedora silverblue on a new laptop but Im unable to boot (and since im a linux noob the first thing i tried was installing it fresh again but that didnt resolve it).

its a single drive partitioned to ext4 and encrypted with luks (its basically the default config from the fedora installation)

any ideas for things to try?

This entry was edited (1 month ago)
Link to source
64 2 1
Max-P
Max-P • lemmy (AP)

The error says /home is a symlink, what if you ls -l /home?

Since this is an atomic distro, /home might be a symlink to /var/home.

Link to source
22 1 1
evasync
evasync • lemmy (AP)
yes it is a symlink to /var/home
Link to source
5 1
mvirts
mvirts • lemmy (AP)
So shouldn't you mount your home partition on /var/home instead?
Link to source
7 1
a2part2
a2part2 • lemmy (AP)
This feels like a winning strategy
Link to source
2 1
evasync
evasync • lemmy (AP)
the command returns my user dir and a lost+found dir
Link to source
1
Possibly linux
Possibly linux • lemmy (AP)
What's in lost+found
Link to source
1 1
lurch (he/him)
lurch (he/him) • lemmy (AP)
that actually sounds like it's already mounted
Link to source
3 1
Nanook
Nanook friendica
rm /home
mkdir /home
make /var/home a symlink to it.
Alternative, edit your /etc/fstab to mount on /var/home.
This entry was edited (1 month ago)
Link to source
Shoreline, WA, USA
7 1
evasync
evasync • lemmy (AP)
editing the /etc/fstab didnt work (I just changed the path but not sure if the uuid plays any part) but ill give the rm/mkdir part a go
Link to source
1 1
data1701d (He/Him)
data1701d (He/Him) • lemmy (AP)
Did you update your initramfs after? The new fstab doesn’t apply until you refresh that
Link to source
1 1
evasync
evasync • lemmy (AP)
No but I rebooted the system after the change.
do still need to update it regardless the reboot?
Link to source
1 1
data1701d (He/Him)
data1701d (He/Him) • lemmy (AP)

Edit: Probably try @nanook@friendica.eskimo.com's solution of systemctl daemon-reload first.

Yes. When booting, your system has an initial image that it boots off of before mounting file systems. You have to make sure the image reflects the updated fstab.

This entry was edited (1 month ago)
Link to source
1 1
Nanook
Nanook friendica
@data1701d (He/Him) @evasync You don't have to reboot to effect that, systemdctl daemon-reload will reload the /etc/fstab file.
Link to source
1 1 1
data1701d (He/Him)
data1701d (He/Him) • lemmy (AP)

You might be right. I was thinking of it in terms of a traditional distro, as I use vanilla Debian where my advice would apply and yours probably wouldn't.

From what I do know, though, I guess /etc would be part of the writable roots overlaid onto the immutable image, so it would make sense if the immutable image was sort of the initramfs and was read when root was mounted or something. Your command is probably the correct one for immutable systems.

Link to source
1
Nanook
Nanook friendica
@data1701d (He/Him) It will work fine with Debian Bookworm, not sure about older releases, I don't know at what point they switched to systemd controlling that but definitely does work in Bookworm. It should work in most other modern Debian or Ubuntu derived systems as well, but not older versions as systemd taking over this functionality is relatively recent.
Link to source
1
data1701d (He/Him)
data1701d (He/Him) • lemmy (AP)
On another note, for actually doing it, it looks like Fedora uses Dracut, so you just need to run sudo dracut -f.
Link to source
1
DNOS
DNOS • lemmy (AP)
Make sure the uuid matches the uuid of the home partition you want, you can list uuids with blkid I think, big noob here too I just spent my last week trying to figure out why it wouldnt mount my boot partition and the problem was the UUID...
Link to source
1
Telorand
Telorand • lemmy (AP)
Don't you have that backwards? This is an atomic distro, and you'd want to mkdir /var/home then symlink /home from that, no? Otherwise, you'll wind up with a home directory that is immutable.
Link to source
4 1 1
Nanook
Nanook friendica
@Telorand I am not familiar with that distro, I am however familiar with how mount works. As far as what is immutable and what is not, you can set with chattr +i file/directory or chattr -i file/directory.
Link to source
6 1 1
ravhall
ravhall • lemmy (AP)
That’s what she said.
Link to source
4 6 1
Possibly linux
Possibly linux • lemmy (AP)
Isn't the default filesystem btrfs? Why did you go with ext4
Link to source
4 2 1
Nanook
Nanook friendica
@Possibly linux @evasync I can't speak for them, but I've had btrfs blow up in ways I could not fix. I didn't just lose a file but the entire file system. I have NEVER had this happen in many years with ext4.
Link to source
5 1 1
psmgx
psmgx • lemmy (AP)
Lol same thing happened to me about 6 months ago. Overheating and/or a failing M2 and system corruption. btrfs got weird and troubleshooting only made it worse.
Link to source
secret300
secret300 • lemmy (AP)
I've only had this happen once and it turned out it was because my ram was shitting out errors that were saved to disk so it ended up not being btrfs's fault
Link to source
2 1
Possibly linux
Possibly linux • lemmy (AP)
Was that in the last 5 years? If it was btrfs is now far more stable. It has never blown up for me and it has in fact saved my data a few times.
Link to source
4 1
Nanook
Nanook friendica
@Possibly linux It was this year. Glad it's working for you. I'll stick with what works for me and has provided adequate performance for years.
Link to source
4 3 1
Nanook
Nanook friendica
There well may be hardware issues, but with ext4 it rarely corrupts the entire file system. You might end up with some data not flushed so you'll have some inodes that don't point to anything that you'll remove with fsck upon boot, but btrfs, I've had it corrupt and lose the entire file system. I've used ext2-through-ext4 for as long as they've existed and never lost a file system though back in the ext2 days I had to hand repair them a few times, but ext2 was sufficiently simple that that was not difficult, but within two weeks of turning up a btrfs file system it shit itself in ways I could not recover anything, the entire file system was lost. If I did not have backups, which of course I always do, I would have been completely fuxored. It is my opinion that btrfs and xfs, both of which have advantages, are also both not sufficiently stable for production use.
Link to source
10 4 1
dustyData
dustyData • lemmy (AP)
Did you reformat the disk before installing? I've seen similar fails when the disk is still encrypted. The installer can't get a hold of a previously encrypted disk. If there's no valuable data in the disk, load up a live distro run gparted and nuke the disk blank and pristine again, as gparted doesn't care about encryption. Then try the installer again.
Link to source
1 1
Nanook
Nanook friendica
@dustyData @evasync When I install, I generally prepare the partitions ahead of time with gparted, whether or not I create an entirely new partition table depends upon whether it is the only OS on the disk or there are multiple. I'm not using any encrypted file systems, I need the machines to be able to boot without my being present to type in a password or pass phrase. So that is not an issue.
This entry was edited (1 month ago)
Link to source
Shoreline, WA, USA
1 1
evasync
evasync • lemmy (AP)
no, I just removed them with the live cd and repartitioned it (Im assuming its doing the same thing under the hood?)
Link to source
1
dustyData
dustyData • lemmy (AP)
You should let the installer do the partitioning. Silverblue and immutable systems are nitpicky about it. Specially if luks is involved. The whole point is that you shouldn't meddle with the system at a low level at all.
Link to source
1
Nanook
Nanook friendica
@dustyData @evasync I've been working with Linux since 1992, I have a better idea of how I want my disks laid out then an installer script.
Link to source
1 1
dustyData
dustyData • lemmy (AP)
Sorry, I was not replying to you (not an insult). I assume you are interacting from Mastodon from the format of the comment, and getting pinged on replies to other comments (?). I mean, you do you, absolutely not going to diss people who want absolute control over their system. But immutable distros are fundamentally an entirely philosophically different approach from how traditional Linux distros have been packaged and managed in the past. That said, I didn't make the installers, I'm just reporting what has been my recent experience toying with immutable distros. The whole point is to automate as much as possible of the deployment and management of an OS, and do the least amount of tedious manual troubleshooting. If you don't like that, all the other distros are still there, they haven't gone anywhere. The current recommendation for Fedora Atomic based distros is to use specialized tools like Universal Blue that allows the user absolute freedom to deterministically configure a Fedora install that results in an immutable OS. And the installer is actually pretty flexible to let you choose how you want the disks laid out. But, the idea is that you should let the installer do its job, that's for what it was made. If you want to do everything by hand just use Arch, that's what Arch is for.

Universal Blue - Powered by the future, delivered today

Universal Blue is a diverse set of images using Fedora Atomic's OCI support as a delivery mechanism. That's nerdspeak for the ultimate Linux client!
universal-blue.org
This entry was edited (1 month ago)
Link to source
1
Urist
Urist • lemmy (AP)
NixOS and ext4 user here with no problems. Care to elaborate?
Link to source
4 1
dhhyfddehhfyy4673
dhhyfddehhfyy4673 • mbin (AP)
Seems like this can be prevented from reaching that point by properly deleting old generations regularly though right?
Link to source
1 1
mvirts
mvirts • lemmy (AP)

20 or 30 generations 😹

I have space for 1 😭

Edit: you've got me worried now, is the behavior you're referring to normal running out of inodes behavior or some sort of bug? Is this specific to ext4 or does it also affect btrfs nix stores?

I've run across the information that ext4 can be created with extra inodes but cannot add inodes to an existing filesystem.

This entry was edited (1 month ago)
Link to source
1
mvirts
mvirts • lemmy (AP)
Thank you for providing the quote!!
Link to source
1

This website uses cookies. If you continue browsing this website, you agree to the usage of cookies.