View previous topic :: View next topic |
Author |
Message |
ExecutorElassus Veteran
Joined: 11 Mar 2004 Posts: 1435 Location: Berlin, Germany
|
Posted: Mon Nov 29, 2021 10:03 am Post subject: |
|
|
Hi Neddy,
I had two ports on the old machine, because the plug on the one on the board wore out. However, here's my problem: the liveUSB names the one on the new board enp3s0, while the machine itself, apparently does not. Without being able to log in, I can't find out what it's named to add it with rc-update. Is there some config file I can use to force the name? It's not eth0, eth1, or enp3s0.
As for logging in, elogind is failing to start. Would that prevent me from being able to log in?
UPDATE: remerging elogind fixed my problem logging in, huzzah! So ow I can log in, and see that there is no network device created at all for the ethernet adapter, despite the specific kernel module being built in. I guess i'll go about fixing that.
UPDATE2: so, 'ls /sys/class/net/' shows two items, lo and "sit0". Would the system really have named the network adapter as "sit0"? That seems … weird. But ifconfig shows no other devices.
UPDATE3: haha, whoops, I forgot to mount /boot before doing 'make install', so the new kernel never got installed on real /boot. got internet working now.
Now I'm trying to get xorg, or rather lightdm, to start
Is there a way to tell the initramfs (I'm using the genkernel one) to run fsck on /, /usr, and /var before switching to "real" root?
Cheers,
EE |
|
Back to top |
|
|
Hu Moderator
Joined: 06 Mar 2007 Posts: 21631
|
Posted: Mon Nov 29, 2021 4:54 pm Post subject: |
|
|
I see that UPDATE3 solved this, but as a note: no, sit0 is not your network card. That is a tunneling device. You probably don't need it to exist. |
|
Back to top |
|
|
pingtoo l33t
Joined: 10 Sep 2021 Posts: 926 Location: Richmond Hill, Canada
|
Posted: Mon Nov 29, 2021 5:01 pm Post subject: |
|
|
ExecutorElassus wrote: | Is there a way to tell the initramfs (I'm using the genkernel one) to run fsck on /, /usr, and /var before switching to "real" root? |
Current genkernel initramfs does not support fsck on any thing. |
|
Back to top |
|
|
ExecutorElassus Veteran
Joined: 11 Mar 2004 Posts: 1435 Location: Berlin, Germany
|
Posted: Mon Nov 29, 2021 5:06 pm Post subject: |
|
|
update: after ploughing through illegal opcode warnings, I got xorg, then my browser, also internet and audio, up and running. Now I'm working my way through the end-point programs like libreoffice, texstudio, etc.
Question about genkernel's initramfs: if it doesn't do fsck on any premounted filesystems, is there a way to do that at any stage? fsck won't check filesystems that are already mounted, so whatever does it would have to do it at the early boot stage, yes?
I'm almost home…
Cheers,
EE |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54237 Location: 56N 3W
|
Posted: Mon Nov 29, 2021 5:35 pm Post subject: |
|
|
ExecutorElassus,
fsck will run on read only filesystem. That's how root is checked and why its mounted read only, then remounted rw later.
If root is rw, it passed its fsck.
You can remount /usr ro to run fsck, as long as there is nothing open for writing.
/var can only be checked when its not mounted.
Be aware that fsck cannot fix everything on a ro filesystem. It may warn you to run it again while its not mounted.
Warning: When fsck finds an inconsistency, it guesses what should be there. When it guesses incorrectly, it wakes things worse, not better and the is no undo.
Do make some backups and validate them before you start. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
pingtoo l33t
Joined: 10 Sep 2021 Posts: 926 Location: Richmond Hill, Canada
|
Posted: Mon Nov 29, 2021 7:25 pm Post subject: |
|
|
ExecutorElassus wrote: | Question about genkernel's initramfs: if it doesn't do fsck on any premounted filesystems, is there a way to do that at any stage? fsck won't check filesystems that are already mounted, so whatever does it would have to do it at the early boot stage, |
May I ask why you think fsck are needed?
If your shutdown are correctly executed your would not need fsck after restart. if you encounter a unexpected situation cause system unclean shutdown it would be better off by manually examine each file system at early boot sequence when init system decide fsck is required
To answer your question about early boot stage fsck, I don't know if it possible with systemd but if you are using OpenRC it is possible.
*** Warning, Doing follow might cause even bigger problem than you expect ***
above will cause /etc/init.d/fsck perform fsck with default options -A -p. The -p will cause for ext4 filesystem automatically fix filesystem where it can safely fixed without human intervention. -A will walk through the /etc/fstab file and try to check all filesystems in one run. |
|
Back to top |
|
|
ExecutorElassus Veteran
Joined: 11 Mar 2004 Posts: 1435 Location: Berlin, Germany
|
Posted: Wed Dec 01, 2021 9:56 am Post subject: |
|
|
well, previously I had problems with the machine randomly switching off when the MOSFETS overheated (the mobo was eight years old, which was why I started this whole ordeal in the first place). So I'd get a lot of orphaned inodes and similar when it rebooted. But if you and Neddy both caution that this isn't necessary/desired on /usr and /var on bootup, I can leave it.
So now the machine is running great, with the only problem being occasionally tripping over SIGILL errors in random programs and then remerging those ebuilds.
(on a side note, brave [the browser] keeps giving errors like this in dmesg:
Code: | [ 312.436830] brave[5631]: segfault at 0 ip 0000557147cc4519 sp 00007fff6cb00100 error 4 in brave[557143958000+9d84000]
[ 312.436839] Code: 00 48 8d 34 03 48 81 c6 00 10 00 00 0f b6 84 03 1e 10 00 00 83 e0 3f 49 89 c5 49 f7 dd 48 c1 e0 05 48 89 f7 48 29 c7 0f 18 0f <80> 3a 02 48 89 7d d0 0f 84 52 01 00 00 80 7a 02 00 0f 84 8a 01 00
|
This is obviously another illegal instruction, but is there any way to track down which specific dependency is causing it? It's not the brave binary itself, because that's brave-bin)
I think I'm almost done with this. Hoo boy, has this been a stressful week!
Cheers,
EE |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|