After the recent update, system is failing to boot and gets stuck at the following message:
system-fsck: info: /dev/sda13: clean, 462221/2876784 files,
mount-fstab: info: starts...
Even when i tried to go into single mode I am getting the same message and not proceeding further.
Can anyone advise what might be going wrong and the solution to fix this?
Thanks.
Boot into live ISO, check out logs and see if the partitions are ok.
If you are positive you didn't change any of the settings, /var/log/pacman.log is one of the logs to check.
https://forum.artixlinux.org/index.php/topic,1923.0.html
What are there in your /etc/fstab?
Following are the fstab entries:
UUID=fa4e74f2-8aab-4307-b0df-8f06152b1959 / ext4 defaults,noatime 0 1
/dev/sda5 none swap defaults 0 0
This file has not been updated since Aug 25.
-rw-r--r-- 1 root root 473 Aug 25 19:41
I boot into this partition mainly and have been restarting every day and never faced this issue before. Only after applying the latest updates yesterday i started facing this issue.
Simple like that. :)
Try to boot the system up using a live iso and check that partition with fsck. If everything is good, I would try to downgrade suite66-script to the previous version to see if it helps. Of coz, check log files following suggestions from
@strajder
I see there is another thread on the same issue though with a different error message (https://forum.artixlinux.org/index.php/topic,3065.0.html), possibly something to do with the s66 startup scripts. Any idea how i can downgrade the suite66 scripts?
Thanks.
In /etc/66/init.conf (I think or similar) change verbosity from 0 to 6 and see if you can screenshot or something of the output. Could be something other than fstab
If the error message is different, there's no reason to believe the issue is the same.
mount-fstab is a part of suite66-scripts (https://gitea.artixlinux.org/packagesS/suite66-scripts), and here's what the current version does (https://gitea.artixlinux.org/artix/66-scripts/src/branch/master/module/boot@/service/mount/mount-fstab):
[main]
@type = oneshot
@version = @vers@
@description = "Mount partition from /etc/fstab file"
@user = ( root )
@depends =( system-fsck )
@options = ( !log )
[start]
@execute =
(
if { 66-yeller -cdp mount-fstab -1 /dev/console starts... }
if -nt {
mount -a -O noauto,no_netdev
}
66-yeller -Fcdp mount-fstab -1 /dev/console failed!
)
Essentially,
mount -a -O noauto,no_netdev
which just mounts all the file systems from /etc/fstab, except those that have the
auto or
_netdev option.
I'd try to enter that command from a live ISO chroot and see if it errors out.
What i mean was the other thread also talked about system not booting and so did mine. I will try out your suggestions and see how it goes.
Thanks.
Now that I'm looking at that command, I wonder what was the reasoning behind changing
mount -a (https://gitea.artixlinux.org/artix/66-scripts/src/commit/b0522d99db6693c584873b562cc3653942953653/module/boot@/service/mount/mount-fstab#L15) to
mount -a -O noauto,no_netdev, and shouldn't that read:
mount -a -O auto,no_netdev
or maybe even
mount -a -O auto,_netdev
I tried the different mount options in live iso and they all executed fine, there were no errors.
I downgraded the suite66-scripts and now I'm able to boot into the system (this post is written from the system that didn't boot earlier). Have set the suite66-scripts to be ignored until it is fixed. The issue is with 1.7.1.6-1.
warning: suite66-scripts: ignoring package upgrade (1.7.1.4-2 => 1.7.1.6-1)
Hmm, mount-fstab seems the same in 1.7.1.4:
https://gitea.artixlinux.org/artix/66-scripts/src/branch/1.7.1.4/module/boot@/service/mount/mount-fstab
Well, the debugging should continue towards bisecting this further between 1.7.1.4 and master. It would be beneficial if you provided /var/log/everything.log (use https://paste.artixlinux.org), /etc/66/init.conf and the contents of
$ su
# 66-env -t boot boot@system
also as ndowens suggested maybe trying to set VERBOSITY=6 in /etc/66/init.conf (a good idea in general when debugging issues in 66) with the latest version and providing the output during boot.
https://paste.artixlinux.org/view/d4acb9cf#NwDCujSLwbZLYdvKmHUVwWU9K5X4B8tp
https://paste.artixlinux.org/8c5a4ae3
https://paste.artixlinux.org/e6fb47ca
I am hoping the above pastbin works.
Can you please test with the latest suite66-scripts and
VERBOSITY=6? If you do that, please clean the everything.log with
$ su
# cat /dev/null > /var/log/everything.log
before reboot, so that only the output from the last boot is captured.
Thanks in advance.
https://paste.artixlinux.org/aa4ffe1a
Everything log with the latest suite 66 scripts and verbosity 6.
Can your system boot now with the latest suite66-scripts?
No, doesn't work. Just upgraded for sharing the logs and then downgraded back to earlier version.
Can you provide the /run/66/log/0/current file after a failed start from the ISO chroot?
Actually providing that file on a successful start as well would be helpful. Presumably a service deadlocks. mount-fstab seems like the culprit but on paper it shouldn't be possible. Anyways a compare/contrast between a successful start and a bad start would help.
https://paste.artixlinux.org/352e8673 - Current bad
https://paste.artixlinux.org/d925e71e - Current good
Not sure if those two files make sense, my machine never booted beyond the "mount-fstab: info: starts..." message but the log seems to say it has proceeded further. Anyway hope it helps.
That is without VERBOSITY=6 in /etc/66/init.conf, can you give the logs with that set?
https://paste.artixlinux.org/b6e310e9 - Current bad
https://paste.artixlinux.org/775892be - Current good
What does your tree look like: 66-intree
Thanks for all the logging so far. So far, I have noticed that mount-swap does not appear to start for you on the new upgrade for some reason. If that oneshot really never starts, that would explain why you hang during boot at least. Verbosity 6 is apparently far too much logging information. Instead, could you set it to 2 and redo the logs one more time for both the bad and good case? Thanks!
https://paste.artixlinux.org/62e29710 - Current bad
https://paste.artixlinux.org/d005e4cf - Current good
VERBOSITY=2
LIVE=/run/66
PATH=/usr/bin:/usr/sbin:/bin:/sbin:/usr/local/bin
TREE=boot
RCINIT=/etc/66/rc.init
RCSHUTDOWN=/etc/66/rc.shutdown
RCSHUTDOWNFINAL=/etc/66/rc.shutdown.final
UMASK=0022
RESCAN=0
CONTAINER=0
RCINIT_CONTAINER=/etc/66/rc.init.container
CATCHLOG=1
What does your tree look like: 66-intree
Name : audio
Initialized : no
Enabled : yes
Starts after : None
Current : yes
Symlinks : svc->source db->backup
Contents : pipewired-log pipewired
That should not be all your trees; there should be boot, and default
@mubh To add the default tree, you can run
# 66-tree -ncE default
# 66-env -t boot boot@system
# 66-enable -t boot -F boot@system
Then populate the default tree. You need at least these services:
acpid dhcpcd dbus logind syslog-ng
plus your network manager, if any, and any other services you are using. You can add just those listed services using the command:
# 66-enable -t default acpid dhcpcd dbus logind syslog-ng
# 66-all -t default up
If you need to add pipewired to the default tree, the commands are:
# 66-enable -t default pipewired
# 66-start -t default pipewired
Update: Information on how to create "boot" and "default" trees from scratch is added to the wiki entry on suite66 (https://wiki.artixlinux.org/Main/Suite66) (TL;DR section at the top).
@strajder - Thanks for the commands on setting up the tree and the services.
For now, I'm not changing anything on my system. I saw there was one more update 1.7.1.7-1 on the suite66-scripts, i upgraded to that version however i was again faced with a non booting system, so downgraded back. I will have to keep ignoring suite66 updates until i find a newer version that works.
We can close this thread if there is no other solution that is likely to fix the issue that i'm having.
Thanks for all your help.
Did you try the commands to recreate the trees on the latest version of suite66-scripts? The latest update was related to properly setting virtual console font and keymap, and should fix one of the issues leaving messages in your boot log (kbd_mode -u).