Arch Linux

fhtagn

Member

From: inside your head.

Registered: 2010-12-21

Posts: 16

[SOLVED] root@(none)

Hello again.

After upgrading my headless server with 'pacman -Syu', I figure I could reboot it. The last update had been 1 day ago and I didn't reboot it, so I did so and waited..
After a while the server starts to sound like it is rebooting a lot, and ssh'ing there does not work, nor does ping. I hook up a monitor and keyboard, reboot it again and get...this...

Just great.
sda1 is /boot, ext2
sda2 is swap, swap
sda3 is /, ext4
sda4 is /home, ext4

After this, and before it reboots again, I try ctrl+C and it says it will take me to runlevel 3. Great. However, I am greeted with

(none) login: 

(none) ?
Logging in as a regular user I get

No directory, logging in with HOME=/

and my bash prompt reads:

<myuser>@(nome):

hmm.. trying to 'sudo su' to root and gives me (but I still manage to become root)

sudo: Can't open /var/lib/sudo/<myuser>/tty4: Read-only file system

(i was on tty4)
By typing

df -h

I find that swap and /home were not mounted.
Trying to mount swap I get

mount: mount point swap does not exist

/home mounts fine.

Finally, dmesg gives me: http://pastebin.com/6tRbA17g

What happened? What can I do?

I will provide as much info as possible, but please guide me a bit. I can't find a way to identify this problem and googling is not helping.
I figure dmesg would keep the earliest report of any error it might have occured, but the first 'error' I can find is the line

Not activating Mandatory Access Control now since /sbin/tomoyo-init doesn't exist.

I don't know what to do.

Last edited by fhtagn (2011-01-28 21:58:14)

ndowens04

Member

Registered: 2009-05-07

Posts: 25

Re: [SOLVED] root@(none)

If you are able to login into the machine, you could use a livecd to recover the system. First thing I would look at is to make sure that your fstab is correct.

---

ArchLinux + ZFS

fhtagn

Member

From: inside your head.

Registered: 2010-12-21

Posts: 16

Re: [SOLVED] root@(none)

Thank you! Problem solved.

I booted from a Arch Live usb

dd if=archlinux-2010.05-core-i686.iso of=/dev/sdb

and ran the tests for memory and CPU, all came out clean.
Afterwards, I ran fsck on all partitions, none gave any error. I checked fstab and re-checked the <pass> field on all the partitions, cause I had been messing with it after adding two external hard drives.

As of now, this is how it is

# /etc/fstab:
# <file system>    <dir>    <type>  <options>    <dump> <pass>

devpts     /dev/pts   devpts  defaults          0  0
shm       /dev/shm   tmpfs   nodev,nosuid  0  0

#swap
UUID=<theuuid>  swap   swap   defaults                0  0
#home
UUID=<theuuid>  /home   ext4     defaults,usrquota  0  2
#boot
UUID=<theuuid>  /boot   ext2     defaults                0  2
#/
UUID=<theuuid>  /   ext4     defaults                0  1

#1T keep
UUID=<theuuid>  /mnt/exthd/1T/keep  ntfs-3g gid=users,umask=0022  0  0
#1T loot
UUID=<theuuid>  /mnt/exthd/1T/loot    ntfs-3g  gid=users,umask=0022  0  0
#1T fatty-rand
UUID=<theuuid>  /mnt/exthd/1T/rand    vfat      gid=users,umask=0022  0  0
#1.5T
UUID=<theuuid>  /mnt/exthd/1.5T       ntfs-3g  gid=users,umask=0022  0  0

Problem was I misread <pass> meaning (wiki).
I had the external harddrive partitions (and the external hdd is not always connected to the server) with a pass value of 2. I figured that would be 'ok' because that is a lower priority than '/'. However, it seems that a pass value other than 0 makes Arch try to check for the partitions. And he doesn't seem to like not having ALL the partitions to check at boot. Maybe that is obvious, but I had no way of knowing it.

If there is anything you'd like to suggest for me to change and improve my setup, I would like to hear it.

Adding [SOLVED] tag. Thank you for your time.

Last edited by fhtagn (2011-01-28 22:01:35)