Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
udevd problem (solved)
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
Kysen
Tux's lil' helper
Tux's lil' helper


Joined: 26 Nov 2004
Posts: 128

PostPosted: Sun Jul 02, 2006 2:50 am    Post subject: udevd problem (solved) Reply with quote

This is the first time i noticed udevd starting up at the boot process and its giving me some problems here. Im so observant :roll:

It first started when i tried to upgrade to the 2.6.17 kernel, when it first came up i got this error

Code:

udevd[1226]: init_udev_socket: error getting socket: Address family not supported by protocol
udevd[1226]: main: error initializing udevd socket: illegal seek


I then recompiled my 2.6.16-r3 kernel to remove old hardware and now im getting the same error with that kernel.

I have no clue where to start with this, maybe a kernel error some how that i missed? Or maybe some udev links? Iono


Last edited by Kysen on Fri Jul 21, 2006 5:14 pm; edited 1 time in total
Back to top
View user's profile Send private message
frozenJim
Guru
Guru


Joined: 18 Jun 2004
Posts: 341
Location: Montreal, Quebec, Canada

PostPosted: Thu Jul 06, 2006 2:26 pm    Post subject: Reply with quote

Similar problem here. I upgraded my kernel to 2.16 and it was fine. I rebooted a few times without error. Then I made a minor change to my network settings in order to get wifi up (enabled generic IEEE under Networking and enabled Intel Pro 2200BG under Devices->Network Devices). I do not think that my wireless changes in the kernel can have caused this:

Now I get the "unsupported protocol" error. To make things worse, it is preventing me from booting so I cannot simply experiment with my kernel config.

Can anyone tell me/us what is causing this? I can't do anything because even though /hda2 was correctly initialized (reiserfs) and shows success in dmesg I cannot load my root directory because /dev/hda2 cannot be mounted. (failed to open /dev/hda2, no such file or directory).

It appears that hda2 is fine, but it's like there is no entry in /dev for it.
Code:
Gentoo Linux; http://www.gentoo.org/
 Copyright 1999-2006 Gentoo Foundation; Distributed under the GPLv2

* Mounting proc at /proc ... [ok]
* Mounting sysfs at /sys ... [ok]
* Mounting /dev for udev ... [ok]
* Seeding /dev with needed nodes ... [ok]
* Setting up proper hotplug agent ... [ok]
*   Using netlink for hotplug events... [ok]
* Starting udevd ...
udevd[1333]: init_udevd_socket: error getting socket: Address family not supported by protocol
udevd[1333]: main: error initializing udevd socket: Illegal seek  [!!]
* Populating /dev with existing devices through uevents ... [ok]
* Letting udev process events ... [ok]
* Finalizing udev configuration ... [ok]
* Mounting devpts at /dev/pts ... [ok]
* Activating (possible) swap ... [ok]
* Checking root filesystem ...
Failed to open the device '/dev/hda2/: No such file or directory

* Filesystem couldn't be fixed :( [!!]
Give root password for maintenance
(or type Control-D to continue):


Dmesg:
Code:

ReiserFS: hda2: found reiserfs format "3.6" with standard journal
ReiserFS: hda2: using ordered data mode
ReiserFS: hda2: journal params: device hda2, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
ReiserFS: hda2: checking transaction log (hda2)
ReiserFS: hda2: Using r5 hash to sort names
VFS: Mounted root (reiserfs filesystem) readonly.
Freeing unused kernel memory: 168k freed
usb 4-2: USB disconnect, address 2


/dev is almost completely empty:
Code:
# ls /dev
console core fd null pts shm stderr stdin stdout zero


When I try to edit my kernel, I am unable:
Code:
# make menuconfig
make: Warning: File `.vmlinux.cmd' has modificatin time 1e+04 s in the future
make[1]: Warning File `.config' has modification time 1e+04 s in the future
make[1]: warning: Clock skew detected.  Your build may be incomplete.
make[1]: Warning: File `scripts/kconfig/.conf.cmd' has modification time 3e+03 s in the future
/usr/src/linus-2.6.16-gentoo-r11/scripts/kconfig/lxdialog/check-lxdialog.sh: line 44:
  1481 Done
    echo "main() []"
  1482 Aborted
    | $cc -xc - -o $tmp 2>/dev/null
 *** Unable to find the ncurses libraries.
 *** make menuconfig require the ncurses libraries
 ***
 *** Install ncurses (ncurses-devel) and try again
 ***
make[2]: *** [scripts/kconfig/lxdialog/dochecklxdialog] Error 1
make[1]: *** [menuconfig] Error 2
make: *** [menuconfig] Error 2


AND... for the first time ever, I deleted my old /boot/bzImage accidentally before backing it up so I don't even have the OLD kernel to boot from!

What's a boy to do?
_________________
Who controls the past, controls the future. Who controls the present, controls the past.
Back to top
View user's profile Send private message
idella4
Veteran
Veteran


Joined: 09 Jun 2006
Posts: 1587
Location: Australia, Perth

PostPosted: Thu Jul 06, 2006 3:35 pm    Post subject: Reply with quote

these might be well off the mark but what the heck

Kysen,
is it possible you have recompiled kernel that was allright 2.6.16 with the config file from troublesome kernel-2.6.17?
an outside chance but admittedly it doesn't make much sense at face value.
The error it gives looks like it comes from a selection in .config not being there but should.

frozenJim,
here's a possibility.
I've seen that appear in the last few days. I had been swapping hard drives. It turned out to be a 'loose' connection of the cable connector into the hard drive.
It didn't appear or feel loose, but with simply re-fitting them, the hard drive re-appeared.
_________________
idella4@aus
Back to top
View user's profile Send private message
frozenJim
Guru
Guru


Joined: 18 Jun 2004
Posts: 341
Location: Montreal, Quebec, Canada

PostPosted: Thu Jul 06, 2006 3:47 pm    Post subject: Reply with quote

idealla4 wrote:
It turned out to be a 'loose' connection of the cable connector into the hard drive.

I considered that, which is why I checked dmesg to confirm that /hda2 was being mounted properly. It's there as far as ReiserFS can tell - and that's probably a pretty solid indicator. :wink:

Unless there is some way that I'm misreading the ReiserFS info... But I doubt that my laptop has a loose HD connection on the day that I upgrade my kernel.

Also, the fact that NONE of my /dev devices have been created is an indicator that it's a more systemic problem.
_________________
Who controls the past, controls the future. Who controls the present, controls the past.
Back to top
View user's profile Send private message
idella4
Veteran
Veteran


Joined: 09 Jun 2006
Posts: 1587
Location: Australia, Perth

PostPosted: Thu Jul 06, 2006 4:34 pm    Post subject: Reply with quote

FrozenJim
from what are you observing your laptop's hardrive? as in is it the same as you are using to post these queries.
The looseness is probably not the cause as you say, though the loosenes refers to now and not to the time you installed.
The main point is do you have concurrent access to the laptop's harddrive from the system you are using?
_________________
idella4@aus
Back to top
View user's profile Send private message
frozenJim
Guru
Guru


Joined: 18 Jun 2004
Posts: 341
Location: Montreal, Quebec, Canada

PostPosted: Thu Jul 06, 2006 4:42 pm    Post subject: Reply with quote

No, I have no access to THAT part of the laptop at all. Can't ssh in, can't login.

Fortunately, I dual-boot XP so that's where I am right now. If I can find out what needs to be fixed, then I will LiveCD in and do the fixing. But first I need to have at least a guess as to what might be wrong.
_________________
Who controls the past, controls the future. Who controls the present, controls the past.
Back to top
View user's profile Send private message
idella4
Veteran
Veteran


Joined: 09 Jun 2006
Posts: 1587
Location: Australia, Perth

PostPosted: Thu Jul 06, 2006 5:27 pm    Post subject: Reply with quote

FrozenJim

You are working from XP which is in fact on the same hard drive, yes?

Seeing you are in Canada, the locall corner computer shop may have just the thing for you.
I got it from a computer magazine a couple of years ago, a superceded freebie.

extfs anywhere

is the name of the program. It is a windows program which ofcourse equips windows to contact linus partitions.
Just the thing. I can't remember the company, Paragon I think. Southerners like me don't have it available in the country.

Otherwise there is Slax or knoppix or Simply Mepis et al. giving access from the cd.
That concurrent access is what I have come to rely upon. Feel crippled without it.
I'd be thinking of repartitioning, making a third partition big enough to get a clone of damaged gentoo onto,
put into it a complete new gentoo kernel complete with .config and make a replacement kernel for the lost one.,
and /or
get a stage3.tar.gz onto the repairer partition 3,
open it up
it can supply a source of spare /dev*

Also, this sudden mass disappearance brings to mind the possibility of hardware corruption.
I have not had it much myself but a recent post from one of your fellow gurus made mention of it. suggested to do a memtest.

good luch
_________________
idella4@aus
Back to top
View user's profile Send private message
Aciel
Tux's lil' helper
Tux's lil' helper


Joined: 07 Jan 2004
Posts: 91

PostPosted: Fri Jul 07, 2006 12:49 am    Post subject: Reply with quote

What does udevd do, anyway? It's not running on my laptop which is a fresh install, and I don't remember reading about it. But I've got it on my desktop, who knows why.
Back to top
View user's profile Send private message
frozenJim
Guru
Guru


Joined: 18 Jun 2004
Posts: 341
Location: Montreal, Quebec, Canada

PostPosted: Fri Jul 07, 2006 10:46 am    Post subject: Reply with quote

en.wikipedia.org/wiki/Udev wrote:
udev is a file system replacement for DevFS on the Linux 2.6 kernel series. Its primary function is managing device nodes in /dev.


I'm thinking that my problem is just a switch I forgot to set in my kernel. I found an old kernel (whew!) and can boot just fine if I use it. It's just my new kernel that causes the problems. So I'm not totally screwed, I booted from the old kernel and emerged some packages and now I am able to make my kernel at least. I'll find it by trial and error eventually.

This is a good sign that there's no hardware problems. It's all my fault - bad kernel config.

I don't know for sure if udev is the problem - but it sure looks like the most likely culprit.
_________________
Who controls the past, controls the future. Who controls the present, controls the past.
Back to top
View user's profile Send private message
Stormbreaker
n00b
n00b


Joined: 30 Oct 2003
Posts: 49

PostPosted: Fri Jul 07, 2006 7:56 pm    Post subject: Reply with quote

You did forget to compile support for UNIX DOMAIN SOCKETS under NETWORK DEVICES in your kernel.
Back to top
View user's profile Send private message
frozenJim
Guru
Guru


Joined: 18 Jun 2004
Posts: 341
Location: Montreal, Quebec, Canada

PostPosted: Sat Jul 08, 2006 12:49 pm    Post subject: Reply with quote

Stormbreaker wrote:
You did forget to compile support for UNIX DOMAIN SOCKETS under NETWORK DEVICES in your kernel.
You must be KIDDING!! sigh... who knew? This is the expected price we must pay for "trimming" the kernel of features we know nothing about.

Yep, that was it - you hit the nail right on the head. Thank you SO MUCH! :D
_________________
Who controls the past, controls the future. Who controls the present, controls the past.
Back to top
View user's profile Send private message
Bigun
Veteran
Veteran


Joined: 21 Sep 2003
Posts: 1805

PostPosted: Thu Jul 13, 2006 1:41 am    Post subject: Reply with quote

It can't be modular either... has to be built in...

pwnt!

*bookmarked*
_________________
Sadhu Sundar Singh - responding to Darwinian Evolution - "...but I am more interested in Divine Selection and survival of the unfit"
Back to top
View user's profile Send private message
random303
n00b
n00b


Joined: 07 Mar 2006
Posts: 7
Location: Belgium

PostPosted: Wed Apr 02, 2008 9:03 pm    Post subject: Reply with quote

Stormbreaker wrote:
You did forget to compile support for UNIX DOMAIN SOCKETS under NETWORK DEVICES in your kernel.


Thank you so much !!! (been looking 2 days to solve this problem)
I'm new (again) to gentoo and I just thought with my fresh install, well let's compile everything for the network after my gentoo boots. Was I wrong. Didn't know it was needed for udevd...
Now I'm for the first time (again) on the gentoo comand prompt ;-)
Back to top
View user's profile Send private message
FilimoniC
n00b
n00b


Joined: 13 Apr 2008
Posts: 6
Location: Russia

PostPosted: Sun Apr 13, 2008 5:10 am    Post subject: Reply with quote

Stormbreaker wrote:
You did forget to compile support for UNIX DOMAIN SOCKETS under NETWORK DEVICES in your kernel.

Thank you!
Back to top
View user's profile Send private message
rockdw
n00b
n00b


Joined: 19 Sep 2003
Posts: 18

PostPosted: Fri Apr 18, 2008 9:45 pm    Post subject: Unix Domain Sockets wasn't enough Reply with quote

I needed to also make sure that Packet socket was selected.

I guess that's what I get for not building a fresh system for a long time. The process keeps changing.

Running like a champ now, though :-)
Back to top
View user's profile Send private message
durand101
n00b
n00b


Joined: 11 Nov 2007
Posts: 24
Location: Nottingham, UK

PostPosted: Sat May 10, 2008 9:15 pm    Post subject: Reply with quote

Stormbreaker wrote:
You did forget to compile support for UNIX DOMAIN SOCKETS under NETWORK DEVICES in your kernel.


THANK YOU SOOOOOOOOOOOOOOO MUCH!!! I've been so confused by this problem!
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Page 1 of 1

 
Jump to:  
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