Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
udev update killed my system - HELP please :-(
View unanswered posts
View posts from last 24 hours

Goto page 1, 2  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
MitsukiP
n00b
n00b


Joined: 19 Mar 2012
Posts: 5
Location: Nürnberg

PostPosted: Mon Mar 19, 2012 7:34 am    Post subject: udev update killed my system - HELP please :-( Reply with quote

Hello,

i have a very big problem.

I made a update hours ago and there was this new udev-update (mask was removed). I read the info message about problems with /usr/ on a different parition and because of that i created a initrd-file with dracut. I don't use genkernel because i compiled all my kernel by myself.

But i can't boot with the initrd file, i only see this error:
Code:
dracut Warning: Unable to process initqueue"
dracut Warning: Unable to process initqueue"
dracut Warning "/dev/md3" does not exist
dracut Warning "/dev/md3" does not exist
Dropping to debug shell.


Then i tried it with my standard kernel, now i can see many error messages and missing files. Then i only see my root Paritition with many empty mountpoints. Funny, and i can't access any parition because mdraid/lvs is not working :-(

How can I fix this? I'm using Gentoo for many years now, but that's the first time i'm stuck. I have no idea what i should do now.

I can't emerge a old udev-version, i can't fix the initrd-file, i can't to anything :-(

I already read https://forums.gentoo.org/viewtopic-t-901206-highlight-udev.html but didn't really help. There was no line with "initramfs=" in my config, i added a initramfs="no" but still the same problem. udev was already in the sysinit runlevel.

Btw, I have /var/ /usr/ /home/ ... on different partitions.

Thanks

Stuck by NeddySeagoon
Back to top
View user's profile Send private message
BillWho
Veteran
Veteran


Joined: 03 Mar 2012
Posts: 1600
Location: US

PostPosted: Mon Mar 19, 2012 1:12 pm    Post subject: Reply with quote

MitsukiP,

My system fried last evening. :cry:

I'm not sure if this will work for you, but what I did as a temporary fix is mask =x11-drivers/xf86-input-evdev-2.7.0 and =x11-base/xorg-server-1.12.0 then rebuild x11-drivers
Code:
emerge -av $(qlist -IC x11-drivers)

Hope this helps you :wink:
Back to top
View user's profile Send private message
MitsukiP
n00b
n00b


Joined: 19 Mar 2012
Posts: 5
Location: Nürnberg

PostPosted: Mon Mar 19, 2012 1:50 pm    Post subject: Reply with quote

I think my problem is bigger, i can't access my hard disks atm and not sure why, because my system don't detect the raid and the lvm :-(
Back to top
View user's profile Send private message
MitsukiP
n00b
n00b


Joined: 19 Mar 2012
Posts: 5
Location: Nürnberg

PostPosted: Mon Mar 19, 2012 4:52 pm    Post subject: Reply with quote

Hello again,

i chrooted my system, masked >=sys-fs/udev-181, emerged it and tried to restart my system.

First i just saw a kernel panic, because my /mnt/md3 is now a /mnt/md122 and my /mnt/md3 is now /mnt/md127 (why?)

I changed the entries in /etc/fstab and /boot/grub/grub.conf and the system is booting. But now with many new errors.
Code:
Starting udev ...
Populating /dev with existing devices through uevents ...
Waiting for uvents to be processed ...
udev[xxx]: failed to execute '/sbin/modprobe' ....


I see dozens of missing modprobe-messages, my network is not working, no env-vars, no modules loaded...

But why? After the boot /sbin/modprobe is there and working.

Now my system is broke and i have no glue what i should do now :-(
I just want my old pre-emerge system back, everything was working fine for years, and now it's broke...

Thanks
Back to top
View user's profile Send private message
ppurka
Advocate
Advocate


Joined: 26 Dec 2004
Posts: 3256

PostPosted: Mon Mar 19, 2012 5:28 pm    Post subject: Reply with quote

Just some questions for you to think about: :)

1. Did you update your configs in /etc after the downgrade?

2. Did you try to get your system to the state before this latest upgrade? For this look at the output of genlop, for example:
Code:
genlop -l -u --date 2 days ago
Then revert all the packages to their versions before the breakage, and then update the configs in your /etc.
_________________
emerge --quiet redefined | E17 vids: I, II | Now using kde5 | e is unstable :-/
Back to top
View user's profile Send private message
BillWho
Veteran
Veteran


Joined: 03 Mar 2012
Posts: 1600
Location: US

PostPosted: Mon Mar 19, 2012 5:58 pm    Post subject: Reply with quote

MitsukiP,

Don't know about the mdraid/lvs mounting but I had a message after the update about DEVTMPFS_MOUNT having to be set! Could that be it?

Good luck :wink:
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54028
Location: 56N 3W

PostPosted: Mon Mar 19, 2012 7:11 pm    Post subject: Reply with quote

This problem arisies now because udev tries to run binaries and scripts located in /usr before /usr is mounted.
/var is a lesser problem, udev tries to restore your mixer settings from /var before its mounted too.

Until udev-181, there was some code that tolerated this and tried to redo items that failed the first time.
The immediate solution is to mask udev-181 until you have time to sort out the mess.

When you miss that opportunity the fallback is to boot a livedCD/USB/DVD ... chroot into your broken install and downgrade udev to a version less than 181, so you have the fallback code once more. This is not a long term soloution. Its just a get you going so you can migrate in your own time.

I've been there and been bitten too. Now all of my raid minor numbers have been scrambeld by booting with SystemRescueCD too.
If you want to make an initrd by hand, looks sane, it does not include raid and lvm2 supprt but if you have an initrd for that already, it looks like its easy to include.

Hint raid assembly and lvm starting goes in the init script at
Code:
# ====================== start doing stuff

As a hand made initrd puts everything under your control and allows you to understand whats happening, its the least worst way ahead.

I use ~ARCH everywhere, because of that, I have buildpkg set in my FEATURES so for me the downgrade was mask udev-181
Code:
emerge -K udev
and fix grub.conf after I found out the hard way the my raid minor numbers had been trashed.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Thistled
Guru
Guru


Joined: 06 Jan 2011
Posts: 572
Location: Scotland

PostPosted: Tue Mar 20, 2012 8:01 pm    Post subject: Reply with quote

Now udev-181 has disappeared from the tree, and udev-182 is hard masked. :o
Are the Gentoo peeps sensing lots of users are about to break their systems? :roll:
I have masked udev-181 and above until some sort of documentation or wiki / walkthrough appears on the internet.
I would advise other Gentoo users (whom like myself) who are not up to speed with the whole concept of all things Gentoo, to do the same.
Had a little dabble today trying to update to udev-181, and experienced warnings about removing "module-init-tools" and lots of file collisions with regards to udev rules.
So not going anywhere on this system until it's all settled down. I don't have /usr on a separate partition either, so I presumed the upgrade would be quite inncocent. But no go for me. :?

Edit: So the on-line package database says [M] but portage says it is [~], but, nonetheless, udev-181 has gone.
_________________
Whatever you do, do it properly!
Back to top
View user's profile Send private message
Naib
Watchman
Watchman


Joined: 21 May 2004
Posts: 6050
Location: Removed by Neddy

PostPosted: Tue Mar 20, 2012 9:42 pm    Post subject: Reply with quote

yup I just got bit by this.
I just modprobe my nic driver, brought up eth0 and emerge the older version without the need to boot of a liveCD


I also do not have /usr partition so was VERY surprised when this bit me
_________________
Quote:
Removed by Chiitoo
Back to top
View user's profile Send private message
Thistled
Guru
Guru


Joined: 06 Jan 2011
Posts: 572
Location: Scotland

PostPosted: Wed Mar 21, 2012 12:03 pm    Post subject: Reply with quote

I didn't get past the upgrade of udev (or to be precise the udev-init-scripts part of it) because of this:
Code:
Detected file collision(s):

   /lib/udev/net.sh
   /lib/udev/write_root_link_rule
   /lib/udev/rules.d/90-network.rules
   /lib/udev/rules.d/40-gentoo.rules

Searching all installed packages for file collisions...

Press Ctrl-C to Stop

sys-fs/udev-171-r5
   /lib/udev/net.sh
   /lib/udev/rules.d/40-gentoo.rules
   /lib/udev/rules.d/90-network.rules
   /lib/udev/write_root_link_rule

Package 'sys-fs/udev-init-scripts-9' NOT merged due to file
collisions. If necessary, refer to your elog messages for the whole
content of the above message.

So in a situation like this, should I just unmerge udev BEFORE beginning the upgrade process?
And would that not be a dangerous thing to do? Removing udev and module-init-tools?
_________________
Whatever you do, do it properly!
Back to top
View user's profile Send private message
albright
Advocate
Advocate


Joined: 16 Nov 2003
Posts: 2588
Location: Near Toronto

PostPosted: Wed Mar 21, 2012 12:57 pm    Post subject: Reply with quote

Quote:
So in a situation like this, should I just unmerge udev BEFORE beginning the upgrade process?


the udev situation right now seems to be a total c***-up but
I think unmerging udev would be suicide
_________________
.... there is nothing - absolutely nothing - half so much worth
doing as simply messing about with Linux ...
(apologies to Kenneth Graeme)
Back to top
View user's profile Send private message
Jaglover
Watchman
Watchman


Joined: 29 May 2005
Posts: 8291
Location: Saint Amant, Acadiana

PostPosted: Wed Mar 21, 2012 1:45 pm    Post subject: Reply with quote

I just moved those files away to avoid collision (and deleted later).

What I really didn't like there were some warnings printed about missing kernel features (fatal), yet emerge carried on with udev upgrade.
_________________
My Gentoo installation notes.
Please learn how to denote units correctly!
Back to top
View user's profile Send private message
gorkypl
Guru
Guru


Joined: 04 Oct 2010
Posts: 444
Location: Kraków, PL

PostPosted: Wed Mar 21, 2012 2:09 pm    Post subject: Reply with quote

Jaglover wrote:

What I really didn't like there were some warnings printed about missing kernel features (fatal), yet emerge carried on with udev upgrade.

It was discussed way back in 2005 IIRC.
User should not be forced to upgrade kernel before upgrading udev - instead, everyone is expected to read the postinst messages, especially warnings.
Furthermore, breaking udev build process with missing kernel features detected would not save you in situation where you have built a new kernel but forgot to move it into /boot, so still attention is needed.
Back to top
View user's profile Send private message
HMC
Tux's lil' helper
Tux's lil' helper


Joined: 06 Jul 2005
Posts: 79
Location: Australia

PostPosted: Sun Mar 25, 2012 1:33 pm    Post subject: Reply with quote

I was also bitten by this one... chrooted in from a rescue cd, issues with lvm and crypt use flags (saw them on the original update. but were still correct, but different behaviour the second time around), fixed a couple of small issues around the USE flags (particularly static), rebuilt packages, paid attention to output notes, rebuilt the kernel with the correct options (note in there somewhere about the initramfs too) and genkernel (also updated - check the config), updated the init and conf files before rebooting (important- forgot that the first time) and it worked.
I don't really remember too much about the details - it has been done.

The next system that I did was a little hardened netbook and it went off without any issues - it was just a little more work than usual. Likewise the update to udev-182 and beyond (currently r2) went smoothly.
Then a couple of servers...

I reckon that I got bitten the first time because I was lazy and half assed on the update and wouldn't have been bitten if I had followed the notes in the output that the devs are so good to provide (which was ultimately used to fix it) and paid a little attention to the details. I didn't think that there was a bug there and I didn't even bother checking bugzilla or the forums. I put it down to idiot error.

Not saying it is like this for everyone... but it certainly was me.
Back to top
View user's profile Send private message
Etal
Veteran
Veteran


Joined: 15 Jul 2005
Posts: 1931

PostPosted: Sun Mar 25, 2012 3:31 pm    Post subject: Reply with quote

Jaglover wrote:
What I really didn't like there were some warnings printed about missing kernel features (fatal), yet emerge carried on with udev upgrade.


It was actually done like that originally, but then reverted. From the ChangeLog:

ChangeLog wrote:
19 Mar 2012; William Hubbs <williamh@g.o> udev-181.ebuild, udev-9999.ebuild:
Revert making the CONFIG_DEVTMPFS check fatal. Checking for kernel config
options cannot be fatal because it breaks build hosts. See
https://bugs.gentoo.org/show_bug.cgi?id=103878#c29

19 Mar 2012; William Hubbs <williamh@g.o> udev-181.ebuild, udev-9999.ebuild:
make check for CONFIG_DEVTMPFS critical. If this is not set, udev will not
run.
.
_________________
“And even in authoritarian countries, information networks are helping people discover new facts and making governments more accountable.”– Hillary Clinton, Jan. 21, 2010
Back to top
View user's profile Send private message
manny15
Guru
Guru


Joined: 01 Dec 2002
Posts: 473
Location: USA

PostPosted: Sun Mar 25, 2012 8:33 pm    Post subject: Reply with quote

I just recovered my broken system. I thought I paid attention to the emerge output and took care of things, but I must have missed something.

When I rebooted after the upgrade something about a missing kernel config displayed on the screen, then some other error, and then the screen went blank. The monitor alerted me there wasn't a signal coming from the PC.

Once I was able to access /var/log/rc.log I found I was missing CONFIG_DEVTMPFS=y. I'm guessing the other missing config is CONFIG_DEVTMPFS_MOUNT=y.

I learned a long time ago, from a failed upgrade like this one, to do a backup first. My rootfs is on an LVM2 volume. Since I still had the snapshot read-only volume I had backed up from, I booted from a Gentoo LiveCD, formatted the rootfs, and cp --archive'ed from the backup read-only volume to the brand new rootfs volume. I didn't have to use the actual backup tarball. I rebooted and now I'm back pre-disaster. I'll try the upgrade again later.
Back to top
View user's profile Send private message
hunky
l33t
l33t


Joined: 19 Nov 2003
Posts: 903
Location: Alaska

PostPosted: Thu Mar 29, 2012 5:52 am    Post subject: Reply with quote

I'm dead right now.. actually booted into rescuesystemcd. I had a trilogy of errors that cause my trouble. First was I had set portage (emerge) warnings to be emailed to me using an smtp username that was deactivated recently without me realizing it - so I wasn't getting emailed any warnings. Next thing was during the emerge world I got a fail on coreutils. I resumed the emerge but that seems to erase any previous warnings when emerge finishes. So no warnings in the shell. I did update all conf files. Third error was before I started the emerge, on an emerge -avuND @world, before I hit yes, I did see a warning about the upgrade to udev-182 (I'm on unstable) but it seemed to apply only if you had /usr on its own partition. Since I have /boot, /, /home, /swap, I didn't think to worry about it. That was third error.

Now when booting (which doesn't complete) I see udev errors during boot but they go by too fast to read - it is obviously not finding some things. Not sure what to do next - will try manny15's ideas on kernel config. I just looked in the emerge log and see: ERROR: setup
CONFIG_DEVTMPFS: is not set when it should be.
WARN: setup

So looks like I will chroot in and rebuild the kernel. Also remove the 70-persistent-net.rules (or whatever it is named) as I saw errors on network when booting.

off to chroot now. /jd

[edit] ok - that worked. Enabled the CONFIG_DEVTMPFS and ...DEVTMPFS_MOUNT. When booting saw some errors about the udev rules beginning with the number 10 - have to look into that. Hopefully back in the saddle. /jd
Back to top
View user's profile Send private message
FizzyWidget
Veteran
Veteran


Joined: 21 Nov 2008
Posts: 1133
Location: 127.0.0.1

PostPosted: Sun Apr 01, 2012 6:50 am    Post subject: Reply with quote

How would i update my initramfs to deal with the change to udev and everything having to be on /

here is my initramfs

Code:
#!/bin/sh
mount -t proc proc /proc
mount -t sysfs sysfs /sys
mount -t devtmpfs devtmpfs /dev
#for a french azerty keyboard
loadkmap < /etc/kmap-fr
rescue() {
   echo "Dropping to rescue shell" >&2
   /bin/sh </dev/tty1 >/dev/tty1 2>&1
}
/bin/cryptsetup luksOpen /dev/sda2 vault || rescue
/bin/lvm pvscan || rescue
/bin/lvm vgscan || rescue
/bin/lvm vgchange -ay vg || rescue
mount -r /dev/mapper/vg-root /newroot || rescue
CMDLINE=`cat /proc/cmdline`
umount /dev
umount /sys
umount /proc
exec /bin/busybox switch_root /newroot /sbin/init ${CMDLINE}


i tried adding mount -t ext4 /dev/mapper/vg-usr but that didnt work, should i do as most people seem to have suggested in various paces and mask the 1.8.x series?
_________________
I know 43 ways to kill with a SKITTLE, so taste my rainbow bitch.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Sun Apr 01, 2012 3:00 pm    Post subject: Re: udev update killed my system - HELP please :-( Reply with quote

MitsukiP wrote:
I already read https://forums.gentoo.org/viewtopic-t-901206-highlight-udev.html but didn't really help. There was no line with "initramfs=" in my config, i added a initramfs="no" but still the same problem. udev was already in the sysinit runlevel.

Er if you're going to use that solution, you need to patch two udev initscripts (/etc/init.d/udev-mount and /etc/init.d/udev) as outlined in the topic, to do something with the new parameter. (And udev has to move to boot runlevel, with udev-mount and sysfs added to sysinit- they are normally depended on by udev so start automatically in that runlevel.) Note that I haven't tested this against udev-181+, as I'm on stable; it's likely that the patch would need to change for a newer version.

edit: remove link to earlymounts initscript as it seems to be dead.


Last edited by steveL on Fri Nov 09, 2012 9:10 pm; edited 1 time in total
Back to top
View user's profile Send private message
DaveDay
n00b
n00b


Joined: 04 Oct 2007
Posts: 29

PostPosted: Wed Apr 04, 2012 8:45 pm    Post subject: Udev and broken system Reply with quote

About a year ago my wife's Windows XP machine that she uses for medical transcription broke and, having a couple spare dual processor boxes lying about, to save time I just gave her my workstation and transferred her hard drive to it and off she went. I transferred my Gentoo drives to a slightly older and less powerful box and we have been carrying on like that ever since.

Well we were given a new machine this week and so we put her drive in the new box and she is quite happy, and I get my original workstation back. Sounded good.

Except that now my Gentoo disks won't boot on my original workstation that I had Gentoo on for several years before.

It did indeed turn out that I was hit by the CONFIG_DEVTMPFS issue but that was only a small part of it. I used a rescue CD and rebuilt the kernel with CONFIG_DEVTMPFS and the "CONFIG_MOUNT_DEVTMPFS (or similar, don't have it handy) but the problem remained.

I took all the cards (USB, NVidia video, SB Audigy, Enet) out and activated the motherboard equivalents to shake up the HW boot environment some, but the behavior remained.

The machine stops booting just after the "Populating /dev by uevents" message.

Getting desperate, I decided to just reinstall from the March 2012 minimal install disk and low and behold........it behaves exactly the same way! So I tried MythBuntu's latest CD ........ they can't boot that machine either. Puppy Linux did boot just fine and it has been a life saver, allowing me to get to the internet and google around for ideas.

So, realizing that it is at least in part a HW issue, I put everything back in the older Dual processor and I am more or less back in business. I have a remaining locking issue that is stopping LVM2 from coming up and Apache2 is complaining about a configuration issue with PHP but I am guessing those are just the usual day to day adventures you get from running ~x86.

So, beware, there are apparently some hardware gotchas out there that will completely hose udev. I went back to udev 171 and still had the issue. Earlier than that and I started getting dependency problems and I didn't want to go down that road right now so I stopped at 171. I have no idea what the issue is but hopefully the devs will figure out how to just spit out a warning and leave something unconfigured rather than allowing the machine to just freeze.

So far the only solution I found was switching to another box.

Cheers,
Dave
Back to top
View user's profile Send private message
gorkypl
Guru
Guru


Joined: 04 Oct 2010
Posts: 444
Location: Kraków, PL

PostPosted: Wed Apr 04, 2012 9:42 pm    Post subject: Re: Udev and broken system Reply with quote

DaveDay wrote:
I have a remaining locking issue that is stopping LVM2 from coming up

I can help you with this one, probably. Just set locking_dir = "/dev/shm/lvm" in /etc/lvm.conf
https://bugs.gentoo.org/show_bug.cgi?id=409921

No idea about the problem with udev however...
_________________
BTW, TWM FTW!
Back to top
View user's profile Send private message
thorbenk
n00b
n00b


Joined: 25 Jan 2005
Posts: 20

PostPosted: Sat Apr 07, 2012 1:02 am    Post subject: Trying to downgrade to older udev, cannot get it to work Reply with quote

Hello all,

I have /usr on an LVM volume and am running ~amd64. After the upgrade to udev-182, I built a initramfs with dracut, which did not work however. I could not boot my system anymore.

For the time being, I just want to downgrade to a working udev again.

I chrooted into the system and downgraded udev and dependencies to the following packages:

udev 171-r5
consolekit 0.4.5-r2
usbutils 005
pciutils 3.1.9-r1

Still, I get the same error on boot, that /usr is not there. What could possibly be still wrong?

Any help would be tremendously appreciated!

Thanks,
Thorben
Back to top
View user's profile Send private message
sDoky
n00b
n00b


Joined: 14 Dec 2010
Posts: 28
Location: CZE

PostPosted: Sat Apr 07, 2012 7:23 am    Post subject: Reply with quote

Hi, I have upgraded to udev-181-r2 with testing pciutils and usbutils. After doing so I was completely unable to connect to any wireless network. I have intel agn in my thinkpad t420. Downgrade does not help. The issue still remains. I have reverted everything the help of genlop. And also I have updated all the config files.

Code:
[   29.961228] br0: no IPv6 routers present
[   31.213676] wlan0: authenticate with 02:1b:b1:02:80:cf (try 1)
[   31.216911] wlan0: authenticated
[   31.217904] wlan0: associate with 02:1b:b1:02:80:cf (try 1)
[   31.222691] wlan0: RX AssocResp from 02:1b:b1:02:80:cf (capab=0x421 status=0 aid=1)
[   31.222694] wlan0: associated
[   31.228837] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[   31.244962] wlan0: deauthenticating from 02:1b:b1:02:80:cf by local choice (reason=3)
[   31.253455] cfg80211: All devices are disconnected, going to restore regulatory settings
[   31.253461] cfg80211: Restoring regulatory settings
[   31.253472] cfg80211: Calling CRDA to update world regulatory domain
[   35.391933] wlan0: authenticate with 5c:da:d4:76:20:58 (try 1)
[   35.394690] wlan0: authenticated
[   35.395708] wlan0: associate with 5c:da:d4:76:20:58 (try 1)
[   35.398412] wlan0: RX AssocResp from 5c:da:d4:76:20:58 (capab=0x411 status=0 aid=1)
[   35.398417] wlan0: associated
[   35.437614] wlan0: deauthenticating from 5c:da:d4:76:20:58 by local choice (reason=3)
[   35.476196] cfg80211: All devices are disconnected, going to restore regulatory settings
[   35.476204] cfg80211: Restoring regulatory settings
[   35.476219] cfg80211: Calling CRDA to update world regulatory domain
[   42.458096] wlan0: authenticate with 00:12:0e:51:fc:97 (try 1)
[   42.657380] wlan0: authenticate with 00:12:0e:51:fc:97 (try 2)
[   42.660229] wlan0: authenticated
[   42.660475] wlan0: associate with 00:12:0e:51:fc:97 (try 1)
[   42.860075] wlan0: associate with 00:12:0e:51:fc:97 (try 2)
[   42.862638] wlan0: RX AssocResp from 00:12:0e:51:fc:97 (capab=0x401 status=0 aid=5)
[   42.862643] wlan0: associated
[   42.862646] wlan0: invalid AID value 0x5; bits 15:14 not set
[   42.884499] wlan0: deauthenticating from 00:12:0e:51:fc:97 by local choice (reason=3)
[   42.903129] cfg80211: All devices are disconnected, going to restore regulatory settings
[   42.903137] cfg80211: Restoring regulatory settings
[   42.903145] cfg80211: Calling CRDA to update world regulatory domain
[   49.397765] wlan0: authenticate with 00:1b:b1:02:80:f4 (try 1)
[   49.402537] wlan0: authenticated
[   49.402762] wlan0: associate with 00:1b:b1:02:80:f4 (try 1)
[   49.415144] wlan0: RX AssocResp from 00:1b:b1:02:80:f4 (capab=0x421 status=0 aid=1)
[   49.415148] wlan0: associated
[   49.438922] wlan0: deauthenticating from 00:1b:b1:02:80:f4 by local choice (reason=3)
[   49.449379] cfg80211: All devices are disconnected, going to restore regulatory settings
[   49.449386] cfg80211: Restoring regulatory settings
[   49.449393] cfg80211: Calling CRDA to update world regulatory domain

[  159.487260] wlan0: authenticate with 5c:da:d4:76:20:58 (try 1)
[  159.489108] wlan0: authenticated
[  159.490605] wlan0: associate with 5c:da:d4:76:20:58 (try 1)
[  159.493362] wlan0: RX AssocResp from 5c:da:d4:76:20:58 (capab=0x411 status=0 aid=1)
[  159.493365] wlan0: associated
[  159.538204] wlan0: deauthenticating from 5c:da:d4:76:20:58 by local choice (reason=3)
[  159.561769] cfg80211: All devices are disconnected, going to restore regulatory settings
[  159.561775] cfg80211: Restoring regulatory settings
[  159.561785] cfg80211: Calling CRDA to update world regulatory domain


Can you please point me the correct direction?
Back to top
View user's profile Send private message
thorbenk
n00b
n00b


Joined: 25 Jan 2005
Posts: 20

PostPosted: Sat Apr 07, 2012 9:49 am    Post subject: Re: Trying to downgrade to older udev, cannot get it to work Reply with quote

thorbenk wrote:
Hello all,

I have /usr on an LVM volume and am running ~amd64. After the upgrade to udev-182, I built a initramfs with dracut, which did not work however. I could not boot my system anymore.

For the time being, I just want to downgrade to a working udev again.

Still, I get the same error on boot, that /usr is not there. What could possibly be still wrong?


So, the problem seems to have been an unrelated update of lvm2, see bug https://bugs.gentoo.org/show_bug.cgi?id=409921

Once I reverted to the previous version, I got my computer to boot again.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Mon Apr 09, 2012 3:05 pm    Post subject: Reply with quote

If you are using lvm, I very much recommend using /dev/mapper/vg-lv in your /etc/fstab rather than /dev/vg/lv. The /dev/mapper/vg-lv links are made by lvm. BTW you need CONFIG_DEVTMPFS for this, but it's required now for udev; CONFIG_DEVTMPFS_MOUNT is also useful, should you ever need to boot into a rescue shell (it means /dev will automatically be mounted by the kernel, which means the /dev/mapper links will be available however you start your machine.)

If you check /etc/init.d/udev in the populate_dev() function, you'll see the following comment:
Code:

# we can speed up booting under these conditions:
#  * using devtmpfs so kernel creates device nodes for us
#  * only using kernel created device nodes at boot
# (in /etc/fstab and elsewhere)

/dev/mapper, and directories below it, are lvm-created nodes, which need device-mapper built into kernel, if lvm is not started in an initramfs.

edit: lvm creates nodes, not kernel.


Last edited by steveL on Mon Jun 11, 2012 9:23 pm; edited 1 time in total
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
Goto page 1, 2  Next
Page 1 of 2

 
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