udev hangs for 30s after upgrade

kcirick · 2012-01-21 06:06:41

Hello,

I just did an upgrade (normal upgrade, using pacman -Syu, nothing fancy) and I found it hangs on boot/reboot. dmesg didn't really show anything except that it hang for 30s, which is udev timeout in /etc/rc.conf. I downgraded udev to 175 (and mkinitcpio) and things are back to normal (no hang).

...
Jan 21 00:45:44 localhost kernel: [    7.586550] drm: registered panic notifier
Jan 21 00:45:44 localhost kernel: [    7.586565] [drm] Initialized radeon 2.12.0 20080528 for 0000:00:01.0 on minor 0
Jan 21 00:45:44 localhost kernel: [    7.586800] snd_hda_intel 0000:00:01.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19
Jan 21 00:45:44 localhost kernel: [    7.586923] snd_hda_intel 0000:00:01.1: irq 41 for MSI/MSI-X
Jan 21 00:45:44 localhost kernel: [    7.586964] snd_hda_intel 0000:00:01.1: setting latency timer to 64
Jan 21 00:45:44 localhost kernel: [    7.977584] HDMI status: Codec=0 Pin=3 Presence_Detect=0 ELD_Valid=0
Jan 21 00:45:44 localhost kernel: [    7.977893] input: HD-Audio Generic HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.1/sound/card0/input9
Jan 21 00:45:44 localhost kernel: [    7.978536] snd_hda_intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Jan 21 00:45:44 localhost kernel: [    7.978640] snd_hda_intel 0000:00:14.2: setting latency timer to 64
Jan 21 00:45:44 localhost kernel: [    8.219857] hda_codec: CX20588: BIOS auto-probing.
Jan 21 00:45:44 localhost kernel: [    8.221007] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input10
Jan 21 00:45:44 localhost kernel: [   38.315309] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
Jan 21 00:45:44 localhost kernel: [   38.561765] EXT4-fs (sda6): re-mounted. Opts: (null)
Jan 21 00:45:44 localhost kernel: [   38.731678] fuse init (API version 7.17)
...

On the bootup screen, it showed some message that looked something like "udevd ... seq 'devices/pci000...' killed " and "worker ... terminated by signal 9" (Sorry for vague error message... i couldn't find these message in any of the log files under /var/log, so I just wrote down by hand what I could write down before it went away).

I was wondering what's up with this? If anyone else has this behaviour? Any known solutions/workarounds? I am using Acer AspireOne 722, if this information helps with anything. I can provide with more information too...

EDIT:
This is the dmesg with downgraded udev/mkinitcpio, where things were working fine.

...
[    8.069505] input: HD-Audio Generic HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.1/sound/card0/input9
[    8.070197] snd_hda_intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[    8.070336] snd_hda_intel 0000:00:14.2: setting latency timer to 64
[    8.312285] hda_codec: CX20588: BIOS auto-probing.
[    8.313306] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input10
[    8.911733] EXT4-fs (sda6): re-mounted. Opts: (null)
[    9.081439] fuse init (API version 7.17)
[   15.414861] NET: Registered protocol family 10
...

Last edited by kcirick (2012-01-21 16:36:06)

vwyodajl · 2012-01-21 08:10:17

I just posted the same thing in the packages part of the forum Did not notice your post first.

Anyways I was able to get the full error in the boot log. I was wondering what was up with it as well, though not noticing anything not working other than the insane delay/hang.

cgarcia · 2012-01-21 10:26:34

Same problem here. Fortunately, udev is waited now for only half a minute (I remember waiting minutes in the past with older boot scripts). Downgrading udev and mkinitcpio seems to fix the problem too.

karol · 2012-01-21 11:04:03

You may want to change the title to add it hangs for 30 seconds. As it is now, I though it hangs forever.

It may be http://projects.archlinux.org/svntogit/ … ab3addfaf7 - the 30 sec timeout that udev 177-3 introduced for broken kernel drivers (description taken from http://projects.archlinux.org/initscrip … 4686d64826 )

bwat47 · 2012-01-21 15:58:42

I've been having a very similar issue since this update:
https://bbs.archlinux.org/viewtopic.php?id=134013

[brandon@brandon-arch ~]$ dmesg | grep 2400
[    6.937508] i2400m_usb 1-1.1:1.0: WiMAX interface wmx0 (64:d4:da:04:85:0c) ready
[   38.161789] i2400m_usb 1-1.1:1.0: boot-mode cmd -1: error waiting for an ack: -4
[   38.161828] i2400m_usb 1-1.1:1.0: bootrom init failed: -4
[   38.161850] i2400m_usb 1-1.1:1.0: i6050-fw-usb-1.5.sbcf: cannot use: -4, skipping
[   38.161942] i2400m_usb 1-1.1:1.0: Could not find a usable firmware image
[   38.161971] i2400m_usb 1-1.1:1.0: cannot bootstrap device: -4
[   38.208629] i2400m_usb 1-1.1:1.0: cannot setup device: -4
[   38.208685] i2400m_usb: probe of 1-1.1:1.0 failed with error -4
[   38.208733] usbcore: registered new interface driver i2400m_usb

The boot hang is really annoying. Didn't get any of these errors before the update. Can't downgrade because I happened to run a pacman -Sc the other day

Last edited by bwat47 (2012-01-21 16:04:08)

swordfish · 2012-01-21 16:48:18

I had the same problem with a Realtek wireless module (rtl8192ce). On this forum I got the hint that placing the module in question into the MODULES section of /etc/rc.conf would be a workaround. I have tried it and it works for me - system boots fast again .

...
# HARDWARE
# --------
MODULES=(rtl8192ce)
USEDMRAID="no"
USEBTRFS="no"
USELVM="no"
...

Last edited by swordfish (2012-01-21 17:27:54)

bwat47 · 2012-01-21 17:18:18

swordfish wrote:

I had the same problem with a Realtek wireless module (rtl8192ce). On this forum I got the hint that placing the module in question into the MODULES section of /etc/rc.conf would be a workaround. I have tried it and it works for me - sytem boots fast again .
...
# HARDWARE
# --------
MODULES=(rtl8192ce)
USEDMRAID="no"
USEBTRFS="no"
USELVM="no"
...

Tried that and it still hung for me (i2400m_usb)

I guess I could just blacklist it since i dont use the wimax anyway (why the hell didn't I think of that before lol)

Last edited by bwat47 (2012-01-21 17:27:49)

kcirick · 2012-01-21 17:31:58

~~The above suggestion (putting the modules in MODULES=() didn't help me either (brcmfmac). I can't really blacklist my wifi module because I need internet~~

~~I'm assuming this is going to get fixed at some point (either in future linux kernel package or udev package)~~

EDIT: Sorry, I was putting in the wrong module name. The above suggestion worked for me. Thanks for the workaround swordfish!

Last edited by kcirick (2012-01-21 17:48:27)

bwat47 · 2012-01-21 17:52:45

Yeah I wasn't quite as lucky, I made sure to put in the right name and got the exact same errors, just under "Loading User Specified Modules" instead. Luckily blacklisting it worked fine for me, and the i2400m_usb module is useless to me anyway Even before the update when it was loading correctly it was adding a few seconds to my boot.

jgreen1tc · 2012-01-21 20:15:40

After I updated udev this morning, I have been seeing this on boot:

Sat Jan 21 13:49:18 2012: :: Starting UDev Daemon    [BUSY]    [DONE] 
Sat Jan 21 13:49:18 2012: :: Triggering UDev uevents    [BUSY]    [DONE] 
Sat Jan 21 13:49:18 2012: :: Waiting for UDev uevents to be processed    [BUSY] udevd[237]: worker [268] timeout, kill it
Sat Jan 21 13:49:18 2012: 
Sat Jan 21 13:49:18 2012: udevd[237]: seq 1253 '/devices/pci0000:00/0000:00:1c.1/0000:02:00.0' killed
Sat Jan 21 13:49:18 2012: 
Sat Jan 21 13:49:18 2012: udevd[237]: worker [268] terminated by signal 9 (Killed)
Sat Jan 21 13:49:18 2012: 
Sat Jan 21 13:49:18 2012:    [DONE]

As far as I can tell so far, nothing is affected but I'm not sure (please correct me if this is cause for concern). It does increase my boot time however and I find that slightly disappointing/inconvenient. I've been searching the web for answers as to what is happening there but I can't seem to find out. Can anyone link me to an article or point me in the right direction so I can try to find out what's going on?

karol · 2012-01-21 20:23:46

See https://bbs.archlinux.org/viewtopic.php?id=134012

ewaller · 2012-01-21 20:31:10

Moderator: Merged threads

Diaz · 2012-01-21 20:59:39

Just started the computer and noticed this problem too.

Is there anything i can do to help sove the problem?

twelveeighty · 2012-01-21 21:33:39

I can confirm that adding my Realtek wireless rtl8192se in the MODULES section of rc.conf did the trick as well.

jgreen1tc · 2012-01-21 21:38:31

twelveeighty wrote:

I can confirm that adding my Realtek wireless rtl8192se in the MODULES section of rc.conf did the trick as well.

+1

My thanks to those who helped provide the solution and reading material.

Fisaulerod · 2012-01-22 00:05:26

I have the same problem but I don't know what I have to add to modules (i don't know what is rtl8192se or i2400m_usb)
thanks!

willbprog127 · 2012-01-22 00:55:43

Yuck, I'm having the same problem. Never had boot issues until now (after the upgrade to stock 3.2.1 ARCH kernel and udev 177. Boot hangs until the 30 second timeout.

Sat Jan 21 16:09:09 2012: :: Waiting for UDev uevents to be processed    [BUSY] udevd[173]: worker [193] timeout, kill it
Sat Jan 21 16:09:09 2012: 
Sat Jan 21 16:09:09 2012: udevd[173]: seq 1096 '/devices/pci0000:00/0000:00:05.0/0000:08:00.0' killed
Sat Jan 21 16:09:09 2012: 
Sat Jan 21 16:09:09 2012: udevd[173]: worker [193] terminated by signal 9 (Killed)

I don't have the slightest clue what to put in my MODULES section of rc.conf, and don't really want to do that anyway because it seems kinda hackish.

kcirick · 2012-01-22 01:06:13

For those who are still having problems: The new udev hang is due to the recent commit to the udev as described here. Reading through it, it seems it mostly affects wireless driver, so if you have no idea what module they need to add/blacklist, it's your wireless. You can search through Arch Wiki to find what wifi card you have.

@willbprog127, modifying rc.conf is nothing really to be scared about. You can also read about it in the wiki if you need to understand better so that you can be more comfortable with what you're editing. I would say it's an essential part of using Arch linux, so get used to it

willbprog127 · 2012-01-22 01:13:14

kcirick wrote:

@willbprog127, modifying rc.conf is nothing really to be scared about. You can also read about it in the wiki if you need to understand better so that you can be more comfortable with what you're editing. I would say it's an essential part of using Arch linux, so get used to it

Ah my friend, I am not scared to edit it...I have made many changes to it through the months I've used Arch. Perhaps I was unclear. I just wish I didn't have to do this to have a smooth boot experience. I will try to find the exact wireless module and add it to MODULES.

Fisaulerod · 2012-01-22 01:33:48

kcirick wrote:

For those who are still having problems: The new udev hang is due to the recent commit to the udev as described here. Reading through it, it seems it mostly affects wireless driver, so if you have no idea what module they need to add/blacklist, it's your wireless. You can search through Arch Wiki to find what wifi card you have.
@willbprog127, modifying rc.conf is nothing really to be scared about. You can also read about it in the wiki if you need to understand better so that you can be more comfortable with what you're editing. I would say it's an essential part of using Arch linux, so get used to it

I have this:
03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 06)

I have to add rtl8111? In /lib/modules/3.2.1-1-ARCH/.../wireless/ i have rt2x00, rtl818x, rtlwifi and others....i feel like a dummy u.u

willbprog127 · 2012-01-22 02:14:51

Adding brcmsmac to MODULES in rc.conf fixed the 30 second hang for me, however I still wonder if this is the 'official resolution' to issues like this, or if the developers will fix this problem...surely this kind of workaround is not part of the elegant 'Arch Way', is it?

kcirick · 2012-01-22 02:43:31

Fisaulerod wrote:

I have this:
03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 06)
I have to add rtl8111? In /lib/modules/3.2.1-1-ARCH/.../wireless/ i have rt2x00, rtl818x, rtlwifi and others....i feel like a dummy u.u

Hi,

I think the kernel module you want to add is rtl818x

try it and see if it works.

willbprog127 · 2012-01-22 02:53:00

willbprog127 wrote:

surely this kind of workaround is not part of the elegant 'Arch Way', is it?

Sorry if that sounded whiny or too harsh...needed some dinner. ;-) I am VERY grateful for all of the hard work that everyone puts into Arch and into Linux.

twelveeighty · 2012-01-22 04:12:34

Fisaulerod wrote:

I have this:
03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 06)
I have to add rtl8111? In /lib/modules/3.2.1-1-ARCH/.../wireless/ i have rt2x00, rtl818x, rtlwifi and others....i feel like a dummy u.u

Don't feel bad - it's a bit of a crapshoot to find the right one, unless you're well-versed in driver-speak. I did the following:

Run as root:
> kmod list

And look through the list - mine had entries like this:
...
rtl8192se 88473 0
rtlwifi 91390 1 rtl8192se
...

Which I presumed was my wireless (wifi). Hence the "rtl8192se"...

bwat47 · 2012-01-22 04:17:33

willbprog127 wrote:

Adding brcmsmac to MODULES in rc.conf fixed the 30 second hang for me, however I still wonder if this is the 'official resolution' to issues like this, or if the developers will fix this problem...surely this kind of workaround is not part of the elegant 'Arch Way', is it?

Its my understanding from the bug report that this is an issue with these effected drivers, and needs to be fixed upstream in the kernel.

https://bugs.archlinux.org/task/27938

(See the last comment)

Last edited by bwat47 (2012-01-22 04:19:12)

Arch Linux

#1 2012-01-21 06:06:41

udev hangs for 30s after upgrade

#2 2012-01-21 08:10:17

Re: udev hangs for 30s after upgrade

#3 2012-01-21 10:26:34

Re: udev hangs for 30s after upgrade

#4 2012-01-21 11:04:03

Re: udev hangs for 30s after upgrade

#5 2012-01-21 15:58:42

Re: udev hangs for 30s after upgrade

#6 2012-01-21 16:48:18

Re: udev hangs for 30s after upgrade

#7 2012-01-21 17:18:18

Re: udev hangs for 30s after upgrade

#8 2012-01-21 17:31:58

Re: udev hangs for 30s after upgrade

#9 2012-01-21 17:52:45

Re: udev hangs for 30s after upgrade

#10 2012-01-21 20:15:40

Re: udev hangs for 30s after upgrade

#11 2012-01-21 20:23:46

Re: udev hangs for 30s after upgrade

#12 2012-01-21 20:31:10

Re: udev hangs for 30s after upgrade

#13 2012-01-21 20:59:39

Re: udev hangs for 30s after upgrade

#14 2012-01-21 21:33:39

Re: udev hangs for 30s after upgrade

#15 2012-01-21 21:38:31

Re: udev hangs for 30s after upgrade

#16 2012-01-22 00:05:26

Re: udev hangs for 30s after upgrade

#17 2012-01-22 00:55:43

Re: udev hangs for 30s after upgrade

#18 2012-01-22 01:06:13

Re: udev hangs for 30s after upgrade

#19 2012-01-22 01:13:14

Re: udev hangs for 30s after upgrade

#20 2012-01-22 01:33:48

Re: udev hangs for 30s after upgrade

#21 2012-01-22 02:14:51

Re: udev hangs for 30s after upgrade

#22 2012-01-22 02:43:31

Re: udev hangs for 30s after upgrade

#23 2012-01-22 02:53:00

Re: udev hangs for 30s after upgrade

#24 2012-01-22 04:12:34

Re: udev hangs for 30s after upgrade

#25 2012-01-22 04:17:33

Re: udev hangs for 30s after upgrade

Board footer