[PlanetCCRMA] Pulling USB cable on FC2 machine causes keyboard and network to crash...

Fri Apr 1 04:36:04 2005

Hi Steve!

On Apr 1, 2005 12:27 AM, Steve Harris <S.W.Harris@ecs.soton.ac.uk> wrote:
> On Thu, Mar 31, 2005 at 10:30:34 -0800, Mark Knecht wrote:
> > Hi,
> >    In the process of cleaning the house this morning my wife pulled
> > the USB cable going to an external USB to Optical device we use. I had
> > assumed  that this would be OK to do. After that the keyboard was
> > dead. I couldn't get to the machine through the network so apparently
> > ndiswrapper got harmed also. Hotplugging the keyboard itself didn't
> > help. Since the power switch does a reasonable shutdown operation we
> > rebooted and everything is up and running again.
> >
> >    Has anyone else seen this? The USB device is card 1. The onboard
> > sound chip is card 0.
> >
> >    I haven't tried repeating as of yet.

and I still haven't.

> 
> If there anything in /var/log/messages? There might be an oops dump or
> similar.

Good point. Here's what I see:

Mar 31 07:58:47 dragonfly su(pam_unix)[7918]: session closed for user news
Mar 31 08:13:47 dragonfly gconfd (matt-4103): Exiting
Mar 31 08:13:47 dragonfly gdm(pam_unix)[3878]: session closed for user matt
Mar 31 08:13:47 dragonfly kernel: mtrr: base(0xf0020000) is not
aligned on a size(0x400000) boundary

Here's where it was (I think) when I did the disconnect. Actually when
we were cleaning we have disconnected the optical cable. I then had
the bright idea of showing my wife that we could disconnect the USB
cable instead. I pulled it an immediately plugged it back in:

Mar 31 09:10:25 dragonfly kernel: usb 3-2: USB disconnect, address 2
Mar 31 09:10:26 dragonfly udev[8239]: removing device node '/udev/input/event2'
Mar 31 09:10:26 dragonfly udev[8176]: removing device node '/udev/dsp1'
Mar 31 09:10:26 dragonfly udev[8171]: removing device node '/udev/mixer1'
Mar 31 09:10:26 dragonfly udev[8190]: removing device node '/udev/audio1'
Mar 31 09:10:26 dragonfly udev[8197]: removing device node '/udev/snd/controlC1'
Mar 31 09:10:26 dragonfly udev[8199]: removing device node '/udev/snd/pcmC1D0p'
Mar 31 09:10:26 dragonfly udev[8201]: removing device node '/udev/snd/pcmC1D0c'

And then the reconnect:

Mar 31 09:10:27 dragonfly kernel: usb 3-2: new full speed USB device
using uhci_hcd and address 3
Mar 31 09:10:27 dragonfly kernel: input: USB HID v1.00 Device
[Burr-Brown from TI               USB Audio CODEC ] on
usb-0000:00:1d.1-2
Mar 31 09:10:29 dragonfly udev[8388]: configured rule in
'/etc/udev/rules.d//50-udev.rules' at line 17 applied, 'pcmC1D0p'
becomes 'snd/%k'
Mar 31 09:10:29 dragonfly udev[8388]: creating device node '/udev/snd/pcmC1D0p'
Mar 31 09:10:29 dragonfly udev[8389]: configured rule in
'/etc/udev/rules.d//50-udev.rules' at line 17 applied, 'pcmC1D0c'
becomes 'snd/%k'
Mar 31 09:10:29 dragonfly udev[8389]: creating device node '/udev/snd/pcmC1D0c'
Mar 31 09:10:29 dragonfly udev[8392]: configured rule in
'/etc/udev/rules.d//50-udev.rules' at line 15 applied, 'controlC1'
becomes 'snd/%k'
Mar 31 09:10:29 dragonfly udev[8392]: creating device node '/udev/snd/controlC1'
Mar 31 09:10:29 dragonfly udev[8390]: creating device node '/udev/dsp1'
Mar 31 09:10:29 dragonfly udev[8393]: creating device node '/udev/mixer1'
Mar 31 09:10:29 dragonfly udev[8391]: creating device node '/udev/audio1'
Mar 31 09:10:39 dragonfly udev[8397]: configured rule in
'/etc/udev/rules.d//50-udev.rules' at line 25 applied, 'event2'
becomes 'input/%k'
Mar 31 09:10:39 dragonfly udev[8397]: creating device node '/udev/input/event2'
Mar 31 09:46:53 dragonfly gdm(pam_unix)[3878]: session opened for user
evelyn by (uid=0)
Mar 31 09:46:54 dragonfly gconfd (evelyn-8515): starting (version
2.6.0), pid 8515 user 'evelyn'
Mar 31 09:46:54 dragonfly gconfd (evelyn-8515): Resolved address
"xml:readonly:/etc/gconf/gconf.xml.mandatory" to a read-only config
source at position 0
Mar 31 09:46:54 dragonfly gconfd (evelyn-8515): Resolved address
"xml:readwrite:/home/evelyn/.gconf" to a writable config source at
position 1
Mar 31 09:46:54 dragonfly gconfd (evelyn-8515): Resolved address
"xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only config
source at position 2
Mar 31 09:46:55 dragonfly bonobo-activation-server (evelyn-8520): iid
OAFIID:BrokenNoType:20000808 has a NULL type
Mar 31 09:46:55 dragonfly bonobo-activation-server (evelyn-8520):
invalid character '#' in iid
'OAFIID:This#!!%$iid%^$%_|~!OAFIID_ContainsBadChars'
Mar 31 09:47:26 dragonfly udev[8660]: removing device node '/udev/mixer1'
Mar 31 09:47:26 dragonfly udev[8671]: removing device node '/udev/dsp1'
Mar 31 09:47:26 dragonfly udev[8682]: removing device node '/udev/audio1'
Mar 31 09:47:26 dragonfly udev[8734]: removing device node '/udev/input/event2'
Mar 31 09:47:26 dragonfly kernel: usb 3-2: USB disconnect, address 3
Mar 31 09:47:34 dragonfly kernel: ndiswrapper (hangcheck_proc:379):
Hangcheck returned true. Resetting wlan0!
Mar 31 09:47:57 dragonfly kernel: usb 3-2: new full speed USB device
using uhci_hcd and address 4
Mar 31 09:47:57 dragonfly kernel: ALSA
/usr/src/rpm/BUILD/alsa-driver-1.0.9rc1/acore/init.c:104: cannot find
the slot for index 1 (range 0-1)
Mar 31 09:47:57 dragonfly kernel: ALSA
/usr/src/rpm/BUILD/alsa-driver-1.0.9rc1/usb/usbaudio.c:3155: cannot
create card instance 0
Mar 31 09:47:57 dragonfly kernel: snd-usb-audio: probe of 3-2:1.0
failed with error -5
Mar 31 09:47:58 dragonfly kernel: input: USB HID v1.00 Device
[Burr-Brown from TI               USB Audio CODEC ] on
usb-0000:00:1d.1-2
Mar 31 09:48:09 dragonfly udev[8895]: configured rule in
'/etc/udev/rules.d//50-udev.rules' at line 25 applied, 'event2'
becomes 'input/%k'
Mar 31 09:48:09 dragonfly udev[8895]: creating device node '/udev/input/event2'
Mar 31 09:48:16 dragonfly kernel: ndiswrapper (hangcheck_proc:379):
Hangcheck returned true. Resetting wlan0!
Mar 31 09:48:49 dragonfly last message repeated 2 times
Mar 31 09:49:59 dragonfly last message repeated 5 times
Mar 31 09:51:08 dragonfly last message repeated 5 times
Mar 31 09:52:14 dragonfly last message repeated 5 times
Mar 31 09:53:29 dragonfly last message repeated 5 times
Mar 31 09:54:35 dragonfly last message repeated 5 times
Mar 31 09:55:50 dragonfly last message repeated 5 times
Mar 31 09:57:11 dragonfly last message repeated 4 times
Mar 31 09:58:14 dragonfly last message repeated 5 times
Mar 31 09:59:29 dragonfly last message repeated 4 times
Mar 31 10:00:32 dragonfly last message repeated 3 times
Mar 31 10:01:41 dragonfly last message repeated 5 times
Mar 31 10:02:50 dragonfly last message repeated 5 times
Mar 31 10:04:00 dragonfly last message repeated 5 times
Mar 31 10:05:15 dragonfly last message repeated 4 times
Mar 31 10:06:21 dragonfly last message repeated 4 times
Mar 31 10:07:30 dragonfly last message repeated 5 times
Mar 31 10:08:39 dragonfly last message repeated 5 times
Mar 31 10:09:48 dragonfly last message repeated 3 times
Mar 31 10:10:06 dragonfly kernel: ndiswrapper (hangcheck_proc:379):
Hangcheck returned true. Resetting wlan0!
Mar 31 10:10:08 dragonfly shutdown: shutting down for system halt
Mar 31 10:10:08 dragonfly init: Switching to runlevel: 0

Some of the stuff above abou t'Bad characters' etc. doesn't look very
good but may or may not be a kernel bug.

Thanks for poking me and getting me to look at this.

cheers,
Mark