Wireless switch position not properly detected

Bug #588366 reported by Kris
20
This bug affects 3 people
Affects Status Importance Assigned to Milestone
rfkill (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

In 32-bit Lucid Lynx I was able to get the broadcom BCM4212 wireless working with the proprietary driver that I installed. There is however a very annoying bug related to the wireless switch on this Dell Vostro 1320 and I am not sure if it is related to the switch itself, with the wireless driver or with network-manager. What happens is the following:

Whenever the computer is turned on (or returning from suspend/hibernate) the current position of the wireless switch is recognized by network-manager (actually likely by rfkill) as being in disabled mode. This is fine if that's in fact the position of the switch is actually disabled as enabling it will turn on the hardware (blue wireless light comes on) and network-manager will identify it as active automatically enabling wireless (unless I deactivate it there of course) allowing we to connect. If the switch is actually in the "on" position when the computer is turned on (which it will be most of the time) the blue wireless light is on and seemingly working while network-manager recognizes it as being off and grays out the option of activating wireless. Flipping the switch now will cause network-manager to think it is on (and so it "activates" wireless) but at the same time the hardware is actually turned off (blue light goes off as well) so no networks are ever found.

This is really annoying since I always have to reboot/suspend all the time to be able to use wireless. It is even more frustrating because the computer is not mine (I am just configuring it for a complete linux/ubuntu newbie) and my friend is waiting for a fully functional ubuntu.

I should also mention that the the switch works as it should in Windows Vista (it's a dual-boot configuration).

Trying the behavior from a live-usb (lucid) gives the same results even before activating the proprietary driver (which leads me to believe the problem is not with the driver itself).

Short of a solution, any workaround would be very welcome. Such as fooling kernel/network-manager that wireless switch is always on etc.

Kristofer

Here is the output of some commands that might be useful:

rfkill - seemingly the culprit, here showing hardware as off even though it was on as happens when turning computer on with switch on:

rfkill list
0: dell-wifi: Wireless LAN
 Soft blocked: no
 Hard blocked: yes

Turning the killswitch off at this point "wrongly" toggles the killswitch to the "Hard blocked: no" state.

$lspci | grep Broadcom
0e:00.0 Network controller: Broadcom Corporation BCM4312 802.11b/g (rev 01)

$ lsmod
Module Size Used by
binfmt_misc 6587 1
ppdev 5259 0
vboxnetadp 6326 0
vboxnetflt 15162 0
vboxdrv 190409 2 vboxnetadp,vboxnetflt
snd_hda_codec_idt 51914 1
fbcon 35102 71
tileblit 2031 1 fbcon
font 7557 1 fbcon
bitblit 4707 1 fbcon
softcursor 1189 1 bitblit
vga16fb 11385 0
vgastate 8961 1 vga16fb
joydev 8708 0
snd_hda_intel 21877 2
snd_hda_codec 74201 2 snd_hda_codec_idt,snd_hda_intel
snd_hwdep 5412 1 snd_hda_codec
snd_pcm_oss 35308 0
snd_mixer_oss 13746 1 snd_pcm_oss
snd_pcm 70662 3 snd_hda_intel,snd_hda_codec,snd_pcm_oss
snd_seq_dummy 1338 0
snd_seq_oss 26726 0
snd_seq_midi 4557 0
snd_rawmidi 19056 1 snd_seq_midi
snd_seq_midi_event 6003 2 snd_seq_oss,snd_seq_midi
snd_seq 47263 6 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_midi_event
snd_timer 19098 2 snd_pcm,snd_seq
snd_seq_device 5700 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_rawmidi,snd_seq
i915 282354 3
drm_kms_helper 29297 1 i915
uvcvideo 56990 0
dell_wmi 1793 0
snd 54148 16 snd_hda_codec_idt,snd_hda_intel,snd_hda_codec,snd_hwdep,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_seq_oss,snd_rawmidi,snd_seq,snd_timer,snd_seq_device
lib80211_crypt_tkip 7596 0
drm 162471 4 i915,drm_kms_helper
psmouse 63245 0
videodev 34361 1 uvcvideo
dell_laptop 6856 0
dcdbas 5422 1 dell_laptop
soundcore 6620 1 snd
snd_page_alloc 7076 2 snd_hda_intel,snd_pcm
sdhci_pci 5470 0
sdhci 15462 1 sdhci_pci
led_class 2864 1 sdhci
intel_agp 24177 2 i915
serio_raw 3978 0
agpgart 31724 2 drm,intel_agp
v4l1_compat 13251 2 uvcvideo,videodev
i2c_algo_bit 5028 1 i915
wl 1959598 0
lib80211 5046 2 lib80211_crypt_tkip,wl
video 17375 1 i915
output 1871 1 video
lp 7028 0
parport 32635 2 ppdev,lp
r8169 33884 0
ohci1394 26950 0
mii 4381 1 r8169
ahci 32008 4
ieee1394 81181 1 ohci1394

$ nm-tool
...
- Device: eth2 -----------------------------------------------------------------
  Type: 802.11 WiFi
  Driver: wl
  State: unavailable
  Default: no
  HW Address: 00:26:5E:82:30:5B

  Capabilities:

  Wireless Properties
    WEP Encryption: yes
    WPA Encryption: yes
    WPA2 Encryption: yes

  Wireless Access Points

Revision history for this message
Leo Arias (elopio) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. This bug did not have a package associated with it, which is important for ensuring that it gets looked at by the proper developers. You can learn more about finding the right package at https://wiki.ubuntu.com/Bugs/FindRightPackage . I have classified this bug as a bug in xorg-server.
(With many applications, you can report a future bug directly on the appropriate package by opening the application's "Help" menu and choosing "Report a Problem". https://wiki.ubuntu.com/ReportingBugs)

affects: ubuntu → xorg-server (Ubuntu)
Revision history for this message
Leo Arias (elopio) wrote :

Sorry, lerning in progress... :D
I made the change in the wrong issue.

affects: xorg-server (Ubuntu) → ubuntu
Revision history for this message
Kris (k-kristoferb) wrote :

Hi, I'm not sure if this is related to xorg-server, some further investigation seems to indicate that it is related to rfkill package but I cannot find where to change this. I say this because when the computer is turned on with the killswitch on, rkill list shows the following:

0: dell-wifi: Wireless LAN
 Soft blocked: no
 Hard blocked: yes

Thus it obviously doesnt recognize the correct position of the hardware switch unless the switch is off when starting the computer. I will also update the bug body with some command outputs..

Kris (k-kristoferb)
description: updated
Revision history for this message
Kris (k-kristoferb) wrote :

This bug does not affect xorg-server but rather rfkill.

affects: ubuntu → rfkill (Ubuntu)
Revision history for this message
Kris (k-kristoferb) wrote :

I managed to work around the bug manually by mounting a tmpfs over /sys/devices/virtual/rfkill/rfkill0 but it still not a solution nor an option for inexperienced users... I detailed the workaround here: http://ubuntu-virginia.ubuntuforums.org/showthread.php?p=9421500#post9421500

Revision history for this message
Christopher (captain-c) wrote :

I am also experiencing this issue. It started just recently.

On Dell D630 with wireless switch off:
rfkill list
0: dell-wifi: Wireless LAN
 Soft blocked: no
 Hard blocked: no

with wireless on:
rfkill list
0: dell-wifi: Wireless LAN
 Soft blocked: no
 Hard blocked: yes

Linux 2.6.32-23-generic #37-Ubuntu SMP Fri Jun 11 08:03:28 UTC 2010 x86_64 GNU/Linux

Revision history for this message
Christopher (captain-c) wrote :

So here's something interesting...
I turned the wireless switch to off and typed in

rfkill unblock 0

and now it can connect to wireless and show up correctly in `rfkill list`

BTW, this issue stops me from being able to connect to wireless networks.

Now to try to read up on rfkill, because that was a big shot in the dark!

Revision history for this message
matias nordin (matias-nordin) wrote :

Interesting stuff..

I just tried out a dual boot system on my wifes new Dell E6400 with Windows 7 / Ubuntu 10.

In Windows the wifi works as normal, but thing is:

Windows seem to screw the wifi hotkey so that in Ubuntu it can't be enabled. From help of the above post, I could enable it with

rfkill unblock 0

As the nob I am, I made a quick solution by binding the above command to a keyboard shortcut and put a note on the laptop, that in order to enable wireless, put hotkey in off and press keyboard shortcut, then put hotkey in on.

Would be nice if anyone could make a more persistent solution to the problem.

Revision history for this message
bradi (bradi-teleline) wrote :

Hi.

I have the same problem in my HP pavilion dv8 since I updated to Maverick. It worked fine in Lucid.

Can we consider that there will be a sooner solution for the problem?

Revision history for this message
Darren Salt (dsalt) wrote :

rfkill(8) reports in human-readable form information which it gets from the kernel, or it tells the kernel to change the soft-block states for some devices. It cannot affect the hard-block state unless there's some kernel or, perhaps, BIOS bug which it's triggering.

Revision history for this message
Jim Parker (jimparker96313) wrote :

After many searches I found a solution that does not involve remounting tmpfs, nor rebooting.

see link.
http://linuxtrap.wordpress.com/2009/11/25/hard-blocked-wireless-kill-switch-on-dell-laptop/

Works only for dell laptops, but for other hardware configurations the idea is that you have to reload the kernel module that interfaces the rf-switch.

ie. in this case
$> sudo rmmod dell_laptop
$> sudo modprobe dell_laptop

Networking should automagickally restart, and rfkill now says
$> rfkill list
1:dell-wifi: Wireless LAN
   Soft blocked: no
   Hard blocked: no

HTH,
--Jim Parker

Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in rfkill (Ubuntu):
status: New → Confirmed
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.