KNetworkManager often ignores wireless

[Wolfgang Jeltsch]

Am Samstag, 27. Februar 2010 15:48:05 schrieb Wolfgang Jeltsch:
> Am Samstag, 27. Februar 2010 10:07:27 schrieb Reinhold Rumberger:
> > I have no clue what "reason 2" might be. But the OP isn't alone with
> > his killswitch problem:
> > http://bbs.archlinux.org/viewtopic.php?id=69987
> 
> I’ve just found the /sbin/rfkill binary. Executing “rfkill list” when the
> switch is set to “on” results in the following output:
> 
> 0: dell-wifi: Wireless LAN
>         Soft blocked: no
>         Hard blocked: yes
> 1: phy0: Wireless LAN
>         Soft blocked: no
>         Hard blocked: no
> 
> After having switched to “off”, I get this:
> 
> 0: dell-wifi: Wireless LAN
>         Soft blocked: no
>         Hard blocked: no
> 1: phy0: Wireless LAN
>         Soft blocked: no
>         Hard blocked: yes
> 
> So rfkill detects two devices and one of them is always blocked. Strange.
> Any ideas what could be the cause for this?

I’ve found a solution!

The virtual file /sys/devices/virtual/rfkill/rfkill0/state shows the state of 
the killswitch. An 1 means that wireless is probably enabled and a 2 means 
that it has been disabled from an outside source. See

    <http://www.mjmwired.net/kernel/Documentation/rfkill.txt>  .

The problem was that these two values were swapped in my case. There were a 1 
when wireless was switched off and a 2 when it was actually switched on. This 
probably caused the wrong values for dell-wifi hard blocked in the “rfkill 
list” outputs.

In addition, I found the bug report at

    <https://bugs.launchpad.net/ubuntu/+source/hal/+bug/307411>

and there I found a link to

    <https://bugs.launchpad.net/bugs/193970>  .

The latter gave me the necessary hint. Obviously, the content of 
/sys/devices/virtual/rfkill/rfkill0/state is always 2 initially. Whenever I 
change the state of the (physical) switch, it is changed from 2 to 1 or from 1 
to 2, respectively. So the solution is to boot with wireless switched off and 
switch it on when the machine is up. Then I get the correct contents in 
/sys/devices/virtual/rfkill/rfkill0/state and the output of “rfkill list” is 
also correct.

When I wanted to report my specific problem as a new bug, I encountered

    <https://bugs.launchpad.net/ubuntu/+source/linux/+bug/523143>  ,

which seems to be exactly my problem. I added a comment.

Thanks for all your helpful hints.

Best wishes,
Wolfgang