Re: [ath5k-devel] "iwlist scan" shows same ap twice

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[Luis R. Rodriguez]

On Tue, Nov 17, 2009 at 11:29 AM, Florian Engelhardt <flo@dotbox.org> wrote:
> Hello list,
>
> i recently have a problem with my ath5k wireless card in my Acer Aspire One
> Netbook. Executing iwlist scan shows some Wifi`s twice, but with different
> address and signal quality, which leads to me not being able to connect to
> this wlan.
> I attached an example output of iwlist scan.
> The WLAN "MA-WLAN" exists twice in the list, none of the Addresses is
> correct. I had the same problem with my WLAN (FOOBAR) until i moved it from
> channel 6 to channel 1.
>
> Can you please help me, what is going on here?
>
> This bug is realy annoying to me, couse every few days i have to change the
> wlan channel for it to work. Unloading an reloading the ath5k module does
> not help.
>
> Kind regards
>
> Florian
>
> PS: Please CC me, couse i am not on the list

That's wild. What kernel are you using? Have you tried:

http://wireless.kernel.org/en/users/Download/stable

or

http://wireless.kernel.org/en/users/Download

Can you provide the output of 'iw list', your dmesg, and your 'iw dev
wlan0 scan'.

  Luis

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[Luis R. Rodriguez]

On Wed, Nov 18, 2009 at 11:09 AM, Florian Engelhardt <flo@dotbox.org> wrote:
> On Tue, 17 Nov 2009 13:35:24 -0800, "Luis R. Rodriguez" <mcgrof@gmail.com>
> wrote:
>
>> That's wild. What kernel are you using?
>
> # uname -a
> Linux eka 2.6.31-ARCH #1 SMP PREEMPT Tue Nov 10 19:48:17 CET 2009 i686
> Intel(R) Atom(TM) CPU N270 @ 1.60GHz GenuineIntel GNU/Linux
>
>> Have you tried:
>> http://wireless.kernel.org/en/users/Download/stable
>> or
>> http://wireless.kernel.org/en/users/Download
>
> Tried both, no change.
>
>> Can you provide the output of 'iw list', your dmesg, and your 'iw dev
>> wlan0 scan'.
>
> See attached files. Despite that i can add one additional information: I
> have a second Laptop with Intel WLAN Chip, and it shows exactly the same
> behaviour, MA-WLAN two, sometimes three times, with none of the addresses
> being correct.
>
> Somehow, i can not reproduce this kind of behaviour with my own wlan. It
> shows up only one time, not multiple, but it did two days ago on the ath5k
> chipset based netbook

As Bob pointed out your scan shows two beacons received on different BSSes:

00:80:48:5c:86:5d
00:80:48:5c:86:48

The TSF on the second one seems odd though. Do you have access to the
AP? Can you change the channel to see if you see the two beacons on a
new channel?

Can you try rebooting it?

BSS 00:80:48:5c:86:5d (on wlan0)
	TSF: 276107264385 usec (3d, 04:41:47)
	freq: 2457
	beacon interval: 100
	capability: ESS Privacy ShortPreamble (0x0031)
	signal: -92.00 dBm
	SSID: MA-WLAN
	Supported rates: 1.0* 2.0* 5.5 11.0
	DS Parameter set: channel 10
	Country: NA (in/outdoor), data: 0a 02 12
	ERP: <no flags>
	Extended supported rates: 6.0 9.0 12.0 18.0 24.0 36.0 48.0 54.0
	RSN:	 * Version: 1
		 * Group cipher: TKIP
		 * Pairwise ciphers: TKIP CCMP
		 * Authentication suites: PSK
		 * Capabilities: PreAuth (0x0001)
BSS 00:80:48:5c:86:48 (on wlan0)
	TSF: 921985 usec (0d, 00:00:00)
	freq: 2457
	beacon interval: 100
	capability: ESS Privacy ShortPreamble (0x0031)
	signal: -99.00 dBm
	SSID: MA-WLAN
	Supported rates: 1.0* 2.0* 5.5 11.0
	DS Parameter set: channel 10
	Country: NA (in/outdoor), data: 0a 02 12
	ERP: <no flags>
	Extended supported rates: 6.0 9.0 12.0 18.0 24.0 36.0 48.0 54.0
	RSN:	 * Version: 1
		 * Group cipher: TKIP
		 * Pairwise ciphers: TKIP CCMP
		 * Authentication suites: PSK
		 * Capabilities: PreAuth (0x0001)

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[Florian Engelhardt]

On Wed, 18 Nov 2009 11:18:01 -0800
"Luis R. Rodriguez" <mcgrof@gmail.com> wrote:

> On Wed, Nov 18, 2009 at 11:09 AM, Florian Engelhardt <flo@dotbox.org>
> wrote:
> > On Tue, 17 Nov 2009 13:35:24 -0800, "Luis R. Rodriguez"
> > <mcgrof@gmail.com> wrote:
> >
> >> That's wild. What kernel are you using?
> >
> > # uname -a
> > Linux eka 2.6.31-ARCH #1 SMP PREEMPT Tue Nov 10 19:48:17 CET 2009
> > i686 Intel(R) Atom(TM) CPU N270 @ 1.60GHz GenuineIntel GNU/Linux
> >
> >> Have you tried:
> >> http://wireless.kernel.org/en/users/Download/stable
> >> or
> >> http://wireless.kernel.org/en/users/Download
> >
> > Tried both, no change.
> >
> >> Can you provide the output of 'iw list', your dmesg, and your 'iw
> >> dev wlan0 scan'.
> >
> > See attached files. Despite that i can add one additional
> > information: I have a second Laptop with Intel WLAN Chip, and it
> > shows exactly the same behaviour, MA-WLAN two, sometimes three
> > times, with none of the addresses being correct.
> >
> > Somehow, i can not reproduce this kind of behaviour with my own
> > wlan. It shows up only one time, not multiple, but it did two days
> > ago on the ath5k chipset based netbook
> 
> As Bob pointed out your scan shows two beacons received on different
> BSSes:
> 
> 00:80:48:5c:86:5d
> 00:80:48:5c:86:48
> 
> The TSF on the second one seems odd though. Do you have access to the
> AP? Can you change the channel to see if you see the two beacons on a
> new channel?
> Can you try rebooting it?

Sorry, no. I have no access to that AP, but i can ask the owner if he
may reboot it, or if he faces any problems with it.

/Florian

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[Luis R. Rodriguez]

On Wed, Nov 18, 2009 at 08:31:56PM +0100, Florian Engelhardt wrote:
> On Wed, 18 Nov 2009 11:18:01 -0800
> "Luis R. Rodriguez" <mcgrof@gmail.com> wrote:
> 
> > On Wed, Nov 18, 2009 at 11:09 AM, Florian Engelhardt <flo@dotbox.org>
> > wrote:
> > > On Tue, 17 Nov 2009 13:35:24 -0800, "Luis R. Rodriguez"
> > > <mcgrof@gmail.com> wrote:
> > >
> > >> That's wild. What kernel are you using?
> > >
> > > # uname -a
> > > Linux eka 2.6.31-ARCH #1 SMP PREEMPT Tue Nov 10 19:48:17 CET 2009
> > > i686 Intel(R) Atom(TM) CPU N270 @ 1.60GHz GenuineIntel GNU/Linux
> > >
> > >> Have you tried:
> > >> http://wireless.kernel.org/en/users/Download/stable
> > >> or
> > >> http://wireless.kernel.org/en/users/Download
> > >
> > > Tried both, no change.
> > >
> > >> Can you provide the output of 'iw list', your dmesg, and your 'iw
> > >> dev wlan0 scan'.
> > >
> > > See attached files. Despite that i can add one additional
> > > information: I have a second Laptop with Intel WLAN Chip, and it
> > > shows exactly the same behaviour, MA-WLAN two, sometimes three
> > > times, with none of the addresses being correct.
> > >
> > > Somehow, i can not reproduce this kind of behaviour with my own
> > > wlan. It shows up only one time, not multiple, but it did two days
> > > ago on the ath5k chipset based netbook
> > 
> > As Bob pointed out your scan shows two beacons received on different
> > BSSes:
> > 
> > 00:80:48:5c:86:5d
> > 00:80:48:5c:86:48
> > 
> > The TSF on the second one seems odd though. Do you have access to the
> > AP? Can you change the channel to see if you see the two beacons on a
> > new channel?
> > Can you try rebooting it?
> 
> Sorry, no. I have no access to that AP, but i can ask the owner if he
> may reboot it, or if he faces any problems with it.

Yeah ask him for the actual MAC address of the BSS you are supposed to
be connecting to. Did you tell him your cards are seeing two beacons
from it?

Might want to get a sniffer an capture two beacons to show this info.

  Luis

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[Luis R. Rodriguez]

On Fri, Dec 4, 2009 at 1:02 PM, Florian Engelhardt <flo@dotbox.org> wrote:
> On Wed, 18 Nov 2009 12:01:08 -0800, "Luis R. Rodriguez" <mcgrof@gmail.com>
> wrote:
>> On Wed, Nov 18, 2009 at 11:42 AM, Bob Copeland <me@bobcopeland.com>
> wrote:
>>>
>>> Ok, fair enough.  We should try to figure out where the 0x5d and 0x43
>>> come from.  It could be memory corruption, for example, but 0x5d and
>>> 0x43 aren't poison values.  You may try turning on slub/slab debugging
>>> just to be sure.
>>
>> He noted he sees this with iwl drivers though., Florian, please
>> correct me if I'm wrong.
>
> You are right. Any news on this?
>
> i attached a new scan from the Laptop with intel chipset (iwlagn module).
> The first Cell ist the correct access point the second cell with the same
> ESSID doesn't exist. I don't know why iwlist scan finds this AP.

Please consider burrying iwlist and iwconfig under some ditch and use iw.

> The
> Problem with all this is very annoying, even unloading iwlagn, iwlcore,
> mac80211 and cfg80211 and reloading all these modules does not help.

Why are you reporting iwlagn stuff on the ath5k devel list?

You also did not provide feedback on the AP, whether you asked its
admin about its config nor if it can be rebooted.

 Luis

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[John W. Linville]

On Fri, Dec 04, 2009 at 01:16:54PM -0800, Luis R. Rodriguez wrote:
> On Fri, Dec 4, 2009 at 1:02 PM, Florian Engelhardt <flo@dotbox.org> wrote:
> > On Wed, 18 Nov 2009 12:01:08 -0800, "Luis R. Rodriguez" <mcgrof@gmail.com>
> > wrote:
> >> On Wed, Nov 18, 2009 at 11:42 AM, Bob Copeland <me@bobcopeland.com>
> > wrote:
> >>>
> >>> Ok, fair enough.  We should try to figure out where the 0x5d and 0x43
> >>> come from.  It could be memory corruption, for example, but 0x5d and
> >>> 0x43 aren't poison values.  You may try turning on slub/slab debugging
> >>> just to be sure.
> >>
> >> He noted he sees this with iwl drivers though., Florian, please
> >> correct me if I'm wrong.
> >
> > You are right. Any news on this?
> >
> > i attached a new scan from the Laptop with intel chipset (iwlagn module).
> > The first Cell ist the correct access point the second cell with the same
> > ESSID doesn't exist. I don't know why iwlist scan finds this AP.
> 
> Please consider burrying iwlist and iwconfig under some ditch and use iw.

Someone suggested reimplementing iwlist, iwconfig, etc as scripts
around iw -- any takers? :-)

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

Re: [ath5k-devel] "iwlist scan" shows same ap twice

[Luis R. Rodriguez]

On Fri, Dec 4, 2009 at 1:49 PM, John W. Linville <linville@tuxdriver.com> wrote:
> On Fri, Dec 04, 2009 at 01:16:54PM -0800, Luis R. Rodriguez wrote:
>> On Fri, Dec 4, 2009 at 1:02 PM, Florian Engelhardt <flo@dotbox.org> wrote:
>> > On Wed, 18 Nov 2009 12:01:08 -0800, "Luis R. Rodriguez" <mcgrof@gmail.com>
>> > wrote:
>> >> On Wed, Nov 18, 2009 at 11:42 AM, Bob Copeland <me@bobcopeland.com>
>> > wrote:
>> >>>
>> >>> Ok, fair enough.  We should try to figure out where the 0x5d and 0x43
>> >>> come from.  It could be memory corruption, for example, but 0x5d and
>> >>> 0x43 aren't poison values.  You may try turning on slub/slab debugging
>> >>> just to be sure.
>> >>
>> >> He noted he sees this with iwl drivers though., Florian, please
>> >> correct me if I'm wrong.
>> >
>> > You are right. Any news on this?
>> >
>> > i attached a new scan from the Laptop with intel chipset (iwlagn module).
>> > The first Cell ist the correct access point the second cell with the same
>> > ESSID doesn't exist. I don't know why iwlist scan finds this AP.
>>
>> Please consider burrying iwlist and iwconfig under some ditch and use iw.
>
> Someone suggested reimplementing iwlist, iwconfig, etc as scripts
> around iw -- any takers? :-)

This may help:

http://wireless.kernel.org/en/users/Documentation/iw/replace-iwconfig

I think the *only* command missing is 'iwconfig wlan0 power off'. And
maybe a generic 'iw link' which lists all link information for all
wiphy interfaces, but that should be easy to implement, just have to
get around to it.

  Luis