headers
David Ehrmann | 1 Jun 2008 09:58
Picon

Re: wpa_supplicant v0.6.3 loses connection, then reconnects to a different AP

Partial update on this.

The symptom was that iwconfig reported it wasn't associated with an AP.

Here's the abbreviated log: http://pastebin.com/f491af16d

I suppose it's possible that the AP goes down, and the disassociation 
isn't a bug.  I'm going to log this more, in hopes of finding out.

David Ehrmann wrote:
> When my connection drops, wpa_supplicant 0.6.3 sometimes reconnects to
> an AP on the same SSID, but with a different BSSID.  I thought that
> specifying the BSSID would force that one to always be used.
> 
> Am I missing something?  Here's my config:
> 
> ctrl_interface=/var/run/wpa_supplicant
> 
> eapol_version=2
> ap_scan=1
> 
> network={
>          ssid="..."
>          scan_ssid=1
>          bssid=...
>          key_mgmt=WPA-EAP
>          pairwise=CCMP TKIP
>          group=CCMP TKIP
>          eap=PEAP TLS
>          identity=...
(Continue reading)

Permalink | Reply | 8 comments |
headers
Christian Rothländer | 1 Jun 2008 10:40

WPA_Supplicant, iwl3945 and la Fonera

Hi there

I've got a weird problem. My wireless is working without any flaw at 
home (AP: Linksys WRT54g using OpenWRT). If I try to use any la Fonera I 
only get connection if I reset the router before trying to connect 
(Power off, power on).
Sometimes I run into trouble at other AP's, too, but usually after a 
iwl3945-module-reload its working again.

This time I recorded the difference between both connection trials. My 
father's Laptop is working without problem, so I conclude it has to be 
an issue with either the iwl3945 (tried several different driver 
versions) or with the wpa_supplicant. My current WPA_supplicant Version 
is 0.5.10 (the one gentoo recommends)

Logs attached, perhaps somebody is able to see the problem. Association 
is failing somewhere during the process. Already tried to change WPA1/2 
and CCMP/TKIP in the config file.

So thanks in advance,

Christian

p.S.: sorry if the first mail comes through, used the wrong email address :(

Config:

ctrl_interface=/var/run/wpa_supplicant
ctrl_interface_group=wheel
#update_config=1
(Continue reading)

Permalink | Reply | 1 comment |
headers
Carolin Latze | 1 Jun 2008 19:01
Picon
Picon
Favicon

[Fwd: Re: [PATCH] enhanced smartcard support]

Hi Jouni,

my patches are too big for the listserver. You can download them at:

http://diuf.unifr.ch/people/latzec/patches/

Regards
Carolin

-------- Original Message --------
Subject: 	Re: [PATCH] enhanced smartcard support
Date: 	Fri, 30 May 2008 17:06:56 +0200
From: 	Carolin Latze <carolin.latze <at> unifr.ch>
To: 	hostap <at> lists.shmoo.com
References: 	<200805020136.09480.dds <at> google.com> 
<20080521072939.GH12378 <at> jm.kir.nu> 
<87d4ng59cj.fsf <at> piyo.tok.corp.google.com> 
<20080523084728.GA5575 <at> jm.kir.nu> <483685E7.4010501 <at> unifr.ch> 
<20080523162259.GB4932 <at> jm.kir.nu>

Hi Jouni,

I'm sorry I didn't have time to apply my patch on the newest version. 
When I started my project, version 0.5.9 was the newest, so that is the 
newest patch I am able to provide at the moment. I will work on a newer 
version when I find the time...

First of all you need trousers in order to access your TPM, but I think, 
you already installed it.
(Continue reading)

Permalink | Reply | 0 comments |
headers
David Ehrmann | 2 Jun 2008 00:01
Picon

Re: wpa_supplicant v0.6.3 loses connection, then reconnects to a different AP

What do these info messages mean?

"RSN: PMKID candidate event - bssid=xx:xx:xx:04:85:aa index=0 preauth=0"

"RSN: Ignored PMKID candidate without preauth flag"
(yes, the previous message indicates that preauth is 0, but why is the 
message ignored)

David Ehrmann wrote:
> Partial update on this.
> 
> The symptom was that iwconfig reported it wasn't associated with an AP.
> 
> Here's the abbreviated log: http://pastebin.com/f491af16d
> 
> I suppose it's possible that the AP goes down, and the disassociation 
> isn't a bug.  I'm going to log this more, in hopes of finding out.
> 
> 
> David Ehrmann wrote:
>> When my connection drops, wpa_supplicant 0.6.3 sometimes reconnects to
>> an AP on the same SSID, but with a different BSSID.  I thought that
>> specifying the BSSID would force that one to always be used.
>>
>> Am I missing something?  Here's my config:
>>
>> ctrl_interface=/var/run/wpa_supplicant
>>
>> eapol_version=2
>> ap_scan=1
(Continue reading)

Permalink | Reply | 7 comments |
headers
Yen-Jung Chang | 2 Jun 2008 04:52
Picon

Re: [PATCH] Fix race conditions in the WPA group-key state machine

Please ignore the previous one patch, it would cause an infinite loop.
Attached is the updated patch.

Regards,
YJ.


2008/5/22 Yen-Jung Chang <yenjung <at> gmail.com>:
Hi Jouni,

I run into the same issue with 0.4.8, so I back port the robust mechanism from 0.5.10 to 0.4.8.

Would you please give me some advice about the attached patch?

Thanks,
YJ.


2008/3/18 Jouni Malinen <j <at> w1.fi>:

On Thu, Mar 06, 2008 at 09:18:46AM +0200, Uri Simchoni wrote:

> Working with madwifi driver, we've encountered some rare conditions in
> which clients can't receive WPA-encrypted multicast packets. This has
> become a more serious lately since Windows Vista uses broadcast packets
> for DHCP by default.
>
> After digging into it for a while, we realized that the group-key state
> machine was stuck in the SETKEYS state, meaning that it negotiates a new
> group key, but doesn't activate the key at the driver. Further
> investigation showed that the GNoStations variable got negative.

Thanks for reporting this.

> The supplied patch (against 0.4.10, sorry...) fixes these issues, and
> also remove the reliance on GNoStations, because it looks like a
> not-very-robust way to determine how many stations need to negotiate
> group key (although it's based on the standard...)

I replaced the group key update management with a more robust mechanism
about a month ago (the new version is included in 0.5.10). It uses a bit
different solution when compared to your patch, but I would expect the
end result to be quite similar since GNoStations variable was removed
completely and only the stations that are really active and ready for
group key update are include in the dynamic count.

I haven't ported the patch to 0.4.x branch yet, but I would prefer to
use the same solution in all branches to minimize amount of code to
maintain in the future.

--
Jouni Malinen                                            PGP id EFC895FA
_______________________________________________
HostAP mailing list
HostAP <at> lists.shmoo.com
http://lists.shmoo.com/mailman/listinfo/hostap


Attachment (robust-group-key-update-0.4.8.patch): application/octet-stream, 2703 bytes
_______________________________________________
HostAP mailing list
HostAP <at> lists.shmoo.com
http://lists.shmoo.com/mailman/listinfo/hostap
Permalink | Reply | 0 comments |
headers
Jouni Malinen | 2 Jun 2008 11:33
Picon

Re: wpa_supplicant v0.6.3 loses connection, then reconnects to a different AP

On Sun, Jun 01, 2008 at 03:01:06PM -0700, David Ehrmann wrote:
> What do these info messages mean?
> 
> "RSN: PMKID candidate event - bssid=xx:xx:xx:04:85:aa index=0 preauth=0"

This means that the driver reported that there is a neighboring AP for
the same SSID.

> "RSN: Ignored PMKID candidate without preauth flag"
> (yes, the previous message indicates that preauth is 0, but why is the 
> message ignored)

WPA2 (IEEE 802.11i) has an option for completing most time consuming
part of authentication prior to connecting to a new network. This is
called preauthentication and supplicant can trigger it based on the
PMKID candidate (i.e., candidate for deriving and caching keys) events.
Since this particular AP does not advertise support for
preauthentication, this event is ignored.

--

-- 
Jouni Malinen                                            PGP id EFC895FA
Permalink | Reply | 6 comments |
headers
Raghavendra | 2 Jun 2008 14:53

Relation between Signal Level, Noise Level and RSSI

Hi,

I have following two data, which display Signal Level = -79dBm, Noise Level 
= -88dBm and RSSI of 177.

Can any one tell me whats the relation between these three? any 
equation..etc. basicaly I want to know how RSSI value is determined as 177?

 # iwconfig eth0
eth0      XXXXX  ESSID:"xxxx"  Nickname:""
          Mode:Managed  Frequency:2.412 GHz  Access Point: 00:1D:7E:0E:3F:54
          Bit Rate:54 Mb/s   Tx-Power=15 dBm
          Retry limit:9   RTS thr=2347 B   Fragment thr=2346 B
          Encryption key:off
          Power Management:off
          Link Quality:0/100  Signal level:-79 dBm  Noise level:-88 dBm
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:2350
          Tx excessive retries:169  Invalid misc:41836   Missed beacon:0

# cat ./sys/class/net/eth0/wireless/level
177

Also if any one have idea about
-90 ~ -80dbm: level 1
-79dbm ~ -69dbm: level 2
-69dbm ~ -59dbm: level 3
-59dbm ~ 0: level4.
what these level stands for?

-Raghu
Permalink | Reply | 1 comment |
headers
Pavel Roskin | 2 Jun 2008 15:03
Picon

Re: Relation between Signal Level, Noise Level and RSSI

On Mon, 2008-06-02 at 18:23 +0530, Raghavendra wrote:
> Hi,
> 
> I have following two data, which display Signal Level = -79dBm, Noise Level 
> = -88dBm and RSSI of 177.
> 
> Can any one tell me whats the relation between these three? any 
> equation..etc. basicaly I want to know how RSSI value is determined as 177?

It looks like RSSI is signed integer, so it's -79, the same as the
signal level.  177=256-79

There is no relation between noise and signal.

All details should probably be asked in the mailing list for your
driver, as they are hardware and driver specific.  If you want to know
more about physics or information theory, Google is your friend.

> Also if any one have idea about
> -90 ~ -80dbm: level 1
> -79dbm ~ -69dbm: level 2
> -69dbm ~ -59dbm: level 3
> -59dbm ~ 0: level4.
> what these level stands for?

That might be empirical levels used for displaying signal quality as
bars for the end users.

--

-- 
Regards,
Pavel Roskin
Permalink | Reply | 0 comments |
headers
Pavel Roskin | 2 Jun 2008 15:20
Picon

Re: WPA_Supplicant, iwl3945 and la Fonera

On Sun, 2008-06-01 at 10:40 +0200, Christian Rothländer wrote:
> network={
>        ssid="Rothlaender"
>        scan_ssid=1
>        psk="psk" <- that's not the real one ;-)
>        key_mgmt=WPA-PSK
>        priority=3
> }

Try adding this line to the above block:

proto=WPA

This would force WPA1.  Fonera uses MadWifi internally, and my
experience shows that some versions of MadWifi don't work with WPA2
correctly when in the master mode.

Just in case, try forcing WPA2 as well:

proto=WPA2

--

-- 
Regards,
Pavel Roskin
Permalink | Reply | 0 comments |
headers
Dan Williams | 2 Jun 2008 19:30
Picon
Favicon

[PATCH] wext: don't overwrite BSS frequency

mac80211 sends _both_ channel and frequency in it's scan results, with
frequency first and channel second (it's since been fixed to send
channel first and frequency second to work around this issue).  This
results in wpa_supplicant getting the right value when the frequency
comes, but overwriting the value with '0' when the channel comes because
wpa_supplicant can't handle 5GHz channel numbers.  So if a valid
previous SIOCGIWFREQ event came in, don't try to overwrite it.

diff --git a/src/drivers/driver_wext.c b/src/drivers/driver_wext.c
index 69aae16..60cdb79 100644
--- a/src/drivers/driver_wext.c
+++ b/src/drivers/driver_wext.c
 <at>  <at>  -1294,8 +1294,15  <at>  <at>  static void wext_get_scan_freq(struct iw_event *iwe,
 		/*
 		 * Some drivers do not report frequency, but a channel.
 		 * Try to map this to frequency by assuming they are using
-		 * IEEE 802.11b/g.
+		 * IEEE 802.11b/g.  But don't overwrite a previously parsed
+		 * frequency if the driver sends both frequency and channel,
+		 * since the driver may be sending an A-band channel that we
+		 * don't handle here.
 		 */
+
+		if (res->res.freq)
+			return;
+
 		if (iwe->u.freq.m >= 1 && iwe->u.freq.m <= 13) {
 			res->res.freq = 2407 + 5 * iwe->u.freq.m;
 			return;
Permalink | Reply | 2 comments |

Gmane