ath + wep issue

Sam Leffler sam at errno.com
Thu Aug 4 05:03:21 GMT 2005


Robert C. Noland III wrote:
> Ok, on the ath card the issue seems to be related to the interaction
> with the AP and the roaming mode...  This problem is exacerbated by the
> fact that I sometimes manage to get the card into a state where I can
> see outbound 802.11 frames with tcpdump, but never see inbound packets,
> i.e. probe responses...  Those are the instances when it scans
> forever... Ejecting the card and re-inserting gets me back to a useful
> state.
> 
> This is long, sorry...
> 
> This is what happens when I plug the card in and wpa_supplicant tries to
> configure the card.
> 
> Aug  3 12:02:39 bbeng-laptop kernel: ath0: <Atheros 5212> mem
> 0xf4010000-0xf401ffff irq 11 at device 0.0 on cardbus1
> Aug  3 12:02:39 bbeng-laptop kernel: ath0: Ethernet address:
> 00:12:17:6e:fc:18
> Aug  3 12:02:39 bbeng-laptop kernel: ath0: mac 5.9 phy 4.3 radio 3.6
> Aug  3 12:03:13 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 12:03:13 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 12:03:18 bbeng-laptop last message repeated 28 times
> Aug  3 12:03:18 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> AUTH
> Aug  3 12:03:23 bbeng-laptop kernel: ath0: ieee80211_newstate: AUTH ->
> SCAN
> Aug  3 12:03:23 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> AUTH
> Aug  3 12:03:28 bbeng-laptop kernel: ath0: ieee80211_newstate: AUTH ->
> SCAN
> Aug  3 12:03:28 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 12:03:28 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 12:03:33 bbeng-laptop last message repeated 28 times
> Aug  3 12:03:39 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 12:03:39 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 12:03:44 bbeng-laptop last message repeated 28 times
> Aug  3 12:03:50 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 12:03:50 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 12:03:55 bbeng-laptop last message repeated 28 times
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> AUTH
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: AUTH ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] assoc
> success: long preamble, long slot time
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> RUN
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: link state changed to UP
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 9)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: RUN ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: link state changed to DOWN
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 7)
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> ASSOC
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: invalid
> transition
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 12:03:56 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 12:03:59 bbeng-laptop last message repeated 15 times
> Aug  3 12:03:59 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> INIT
> 
> It continues scanning, but never associates again...

The error responses from the ap seem to indicate dropped frames.  What 
does ifconfig ath0 list scan show for the rssi?  If possible you might 
try moving the ap to channel 6 or 11.

> 
> When I "wpa_cli term" wpa_supplicant, it leaves the card like this...
> 
> ath0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> mtu 1500
>         inet6 fe80::212:17ff:fe6e:fc18%ath0 prefixlen 64 scopeid 0x3 
>         inet 0.0.0.0 netmask 0xff000000 broadcast 255.255.255.255
>         ether 00:12:17:6e:fc:18
>         media: IEEE 802.11 Wireless Ethernet autoselect (autoselect)
>         status: no carrier
>         ssid ""
>         authmode OPEN privacy OFF deftxkey 1 txpowmax 34 roaming DEVICE
>         bintval 100
> 
> NOTE: roaming is set to DEVICE

That's just wpa_supplicant; it should be fixed to restore the previous 
settings instead of blindly forcing a fixed value on cleanup.

> 
> bbeng-laptop# ifconfig ath0 ssid "test001" authmode shared wepmode on
> weptxkey 1 wepkey "xxxxx"
> 
> Aug  3 13:50:20 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 13:50:20 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 13:50:51 bbeng-laptop last message repeated 151 times
> Aug  3 13:52:52 bbeng-laptop last message repeated 593 times
> 
> card not seeing probe responses... later test same state below...
> 
> Aug  3 15:29:00 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 15:29:00 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 15:29:06 bbeng-laptop last message repeated 28 times
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> AUTH
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: ieee80211_newstate: AUTH ->
> ASSOC
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] assoc
> success: long preamble, long slot time
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> RUN
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: link state changed to UP
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 9)
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: ieee80211_newstate: RUN ->
> ASSOC
> Aug  3 15:29:06 bbeng-laptop kernel: ath0: link state changed to DOWN
> 
> ath0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
>         inet6 fe80::212:17ff:fe6e:fc18%ath0 prefixlen 64 scopeid 0x3 
>         inet 0.0.0.0 netmask 0xff000000 broadcast 255.255.255.255
>         ether 00:12:17:6e:fc:18
>         media: IEEE 802.11 Wireless Ethernet autoselect (DS/1Mbps)
>         status: no carrier
>         ssid test001 channel 1
>         authmode SHARED privacy ON deftxkey 1 wepkey 1:104-bit txpowmax
> 46
>         protmode CTS roaming DEVICE bintval 100
> 
> lights on card blinking like it is associated and no further scanning.

The card was left in ASSOC state so the lights reflect that.  Because 
roaming was set to device nothing progressed.  But the basic problem is 
still you don't appear to be getting frames through to the ap.

> 
> bbeng-laptop# ifconfig ath0 roaming auto
> 
> Aug  3 14:07:36 bbeng-laptop kernel: ath0: ieee80211_newstate: INIT ->
> SCAN
> Aug  3 14:07:36 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> SCAN
> Aug  3 14:07:42 bbeng-laptop last message repeated 28 times
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: SCAN ->
> AUTH
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: AUTH ->
> ASSOC
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] assoc
> success: long preamble, long slot time
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> RUN
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: link state changed to UP
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 9)
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: RUN ->
> ASSOC
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: link state changed to DOWN
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] assoc
> success: long preamble, long slot time
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> RUN
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: link state changed to UP
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] recv
> disassociated (reason 9)
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: RUN ->
> ASSOC
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: link state changed to DOWN
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: [00:0d:93:e9:cf:d4] assoc
> success: long preamble, long slot time
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: ieee80211_newstate: ASSOC ->
> RUN
> Aug  3 14:07:42 bbeng-laptop kernel: ath0: link state changed to UP
> 
> It's all good now... 
> 
> ath0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
>         inet6 fe80::212:17ff:fe6e:fc18%ath0 prefixlen 64 scopeid 0x3 
>         inet 0.0.0.0 netmask 0xff000000 broadcast 255.255.255.255
>         ether 00:12:17:6e:fc:18
>         media: IEEE 802.11 Wireless Ethernet autoselect (DS/1Mbps)
>         status: associated
>         ssid test001 channel 1 bssid 00:0d:93:e9:cf:d4
>         authmode SHARED privacy ON deftxkey 1 wepkey 1:104-bit txpowmax
> 46
>         protmode CTS bintval 100
> 
> I either have or can provide tcpdumps of each specific case if that
> helps more, but I figure this message is long enough already.

Better would be a 3rd sta sniffing traffic and also recording rssi 
(collect -y IEEE802_11_RADIO).  Your problem seems unrelated to wep or 
roaming mode; you appear to just not get frames through reliably.  When 
this happens I look at stats on the sta and the ap.  athstats can be 
useful here; something like

athstats 1

will give you a rolling display of stats together with current rssi. 
Sometimes you can see obvious problems like high retransmit rates or big 
bursts of noise.  If the frame count on the ap goes up as you transmit 
then it's likely you're not hearing the ACK's coming back.  If the ap 
doesn't hear your frames (as appears to be happening) then it can either 
be noise, misconfig (e.g. 11g parameters like protection wrong in a 
mixed b/g bss), or possibly low tx power by the sta.  The latter would 
appear as low rssi on recv'd frames at the ap and/or a 3rd sta.

I don't see an indication of what you're using for an ap.  Also remove 
everything like crypto and shared key auth for testing--get 
communication working before adding more variables.  Figuring out 
communication problems can be hard w/o a good test environment and tools.

> 
> Now, off to debug the other issues with the wi card, sigh...
> 
> robert.
> 
> On Tue, 2005-08-02 at 19:54 -0700, Sam Leffler wrote:
> 
>>Robert C. Noland III wrote:
>>
>>>I am having an issue with static wep on my ath card, -current sources as
>>>of this morning EST.  It is a Linksys WPC55ag card.  I also have an old
>>>Netgear wi card that works shown at the bottom.
>>>
>>>ath_hal: 0.9.14.9 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413)
>>>...
>>>ath0: <Atheros 5212> mem 0xf4010000-0xf401ffff irq 11 at device 0.0 on
>>>cardbus0
>>>ath0: Ethernet address: 00:12:17:6e:fc:18
>>>ath0: mac 5.9 phy 4.3 radio 3.6
>>>
>>>laptop# ifconfig ath0 ssid "test001" wepmode on weptxkey 1 wepkey "xxxx"
>>>laptop# ifconfig ath0 up
>>>
>>>Tue Aug  2 17:27:19 RTM_IFINFO: if# 3, link: down,
>>>flags:<UP,BROADCAST,SIMPLEX,MULTICAST>
>>>Tue Aug  2 17:27:25 RTM_IEEE80211: scan complete
>>>Tue Aug  2 17:27:25 RTM_IEEE80211: associate with 00:0d:93:e9:cf:d4
>>>Tue Aug  2 17:27:25 RTM_IFINFO: if# 3, link: up,
>>>flags:<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST>
>>>Tue Aug  2 17:27:25 RTM_IEEE80211: disassociate
>>>Tue Aug  2 17:27:25 RTM_IFINFO: if# 3, link: down,
>>>flags:<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST>
>>
>>You associated with the ap then fell off either because the ap dropped 
>>you or your station initiated it.  The easiest way to see which is to 
>>enabling debugging; I prefer to use 80211debug from tools/tools/ath; e.g.
>>
>>80211debug state+assoc
>>
>>or probably
>>
>>ifconfig ath0 debug
>>
>>will give you enough info.  The output of 80211stats should also tell 
>>you what happened.
>>
>>
>>>The disassociate message concerns me, but it should work...
>>>
>>>ath0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
>>>        inet6 fe80::212:17ff:fe6e:fc18%ath0 prefixlen 64 scopeid 0x3 
>>>        ether 00:12:17:6e:fc:18
>>>        media: IEEE 802.11 Wireless Ethernet autoselect (DS/1Mbps)
>>>        status: no carrier
>>>        ssid test001 channel 1
>>>        authmode OPEN privacy ON deftxkey 1 wepkey 1:104-bit txpowmax 46
>>>        protmode CTS roaming DEVICE bintval 100
>>>
>>>At this point, the card is no longer scanning and the lights on the card
>>>continue blinking as if it is associated (slowly in sync, as opposed to
>>>alternately blinking like mad when it is scanning) yet status shows no
>>>carrier.  dhclient will not work when the interface is in this state.
>>>It seems to just broadcast and eventually die.
>>>
>>>wi0: <NETGEAR MA401 Wireless PC Card> at port 0xe000-0xe03f irq 11
>>>function 0 config 1 on pccard0
>>>wi0: using RF:PRISM2 MAC:HFA3841 CARD:HWB3163-SST-flash
>>>wi0: Intersil Firmware: Primary (0.3.0), Station (1.3.4)
>>>wi0: Ethernet address: 00:30:ab:07:e4:7b
>>>
>>>laptop# ifconfig wi0 ssid "test001" wepmode on weptxkey 1 wepkey "xxxx"
>>>laptop# ifconfig wi0 up
>>>
>>>Tue Aug  2 17:35:57 RTM_IFANNOUNCE: if# 3, what: arrival
>>>Tue Aug  2 17:39:06 RTM_IFINFO: if# 3, link: unknown,
>>>flags:<UP,BROADCAST,SIMPLEX,MULTICAST>
>>>Tue Aug  2 17:39:06 RTM_IEEE80211: disassociate
>>>Tue Aug  2 17:39:06 RTM_IFINFO: if# 3, link: down,
>>>flags:<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST>
>>>
>>>again with the disassociate message.  This seems to force wpa_supplicant
>>>to start scanning again, so I can't currently use wpa_supplicant on this
>>>network.  Actually, wpa_supplicant doesn't ever seem to be able to
>>>associate on the wi card, but static configuration works.
>>>wpa_supplicant on the ath card seems to associate, then catches the
>>>disassociate and starts scanning again.  lather, rinse, repeat...
>>>
>>>wi0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
>>>        inet6 fe80::230:abff:fe07:e47b%wi0 prefixlen 64 scopeid 0x3 
>>>        ether 00:30:ab:07:e4:7b
>>>        media: IEEE 802.11 Wireless Ethernet autoselect (DS/11Mbps)
>>>        status: associated
>>>        ssid test001 channel 1 bssid 00:0d:93:e9:cf:d4
>>>        stationname "FreeBSD WaveLAN/IEEE node"
>>>        authmode OPEN privacy MIXED deftxkey 1 wepkey 1:104-bit txpowmax
>>>100
>>>        bintval 100
>>>
>>>dhclient does work here, and then I can start the vpn and roam about the
>>>office...
>>
>>I've been testing ath w/ static key wep today and seen no problems. 
>>Please get some more info.
>>
>>	Sam
>>_______________________________________________
>>freebsd-current at freebsd.org mailing list
>>http://lists.freebsd.org/mailman/listinfo/freebsd-current
>>To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"
> 
> 
> _______________________________________________
> freebsd-current at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"
> 



More information about the freebsd-current mailing list