[Click] RXStats problem with Mac addresses

[Saquib Razak]

hello,

I am using RXStats Element in my script and when I read the stats handler, most of the mac addresses in the received data seem to be corrupted. 

When I list scan on my ath device, I get the following access points

>click-1.6.0/conf/wifi# wlanconfig ath1 list scan
SSID            BSSID              CHAN RATE  S:N   INT CAPS
Titanium        00:1a:70:ab:19:5f    6   54M 11:0   100 EP  
Titanium        00:18:39:33:35:16    6   54M  8:0   100 EPSs
SpeedStream     00:13:a3:e3:ae:5d   11   54M 17:0   200 Es  
USR5461         00:14:c1:1d:31:64   11   54M 10:0   100 E   
ahideen         00:18:d1:0c:3b:6f   11   54M  6:0   200 EPs  WPA


>From Click, I am running dump.click with minor modifications. 
FromDevice(ath1) 
-> prism2_decap :: Prism2Decap()
-> extra_decap :: ExtraDecap()
-> countme :: Counter()
-> tap_decap :: RadiotapDecap()
-> c :: RXStats()
-> PrintWifi() 
-> Classifier(!0/80%f0) // filter out beacons
-> WifiDecap()
-> Strip(14)
-> CheckIPHeader()
-> IPPrint(ip)
-> Discard;

The print wifi element shows me the access points properly. Following is a sample of the output

78 |  2Mb + 7/170 | mgmt beacon     00:18:D1:2A:37:6D SpeedStream chan 11 b_int 200 [ ESS ] ({2 4 11 22} 44 12 18 24 36 48 72 96 108)
92 |  2Mb +13/171 | mgmt probe_resp 00:18:D1:0C:3B:6F ahideen chan 11 b_int 200 [ ESS PRIVACY ] ({2 4 11 22} 44 12 18 24 36 48 72 96 108)
78 |  1Mb + 9/171 | mgmt probe_resp 00:14:C1:1D:31:64 USR5461 chan 11 b_int 100 [ ESS ] ({2 4 11 22} 36 48 72 108 12 18 24 96)

but when I read these values from RXStats, all mac addresses seem to be messed up. Here is a sample of the output

D1:0C:FF:FF:00:18 rate 4 signal 12 noise 177 avg_signal 12.250000 avg_noise 177.500000 total_signal 98 total_noise 1420 packets 8 last_received 6.278450
D1:2A:E1:1B:00:18 rate 4 signal 12 noise 177 avg_signal 11.750000 avg_noise 177.500000 total_signal 47 total_noise 710 packets 4 last_received 6.852487

The first line seems to be access point "ahideen" (its just a guess because the mac addresses are quite similar with some FF's and some bytes rotated!)
The second line seems to be access point "Speed Stream".

Has anybody experience something similar with RXStats or am I missing something here. This test was conducted on the 20th floor of a high rise building with 46 floors. There are several APs around so it could be an interference issue but print wifi seems to be spitting right data. 

Thanks
Saquib




      ____________________________________________________________________________________
Never miss a thing.  Make Yahoo your home page. 
http://www.yahoo.com/r/hs