[Click] gen_config_ap.pl problem

[John Bicket]

Hi Nels,

I just updated the config with some more debug messages, and a possible
fix - proberesponses include what channel the ap is on, so if it's not
in sync with the device, then clients will ignore the probes.

Anyway, could you try it again and if it doesn't work mail me 
/click/messages again?

I tried it with a colleague's windows laptop and it associated.

--john

Nels Anderson [Nels.Anderson at colorado.edu] wrote:
>Aloha,
>
>I grabbed the new script, but the station still can't see the ap.  It continues
>to sees packets from it, but it isn't associating.  Nothing in /click/errors. 
>The contents of /click/messages is this:
>
>expensive Packet::push; have 8 wanted 24
>expensive Packet::push; have 8 wanted 24
>expensive Packet::push; have 8 wanted 24
>expensive Packet::push; have 8 wanted 24
>expensive Packet::push; have 8 wanted 24
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid  rates { * 2 * 4 11 22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid  rates { * 2 * 4 11 22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid  rates { * 2 * 4 11 22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>pr :: ProbeResponder: ProbeReq: 00:04:e2:80:45:42 ssid test rates { * 2 * 4 11
>22 }
>#This line continues to repeat.
>
>That line didn't show up in /click/messages until I set the essid on the station
>and it tried to associate, so it appears they're at least able to poke at each
>other.  If you need anything else from me, just let me know what.  Thanks again
>for the assistance,
>
>-Nels
>
>Quoting John Bicket <jbicket at amsterdam.lcs.mit.edu>:
>
>> Hi Nels,
>>
>> I just checked in a few more changes to that script. Can you try it
>> again, and if it doesn't work can you email me the contents of
>> /click/messages?  I should be able to debug it from there.
>>
>> --john
>>
>>
>>
>> Nels Anderson [Nels.Anderson at colorado.edu] wrote:
>> >Aloha,
>> >
>> >Excellent, thanks.  The router initializes now, but now I'm not able to get
>> >stations nearby (as in less than a meter) to associate with the click ap.
>> Is
>> >there anything special that needs to be done, or should it just sorta go?
>> >Setting the essid on the stations just results in nothing, but looking at a
>> >tcpdump of the stations traffic shows that it's at least seeing packets from
>> >the ap.  Again, thanks for all the assistance,
>> >
>> >-Nels
>> >
>> >Quoting John Bicket <jbicket at amsterdam.lcs.mit.edu>:
>> >
>> >>
>> >> Hi Nels,
>> >>
>> >> Sorry that's a typo. I fixed it just now, and updated cvs.
>> >>
>> >> --john
>> >>
>> >>
>> >> Nels Anderson [Nels.Anderson at colorado.edu] wrote:
>> >> >Aloha,
>> >> >
>> >> >Hello again.  Having succeeded with getting madwifi.stripped to run my
>> >> >wireless as a station, I'm now trying to get it to work as a access
>> point.
>> >> >However, when I use the gen_config_ap.pl script, the router fails to
>> >> >initialize.  The script is using rather normal arguments (ath0, test, 4).
>>  A
>> >> >cat on /click/errors comes up with this:
>> >> >
>> >> >line 74: While configuring 'bs :: BeaconScanner':
>> >> >  AvailableRates element is not provided or not a AvailableRates
>> >> >Router could not be initialized!
>> >> >
>> >> >Line 74 from the config file is this one:
>> >> >
>> >> >management_cl [4] -> bs :: BeaconScanner() ->  Discard;
>> >> >
>> >> >Taking a look at the source for the BeaconScanner element, it looks like
>> it
>> >> got
>> >> >changed but the script didn't.  I think it's now looking for some
>> arguments
>> >> >that aren't provided in the script.  I'm not totally sure, but that's my
>> >> >suspicion.  If this is the problem, how can it be corrected?  If not,
>> what
>> >> else
>> >> >could be the culprit?  As always, any assistance is greatly appreciated.
>> >> >Thanks,
>> >> >
>> >> >-Nels
>_______________________________________________
>click mailing list
>click at amsterdam.lcs.mit.edu
>https://amsterdam.lcs.mit.edu/mailman/listinfo/click