[mc1322x] 6LowPAN data payload

Daniel Berenguer dberenguer at usapiens.com
Mon Jul 19 10:32:17 EDT 2010


Thanks a lot guys!

Mar was right, Building tunslip6 again from the latest sources solved
the problem. Now, border-router seems to work as expected, even with
packet fragmentation enabled :-)

Delays between packets are now much smaller so comms seem to work faster.

Daniel.


On 18 July 2010 17:40, Mariano Alvira <mar at devl.org> wrote:
> I had this problem a while ago --- there were some changes to tunslip6
> to send the prefix. Try deleting tunslip6 and making a new one from
> the current source.
>
> -Mar.
>
>
> On Sun, Jul 18, 2010 at 01:54:51PM +0200, Joakim Eriksson wrote:
>> Ok, so it does not stop waiting when you start tunslip6?
>>
>> Tunslip is supposed to send a prefix to the border-router, and
>> then it should start.
>>
>> Best regards,
>> -- Joakim Eriksson, SICS
>>
>> Daniel Berenguer skrev 2010-07-18 09:34:
>>> OK, so should border-router wait for the prefix until tunslip6 is
>>> started... ? This is not the case.
>>>
>>> Daniel.
>>>
>>>
>>> On 18 July 2010 00:52, Joakim Eriksson<joakime at sics.se>  wrote:
>>>> Hello Daniel,
>>>>
>>>> The prefix is for enabling autoconfiguration of a full IPv6 address
>>>> (without hardcoding anything) for all nodes connecting to the network
>>>> and to the border-router.
>>>>
>>>> the full IP is:<prefix>:<802.15.4 MAC address>.
>>>>
>>>> You will set the prefix if you start the tunslip6 program.
>>>> (or you hard-code a prefix instead).
>>>>
>>>> Best regards,
>>>> -- Joakim Eriksson, SICS
>>>>
>>>> Daniel Berenguer skrev 2010-07-18 00:45:
>>>>>
>>>>> Thanks Mar and George.
>>>>>
>>>>> I've finally tested udp-server on my first Econotag with udp-client on
>>>>> my second Econotag and everything seems to work just fine. Then,
>>>>> uploading border-router into the first Econotag (instead of
>>>>> udp-server) disgraces the whole thing:
>>>>>
>>>>> slip_input_callback is never called after calling slip_send() from
>>>>> request_prefix() in border-router.c
>>>>> As result, border-router gets blocked in the "Request prefix" loop so
>>>>> it will even never respond to a "global repair" button press.
>>>>>
>>>>> PROCESS_THREAD(border_router_process, ev, data)
>>>>> {
>>>>> ...
>>>>>    /* Request prefix until it has been received */
>>>>>    while(!prefix_set) {
>>>>>      etimer_set(&et, CLOCK_SECOND);
>>>>>      request_prefix();
>>>>>      PROCESS_WAIT_EVENT_UNTIL(etimer_expired(&et));
>>>>>    }
>>>>> ...
>>>>>
>>>>> What is the purpose of "requesting the prefix" in border-router?
>>>>> Sorry, I know this is an IPv6-related question but I've found very
>>>>> little information in the net.
>>>>>
>>>>> Daniel.
>>>>>
>>>>>
>>>>> On 15 July 2010 23:31, George Lu<lu at goodxense.com>    wrote:
>>>>>>
>>>>>>
>>>>>> On Thu, Jul 15, 2010 at 2:22 PM, Mariano Alvira<mar at devl.org>    wrote:
>>>>>>>
>>>>>>> Do you guys know about the global repair button? Pressing S3, KBI0-4
>>>>>>> button does "Initiating global repair", which should refresh the
>>>>>>> routes etc...
>>>>>>>
>>>>>>
>>>>>> I had tried the global repair button a while ago. Also had a modified
>>>>>> rpl-border-router.c that would initiate global repair periodically. I
>>>>>> didn't
>>>>>> find it effective. But then that was before the big improvement that came
>>>>>> with increasing NUM_PACKETS to 32.
>>>>>> George
>>>>>
>>>>> _______________________________________________
>>>>> mc1322x mailing list
>>>>> mc1322x at devl.org
>>>>> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>>>>
>>>>
>>>> _______________________________________________
>>>> mc1322x mailing list
>>>> mc1322x at devl.org
>>>> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>>>>
>>>
>>> _______________________________________________
>>> mc1322x mailing list
>>> mc1322x at devl.org
>>> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>>
>>
>> _______________________________________________
>> mc1322x mailing list
>> mc1322x at devl.org
>> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>
> _______________________________________________
> mc1322x mailing list
> mc1322x at devl.org
> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>



More information about the mc1322x mailing list