[Winpcap-users] Strange behavior of pcap_next_ex

[Jens Munk]

Thank you so much. Now it works and give full speed.

I have an additional question. The data can be pulled out with both 
"pcap_next_ex" and a call back function loaded with "pcap_loop". Is 
there any advantage of one over the other? My main concern is lost 
packages. I do really, really need to get them all.

Jens.


>>
>> In my application, I am using pcap_sendpacket to send a raw packagage to
>> a networked microcontroller that is within a millisecond or so sending
>> another raw package as a reply (old microcontroller -> not IP), and I
>> can verify with the WireShark that this is working just fine. However,
>> the behavior of pcap_next_ex is not the way I expect. Due to its
>> blocking nature I have it sitting in a separate thread, so that should
>> be OK, but it doesn't return anything before it sees some kind of an IP
>> message like an ARP. Then, however, it keeps returning all the raw
>> messages since the last appearance of an IP message.
>>
>> Is this the way it is supposed to be or am I doing something wrong?
>>
>
> Please remember that there is a parameter called mintocopy. 
> pcap_next_ex will not return until
> - the timeout elapses (timeout that you chose in pcap_open_live)
> - at least mintocopy bytes have been received (default value = 
> 16000bytes)
> (whatever happens first).
>
> I would try to change the mintocopy value (pcap_setmintocopy) to a 
> very small value and see what happens.
>
> Have a nice day
> GV
>
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: j_munk.vcf
Type: text/x-vcard
Size: 321 bytes
Desc: not available
Url : http://www.winpcap.org/pipermail/winpcap-users/attachments/20081205/15eb1287/j_munk.vcf