[Nfd-dev] Nack receive duplicate on ndnperf

[Junxiao Shi]

Hi Ishita

I looked at ndnperf source code
<https://github.com/Kanemochi/ndnperf/blob/cec1ae7a300a7c9560c1c8b58528516cd0c36ac5/c%2B%2B/client/client.cpp>
and don't see any obvious problem there. Therefore, the problem could be
between topology and forwarding strategy.
Please give complete information to the following questions:

   1. How are the nodes connected? Give a topology diagram. Write it as
   ASCIIart in email body, don't send attachments.
   2. How did you setup the RIB/FIB on each node, for the prefix being used
   in the test?
   3. What forwarding strategy is being used? If you have developed your
   own strategy, give a Gist link. Otherwise, give NFD's exact version
   (including git commit hash) and the strategy name.


Yours, Junxiao

On Wed, Dec 19, 2018 at 10:01 AM Ishita Dasgupta <ishita.dasgupta at gmail.com>
wrote:

> Hi All,
>
> I am trying to download a file using ndnperf client on 10 different nodes
> and some of them ressult in incomplete downloads. On simply running a
> "ndnperf prefix" to test the network stats, I see some of the nodes return
> "Nack receive : Duplicate" after one point.
>
> root at ca39c32f18d3:/ndnperf/c++/client/bin# ./ndnperf -p /edu/ishita
>> Client start with window = 32
>> Server signature type = DigestSha256
>> Server packet size = 8192
>> Wed Dec 19 14:53:14 2018 - 84864 Kbps ( 1326 pkt/s) - latency = 47951 us
>> Wed Dec 19 14:53:16 2018 - 84978 Kbps ( 1333 pkt/s) - latency = 47978 us
>> Wed Dec 19 14:53:18 2018 - 81318 Kbps ( 4737 pkt/s) - latency = 12513 us
>> Wed Dec 19 14:53:20 2018 - 77736 Kbps ( 6633 pkt/s) - latency = 5995 us
>> Wed Dec 19 14:53:22 2018 - 61664 Kbps ( 5262 pkt/s) - latency = 6001 us
>> Nack receive : Duplicate
>
>
> Could this be the reason behind the incomplete file downloads? Does anyone
> know when and why do we see "Nack receive: Duplicate" message on ndnperf?
>
> Regards,
> Ishita
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20181219/c6ac7f03/attachment.html>