[Nfd-dev] Help needed with debugging duplicate Nonce

[Ashlesh Gawande (agawande)]

Alright, thanks.


Can you please tell me how to deduce PIT entry contents by looking at the NFD log?


Ashlesh

________________________________
From: Junxiao Shi <shijunxiao at email.arizona.edu>
Sent: Tuesday, August 9, 2016 11:50:47 PM
To: Ashlesh Gawande (agawande)
Cc: Lan Wang (lanwang); nfd-dev at lists.cs.ucla.edu
Subject: Re: [Nfd-dev] Help needed with debugging duplicate Nonce

Hi Ashlesh

This typically means either the design or the implementation is wrong. You may find design problems through theoretical analysis; you may find implementation problems through unit testing.
Also, don't combine two different features into the same commit. They need to be in two separate commits, as they don't depend on each other.

Yours, Junxiao

On Tue, Aug 2, 2016 at 2:19 PM, Ashlesh Gawande (agawande) <agawande at memphis.edu<mailto:agawande at memphis.edu>> wrote:

Hi Junxiao


I am trying to test multicast v2 (https://gerrit.named-data.net/#/c/3013/1) before I push the code review ready patches to resolve our issue.

(Issue 3176 and issue 2602).


Number of duplicate NACKs have increased.

For 3 node topology:

NACKs increase from 58 (v1) to 132 (v2) for 30 second convergence time.


For a 58 node topology:

In multicast v2, NACKs increase from 913073 (v1) to 5624592(v2) and topology does not converge in 30 seconds as it does with v1 (most of them are duplicate NACK). The topology does not converge even in 3 minutes. The increased NACK processing seems to take the blame for increased convergence time.


I have attached the 3 node topology test with ndnump and logs.

Maybe I am doing something wrong in the patch. Could you please take a look?


Thanks

Ashlesh

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20160810/f3098d35/attachment.html>