[Nfd-dev] strange results

Sun Mar 1 10:28:17 PST 2015

On Mar 1, 2015, at 12:23 PM, Lan Wang (lanwang) <lanwang at memphis.edu<mailto:lanwang at memphis.edu>> wrote:

On Mar 1, 2015, at 11:42 AM, Lixia Zhang <lixia at cs.ucla.edu<mailto:lixia at cs.ucla.edu>> wrote:

On Feb 28, 2015, at 10:28 PM, Lan Wang (lanwang) <lanwang at memphis.edu<mailto:lanwang at memphis.edu>> wrote:

In case you are wondering why this is a problem.  Look at the UMich node.  It first forwarded the Interest to CSU, but that one didn't bring back the data (perhaps it was correctly suppressed).  However, its second Interest to UIUC did bring back data (which should not have happened).  So from that point on, it's going to use UIUC instead of the better path CSU to forward Interest to CAIDA.  We did observe this problem.

Lan

Lan, I do not know exactly how your ping measurement works, but from the above description, it seems that,

- for each ping interest with nonce=N, for any node to get duplicates, there ought to be a branching point along its (the interest’s) path

- UMich is not this point, since CSU already detected the interest from UMich as duplicate (So CSU node has seen interest(N) from some other neighbor)

UMich was on one of the several branches that didn't get the data due to duplicate suppression, so it retransmitted the interest over a different  (worse) path.   This duplicate interest hit CSU's cache and was able to bring data back due to incorrect duplicate suppression. From this point on, Umich switched to the worse path.

Junxiao has a good example with simpler topology at http://redmine.named-data.net/issues/2592

In note 2.

Lan

Lan

- this branching point (could be the origin consumer?) should be able to tell a good path from a worse path, right?

Lixia

On Mar 1, 2015, at 12:02 AM, Lan Wang (lanwang) <lanwang at memphis.edu<mailto:lanwang at memphis.edu>> wrote:

Hi all,

We're seeing strange results from our latest experiments.  Below is an ndndump trace of one ndnping packet from UCLA to CAIDA.  We ran the experiments using mini-ndn.  I'm also attaching a png file that shows where the Interests and Data packets were forwarded (white is the original transmission of the Interest,  yellow and blue indicate retransmissions of this Interest with the same nonce by different routers).

[Lans-MacBook-Air:test-ucla-ping-cpu-alloc-withNLSRlog-timing/ls/faces-2] lanwang1% grep "/ndn/edu/caida/ping/251436634" */dump* | sort -t ":" -k 2 | more
ucla/dump.0ucla:1425075159.119036 From: 1.0.0.42, To: 1.0.0.41, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
csu/dump.0csu:1425075159.119112 From: 1.0.0.42, To: 1.0.0.41, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
csu/dump.4csu:1425075159.140391 From: 1.0.0.29, To: 1.0.0.30, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
memphis/dump.1memphis:1425075159.140488 From: 1.0.0.29, To: 1.0.0.30, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
ucla/dump.1ucla:1425075159.164108 From: 1.0.0.73, To: 1.0.0.74, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
umich/dump.0umich:1425075159.164185 From: 1.0.0.73, To: 1.0.0.74, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
umich/dump.2umich:1425075159.179542 From: 1.0.0.50, To: 1.0.0.49, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
csu/dump.6csu:1425075159.179620 From: 1.0.0.50, To: 1.0.0.49, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
memphis/dump.2memphis:1425075159.181041 From: 1.0.0.18, To: 1.0.0.17, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
caida/dump.4caida:1425075159.181138 From: 1.0.0.18, To: 1.0.0.17, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
caida/dump.4caida:1425075159.224372 From: 1.0.0.17, To: 1.0.0.18, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
memphis/dump.2memphis:1425075159.224444 From: 1.0.0.17, To: 1.0.0.18, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
memphis/dump.1memphis:1425075159.245821 From: 1.0.0.30, To: 1.0.0.29, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
csu/dump.4csu:1425075159.245887 From: 1.0.0.30, To: 1.0.0.29, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
csu/dump.0csu:1425075159.262268 From: 1.0.0.41, To: 1.0.0.42, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
ucla/dump.0ucla:1425075159.262344 From: 1.0.0.41, To: 1.0.0.42, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
umich/dump.4umich:1425075159.330700 From: 1.0.0.82, To: 1.0.0.81, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
uiuc/dump.2uiuc:1425075159.330771 From: 1.0.0.82, To: 1.0.0.81, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
uiuc/dump.0uiuc:1425075159.345166 From: 1.0.0.46, To: 1.0.0.45, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
csu/dump.5csu:1425075159.345243 From: 1.0.0.46, To: 1.0.0.45, Tunnel Type: UDP, INTEREST: /ndn/edu/caida/ping/251436634?ndn.MustBeFresh=1&ndn.Nonce=251436634, size: 48
csu/dump.5csu:1425075159.359534 From: 1.0.0.45, To: 1.0.0.46, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
uiuc/dump.0uiuc:1425075159.359617 From: 1.0.0.45, To: 1.0.0.46, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
uiuc/dump.2uiuc:1425075159.365972 From: 1.0.0.81, To: 1.0.0.82, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
umich/dump.4umich:1425075159.366036 From: 1.0.0.81, To: 1.0.0.82, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
umich/dump.0umich:1425075159.399394 From: 1.0.0.74, To: 1.0.0.73, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392
ucla/dump.1ucla:1425075159.399522 From: 1.0.0.74, To: 1.0.0.73, Tunnel Type: UDP, DATA: /ndn/edu/caida/ping/251436634, size: 392

Since all these Interests were duplicates of the original Interest, there should be only one copy of the Data packet going back to the ndnping client, but it seems that the Data came back in two paths:

- CAIDA -> Memphis -> CSU -> UCLA
- CAIDA -> Memphis -> CSU -> UIUC -> UMich -> UCLA

Seems that CSU was not doing duplicate Interest suppression correctly.   The Interest from UIUC to CSU received at time 1425075159.345243 should have been dropped because it's a duplicate of the earlier Interest from UCLA to CSU received at time 1425075159.119112, but it was not and instead brought data back.  The time difference between the two Interests is 226ms.  I'm not sure how the duplicate suppression is implemented.  Someone more familiar with the forwarding pipeline please help explain the above trace.  Thanks.

Lan
<ucla-ip.png>
_______________________________________________
Nfd-dev mailing list
Nfd-dev at lists.cs.ucla.edu<mailto:Nfd-dev at lists.cs.ucla.edu>
http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev

_______________________________________________
Nfd-dev mailing list
Nfd-dev at lists.cs.ucla.edu<mailto:Nfd-dev at lists.cs.ucla.edu>
http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20150301/ac9f60a3/attachment.html>