[Nfd-dev] weird behaviour: producer in the presence of repo (another producer)
Lan Wang (lanwang)
lanwang at memphis.edu
Sun Dec 16 10:46:45 PST 2018
On Dec 15, 2018, at 10:35 PM, Gusev, Peter <peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>> wrote:
Yes, those in the original email were copied from terminal and was another run. I copied the blocks of repeated output. So after the line you asked about, it was again message about the received incoming interest.
The interest was sent to the producer, or at least a PIT entry was created (face 270). Did the producer actually get the interest? Did it respond?
It definitely looks like there’s no producer app that replies with data, but there is.
I don’t touch producer, but once the repo is killed and consumer is restarted, it receives the data back. This can be seen in the logs provided, at the end.
This log doesn’t even forward the interest to the producer, as there’s no next hop. Something was not right.
I suggest that you rerun this experiment and recapture both the terminal output and the nfd logs. Also some logs of the producer app (whether the interest was received) would be useful.
Lan
Thanks,
--
Peter Gusev
peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>
+1 213 5872748<tel:+1%20213%205872748>
peetonn_ (skype)
Software Engineer/Programmer Analyst @ REMAP UCLA
Video streaming/ICN networks/Creative Development
On Dec 15, 2018, at 6:34 PM, Lan Wang (lanwang) <lanwang at memphis.edu<mailto:lanwang at memphis.edu>> wrote:
The logs you attached have different timestamps than the ones you quoted in the email.
In the producer log, you have
1544831132.769269 DEBUG: [nfd.BestRouteStrategy2] /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%AE%F1%DF%16/t/_latest?ndn.MustBeFresh=true&ndn.InterestLifetime=2000&ndn.Nonce=2938754063 from=264 noNextHop
It seems that there is no application receiving this data?
——
In the quoted log, it will be interesting to see what’s after "1544826114.660746 DEBUG: [nfd.BestRouteStrategy2] /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest?ndn.MustBeFresh=true&ndn.InterestLifetime=2000&ndn.Nonce=2176110425 from=264 newPitEntry-to=270”. But you didn’t include those.
Lan
On Dec 14, 2018, at 6:08 PM, Gusev, Peter <peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>> wrote:
please, see the nfd.log files for all three attached
Thanks,
--
Peter Gusev
peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>
+1 213 5872748
Research Scholar @ REMAP UCLA<http://remap.ucla.edu/>
Video streaming/ICN networks/Creative Development
On Dec 14, 2018, at 2:55 PM, Gusev, Peter <peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>> wrote:
sorry, the diagram got skewed.
all three nodes are connected to the same LAN network segment and I use ethernet multicast faces when registering routes. so it’s NDN over Layer2
Thanks,
--
Peter Gusev
peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>
+1 213 5872748
Research Scholar @ REMAP UCLA<http://remap.ucla.edu/>
Video streaming/ICN networks/Creative Development
On Dec 14, 2018, at 2:43 PM, Gusev, Peter <peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>> wrote:
Hi devs,
I have the following setup:
```
PRODUCER REPO CONSUMER
+ + +
| | |
+——-[ LAYER 2 (multicast) ]——+
```
where producer serves prefixes (from nfd-status, so has other routes too):
```
/touchdesigner/lores nexthops={faceid=270 (cost=0)}
/touchdesigner/lores/KEY nexthops={faceid=270 (cost=0)}
/touchdesigner/KEY nexthops={faceid=270 (cost=0)}
```
repo (think, another producer) serves prefixes:
```
/eb/repo/control routes={nexthop=274 origin=app cost=0 flags=child-inherit expires=never}
/eb/repo/control/status routes={nexthop=274 origin=app cost=0 flags=none expires=never}
/touchdesigner/lores/ndnrtc/%FD%03/video/s routes={nexthop=274 origin=app cost=0 flags=child-inherit expires=never}
/touchdesigner routes={nexthop=256 origin=static cost=0 flags=child-inherit expires=never}
/eb routes={nexthop=256 origin=static cost=0 flags=child-inherit expires=never}
```
all three connected to same unmanaged switch and NFD on consumer and repo are setup to use multicast faces for the NICs connected to the switch.
the weird behaviour I’m having is the following
when both REPO and PRODUCER are present, consumer (ndnrtc) is not able to bootstrap video stream because it receives timeout for bootstrap interest:
CONSUMER NFD:
```
1544826176.112756 DEBUG: [nfd.Forwarder] onIncomingInterest face=274 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826176.112772 TRACE: [nfd.NameTree] lookup(/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest, 9)
1544826176.112871 TRACE: [nfd.NameTreeHashtable] insert /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest hash=5592790012398206243 bucket=291
1544826176.112883 DEBUG: [nfd.ContentStore] find /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest L
1544826176.112918 DEBUG: [nfd.Forwarder] onContentStoreMiss interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826176.112935 DEBUG: [nfd.Forwarder] onOutgoingInterest face=261 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826176.113021 DEBUG: [nfd.BestRouteStrategy2] /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest?ndn.MustBeFresh=true&ndn.InterestLifetime=2000&ndn.Nonce=3072391821 from=274 newPitEntry-to=261
<after 2 seconds - repeats>
1544826178.112976 DEBUG: [nfd.Forwarder] onIncomingInterest face=274 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
```
PRODUCER NFD:
```
1544826114.655911 DEBUG: [nfd.Forwarder] onInterestFinalize interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest unsatisfied
1544826114.655933 TRACE: [nfd.NameTreeHashtable] erase /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest hash=5592790012398206243 bucket=3363
1544826114.660585 DEBUG: [nfd.Forwarder] onIncomingInterest face=264 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826114.660592 TRACE: [nfd.NameTree] lookup(/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest, 9)
1544826114.660673 TRACE: [nfd.NameTreeHashtable] insert /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest hash=5592790012398206243 bucket=3363
1544826114.660680 DEBUG: [nfd.ContentStore] find /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest L
1544826114.660706 DEBUG: [nfd.Forwarder] onContentStoreMiss interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826114.660722 DEBUG: [nfd.Forwarder] onOutgoingInterest face=270 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826114.660746 DEBUG: [nfd.BestRouteStrategy2] /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest?ndn.MustBeFresh=true&ndn.InterestLifetime=2000&ndn.Nonce=2176110425 from=264 newPitEntry-to=270
```
REPO NFD:
```
1544826088.610526 DEBUG: [nfd.Forwarder] onIncomingInterest face=256 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826088.610543 TRACE: [nfd.NameTree] lookup(/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest, 9)
1544826088.610657 TRACE: [nfd.NameTreeHashtable] insert /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest hash=5592790012398206243 bucket=1315
1544826088.610676 DEBUG: [nfd.ContentStore] find /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest L
1544826088.610726 DEBUG: [nfd.Forwarder] onContentStoreMiss interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826088.610786 DEBUG: [nfd.Forwarder] onOutgoingInterest face=272 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
1544826088.610858 DEBUG: [nfd.BestRouteStrategy2] /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest?ndn.MustBeFresh=true&ndn.InterestLifetime=2000&ndn.Nonce=212708639 from=256 newPitEntry-to=272
1544826090.610550 DEBUG: [nfd.Forwarder] onInterestFinalize interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest unsatisfied
1544826090.610567 TRACE: [nfd.NameTreeHashtable] erase /touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest hash=5592790012398206243 bucket=1315
1544826090.614587 DEBUG: [nfd.Forwarder] onIncomingInterest face=256 interest=/touchdesigner/lores/ndnrtc/%FD%03/video/s/%FC%00%00%01g%A3%E0%5E%8E/t/_latest
```
once I kill REPO process AND restart consumer process, it is able to bootstrap. however, if REPO is killed BUT consumer is not restarted, it still receives timeouts.
Looking for any insights in what might be happening here.
Thanks,
--
Peter Gusev
peter at remap.ucla.edu<mailto:peter at remap.ucla.edu>
+1 213 5872748
Research Scholar @ REMAP UCLA<http://remap.ucla.edu/>
Video streaming/ICN networks/Creative Development
_______________________________________________
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
<nfd-logs.zip>_______________________________________________
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/20181216/42adf740/attachment-0001.html>
More information about the Nfd-dev
mailing list