NDN project issue tracking system: Issueshttps://redmine.named-data.net/https://redmine.named-data.net/favicon.ico?14759811232019-03-13T23:03:19ZNDN project issue tracking system
Redmine ndn-cxx - Task #4882 (New): Improve "NDN Regular Expression" documentationhttps://redmine.named-data.net/issues/48822019-03-13T23:03:19ZPeter Gusevpeter@remap.ucla.edu
<p>From my personal experience, I wanted to learn how to match typed components using NDN regexes. However, this information can not be found on the <a href="https://docs.named-data.net/ndn-cxx/0.8.1/tutorials/utils-ndn-regex.html" class="external">NDN Regular Expression page</a>.<br>
I think this info, if available, can be added to make this page more useful. Also, pointers to sample code and examples can be very useful too.</p>
<p>As an example, one can use ndnrtc v4 namespace (see attached), to demonstrate using NDN regex(es):</p>
<ul>
<li>stream <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream</code></li>
<li>stream meta <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/_meta</code></li>
<li>stream live meta <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/_live</code></li>
<li>stream latest pointer interest <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/_latest</code></li>
<li>stream latest pointer data <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/_latest/%FE%01%A2</code></li>
<li>stream gop-start <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/_gop/%FE2/start</code> </li>
<li>stream gop-end <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/_gop/%FE2/end</code> </li>
<li>stream frame meta <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/%FE%0DF/_meta</code></li>
<li>stream frame manifest <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/%FE%0DF/_manifest</code></li>
<li>srteam frame payload <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/%FE%0DF/%00%01</code></li>
<li>stream frame parity data <code>/baseprefix/ndnrtc/%FD%04/%FC%03%18%26%60/stream/%FE%0DF/_parity/%00%01</code></li>
</ul>
NFD-android - Bug #4632 (Closed): Cannot add route to existing facehttps://redmine.named-data.net/issues/46322018-06-11T10:39:59ZPeter Gusevpeter@remap.ucla.edu
<p><strong>Description</strong></p>
<ol>
<li>Make sure remote face already exists (for example, udp://10.0.0.1)</li>
<li>Go to "Routes" and add a new route with arbitrary prefix and same remote URI (udp://10.0.0.1).</li>
</ol>
<p><strong>Result</strong><br>
App fails to add route and displays message:<br>
<em>"Error communicating with NFD (Action failed, forwarder returned: 409 Face with remote URI already exists)"</em></p>
<p><strong>Expected result</strong><br>
App adds route to the existing face.</p>
<p><strong>Additional info</strong><br>
See screen capture for this bug -- <a href="https://youtu.be/Ewh8WhEAqkE">https://youtu.be/Ewh8WhEAqkE</a>.</p>
repo-ng - Task #4374 (New): Can't insert segmented datahttps://redmine.named-data.net/issues/43742017-11-01T09:33:59ZPeter Gusevpeter@remap.ucla.edu
<p>Can't insert segmented data into the repo, default configuration (attached).</p>
<p><strong>Steps to reproduce</strong></p>
<ul>
<li><p>Start repo</p>
<pre><code>$ sudo ndn-repo-ng
</code></pre></li>
<li><p>In a separate console, insert a file (50-300KB):</p>
<pre><code>$ ndnputfile /icear/content-provider/1 /icear/content-provider/data/chair chair.jpg
</code></pre></li>
</ul>
<p><strong>Result</strong></p>
<pre><code>ERROR: insert command failed with code 402
</code></pre>
<p><strong>Expected</strong></p>
<p>File is segmented and segments are inserted into the repo.</p>
NDN Control Center - Task #3890 (New): Identity management modulehttps://redmine.named-data.net/issues/38902016-12-14T12:32:15ZPeter Gusevpeter@remap.ucla.edu
<p>Identity management module is responsible for providing signing identities for NDN applications upon request. Signing identity is then used by application to generate long-lived application identity and short-lived instance identities. The module also stores application identity certificate and serves it upon incoming interest.</p>
<blockquote>
<p>For example, in the context of the Flume application, once launched for the first time, application makes an API call for requesting signing identity. The Identity Management Module asks user to choose either existing installed identity or to request a new identity from a list of identity providers.<br>
Let’s assume, Flume application received signing identity /ndn/edu/ucla/remap/peter as a result of API call to NFDcc. Application then generates long-lived app identity /ndn/edu/ucla/remap/peter/flume and short-lived instance identity /ndn/edu/ucla/remap/peter/flume/. App identity certificate is then passed back to the Identity Management Module (via API call) which stores it locally and answers incoming interests that correspond to this certificate name. Instance identity certificate is served by the running application.</p>
</blockquote>
<p><em>REQUIREMENTS</em></p>
<ol>
<li><p>Provide local API calls for applications for:</p>
<p>1.1. requesting new signing identity;<br>
1.2. storing application identity certificate in NFDcc;<br>
1.3. updating application identity certificate.</p></li>
<li><p>Provide user interface for application identity setup:</p>
<p>2.1. allow user to select identity provider for a new identity;<br>
2.2. allow user to select identity from the list of already installed identities.</p></li>
</ol>
ndn-cxx - Feature #3816 (New): KeyRollover for high-rate datahttps://redmine.named-data.net/issues/38162016-10-17T13:14:12ZPeter Gusevpeter@remap.ucla.edu
<p>Need a mechanism for keyrollover for high-rate data (50fps and up) where producer can let consumers fetch newly generated certificate before this certificate is used to sign the <br>
data.</p>
<p><strong>Why it's needed:</strong> If producer will change certificate on-the-fly, consumer will trigger fetching new certificate for real-time data signed by this certificate, which is time-sensitive and certificate may not arrive in time for consumer to verify received packets. Thus, consumer should fetch new certificate prior to it being used by the publisher.</p>
<p>Original e-mail by <a href="mailto:peter@remap.ucla.edu">peter@remap.ucla.edu</a>:</p>
<blockquote>
<p>We talked with Zhehao today about key rollover and he mentioned that there was a proposal from Yingdi. However, I couldn’t find it. Does anyone know where to find it?<br>
Any links on the materials covering the issue would be appreciated as well.</p>
<p>Currently, in ndncon I didn’t plan for writing any special roll-over mechanism/module and wanted to keep things simple (maybe not in a very elegant way - that’s what triggered our discussion about rollover with Zhehao today).</p>
<p>My idea right now is to have ndncon producer to generate new instance certificate every hour. This certificate is installed in the instance keychain, but is not set as a default certificate yet. Media stream packets are always signed with the default certificate. Once new certificate is generated, it is used for signing low-rate data generated by the discovery library. That way, consumers will be able to receive discovery data signed with the new certificate, fetch it according to the certificate chain, verify and have it cached locally. After some delay (60 seconds for instance) producer sets newly generated certificate as a default for the instance keychain and media data packets now will get signed with the new certificate. This won’t trigger fetching certificate on consumer sides as they already have this certificate fetched when they received discovery data earlier. That way, verification won’t trigger delays for the time-sensistive media streaming data.</p>
<p>I’m reaching out for your thoughts on this approach. How does this conflict with the key roll-over concept and whether we should to proceed with this approach (for now/for good)?</p>
</blockquote>
ndnrtc - Task #3534 (New): [ndncon/ndnrtc] Bugfixing and optimizationhttps://redmine.named-data.net/issues/35342016-03-09T17:33:46ZPeter Gusevpeter@remap.ucla.edu
<p>Parent task for NDN-RTC and <em>ndncon</em> bugfixing and optimization</p>
ndnrtc - Task #3532 (Closed): [ndnrtc-oi] Design consumer analysis dashboardhttps://redmine.named-data.net/issues/35322016-03-09T17:32:08ZPeter Gusevpeter@remap.ucla.edu
<p>Need to design a dashboard for analyzing performance of individual consumers. Such a dashboard should allow for easier troubleshooting and <em>ndncon</em> bugfixing.<br>
Some of the possible metrics that can be included:</p>
<ul>
<li><em>ndncon</em> states</li>
<li>buffer levels</li>
<li>interest rate</li>
<li>data rate</li>
<li>DRD and lambda</li>
<li>number of streams fetched</li>
</ul>
ndnrtc - Task #3508 (Closed): [ndncon] Chronosync misconfiguration leads to extensive CPU usage f...https://redmine.named-data.net/issues/35082016-03-04T11:20:47ZPeter Gusevpeter@remap.ucla.edu
<p>Observed behavior:</p>
<ol>
<li>Start producer(s) on REMAP hub</li>
<li>Start ndncon connected to REMAP</li>
<li>Create chat </li>
<li>Wait for some time</li>
</ol>
<p>Result:<br>
NFD and ndncon CPU usage ramps up to 80% and 100% accordingly.</p>
<p>Problem description by Zhehao:</p>
<blockquote>
<p>Given the current implementation, a problematic case could be when sync broadcast interest brings back data that does not result in digest root update (though this, by design, should not happen), and the same interest would be sent again, bringing back the same data thus causing the same interest to be sent again with no interval, and quickly flooding (only) the local NFD. The spike would go away after one sync data times out.<br>
For example, in ChronoSync2013 in ndn-cpp's onData, if the update call does not cause update in root digest, the same interest gets issued again instantly. Similar logic exists in my discovery.<br>
Our previous (deterministic) spike issue, caused by unintended namespace conflict, is the same thing: essentially a different application is able to answer a sync interest with data that does not make sense, thus causing the sync interest sender to instantly retransmit the same interest, and floods its local NFD.<br>
(So as a side note, if anyone in the call today's using a version of NdnCon before your update, this spike should happen for someone else in the call; assuming that everyone has the update, I'll still need to figure out the exact scenario that can cause this.)</p>
</blockquote>
<p>The issue was considered to be resolved with this <a href="https://github.com/remap/ndncon/commit/2c44bf6ec3b34a7c25a2a10d54b3e005fc8e8887" class="external">commit</a>. Though it was observed again right before the 3/2/2016 seminar. Need to check again.</p>
ndnrtc - Feature #2713 (Abandoned): [NdnCon] Provide an indicator of statistics uploadinghttps://redmine.named-data.net/issues/27132015-03-30T13:01:28ZPeter Gusevpeter@remap.ucla.edu
<p>There should be an indicator which tells the progress of uploading statistics onto the server somewhere in UI.</p>
ndnrtc - Bug #2708 (Resolved): [NdnCon] CPU load improvementshttps://redmine.named-data.net/issues/27082015-03-30T09:34:30ZPeter Gusevpeter@remap.ucla.edu
<p>Need to test NdnCon for big conversations and try to reduce CPU load by optimizing the code.</p>
<p>From JeffB:</p>
<blockquote>
<p>High load for big conversations - Anything we can do? Where is CPU load taken primarily now?* Can we switch to only fetching keyframes for thumbnails (I.e., videos that are not the main chat window)</p>
</blockquote>
ndnrtc - Feature #2706 (Resolved): [NdnCon] Check stream status updates https://redmine.named-data.net/issues/27062015-03-30T09:31:26ZPeter Gusevpeter@remap.ucla.edu
<p>From JeffB:</p>
<blockquote>
<p>After closing a stream, the status updates (green/amber led) doesn't seem correct</p>
</blockquote>
ndnrtc - Bug #2703 (New): [NdnCon] Make possible to choose different than system audio input.https://redmine.named-data.net/issues/27032015-03-30T09:23:05ZPeter Gusevpeter@remap.ucla.edu
<p>Currently, NdnCon allows choosing any audio input while configuring media streams, however, only default system input is used when publishing. This should be fixed and audio should be published from the source chosen in preferences.</p>
ndnrtc - Feature #2689 (Resolved): [NdnCon] Status for fetching audio streams.https://redmine.named-data.net/issues/26892015-03-23T12:40:15ZPeter Gusevpeter@remap.ucla.edu
<p>Provide status indicator for audio streams (like in video streams).</p>
ndnrtc - Bug #2688 (Resolved): [NdnCon] Adding new userhttps://redmine.named-data.net/issues/26882015-03-23T12:38:54ZPeter Gusevpeter@remap.ucla.edu
<p>After new user has been added, one hsould click away and chose this user again to start fetch from it. </p>
<p>This "click-away" should be eliminated. </p>
ndnrtc - Feature #2686 (Closed): [NdnCon] NdnCon takes too much time for closing/stop publishing.https://redmine.named-data.net/issues/26862015-03-23T12:34:14ZPeter Gusevpeter@remap.ucla.edu
<p>From Josh:</p>
<blockquote>
<p>often when I end the conversation, NndCon takes a long<br>
time to complete. Sometimes a minute or more such that it seems<br>
I might have to "Force Quit" the application. I have learned to be<br>
more patient but I can imagine users feeling that NdnCon has hung<br>
and killing manually.</p>
</blockquote>