NDN project issue tracking system: Issueshttps://redmine.named-data.net/https://redmine.named-data.net/favicon.ico?14759811232024-02-23T19:02:52ZNDN project issue tracking system
Redmine ndn-cxx - Feature #5313 (New): ValidatorConfig: specify signed Interest fieldshttps://redmine.named-data.net/issues/53132024-02-23T19:02:52ZJunxiao Shi
<p>Currently, <strong>ValidatorConfig</strong>, through the usage of <code>ValidationPolicyCommandInterest</code> in its composition, requires every signed Interest to have <strong>SignatureTime</strong> field.<br>
This should be made configurable: the configuration should be able to specify the presence and uniqueness requirements of <strong>SignatureTime</strong> <strong>SignatureSeqNum</strong> <strong>SignatureNonce</strong> fields.</p>
PSync - Task #5312 (New): Improve handling of stale sync data/"no new update"https://redmine.named-data.net/issues/53122024-02-12T21:02:51ZAlexander Lane
<p>Quoting Ashlesh Gawande: "During high update frequency PSync nodes can send data that contain nothing new. This blocks the receiver nodes from syncing leading to high delays." The proposed resolution is to append the number of elements inserted into the IBF to the sync interest to improve reaction to decode failures and dropping the freshness on data containing the full sync dataset to minimize the time potentially unhelpful data occupies the content store.</p>
mGuard - Bug #5307 (New): ck+encrypted data bughttps://redmine.named-data.net/issues/53072024-01-09T21:31:55ZAdam Thieme
<p>publisher.cpp assumes the number of encrypted data segments is the same as the number of ck data segments, which might not always be the case</p>
<p>things might break if we do key or data -size experiments</p>
PSync - Task #5305 (New): Enhance HELLO protol to notify consumers of new prefixes when they are ...https://redmine.named-data.net/issues/53052023-12-21T18:21:54ZAdam Thieme
<p>After getting their initial HELLO data, consumers are not notified of new stream prefixes unless they send regular HELLO interests, which increases bandwidth usage.</p>
ndn-cxx - Feature #5303 (In Progress): Dockerfile for ndn-cxx ecosystemhttps://redmine.named-data.net/issues/53032023-11-27T00:32:16ZJunxiao Shi
<p>Continuation of 15th-ndn-hackathon <em>Containerize all the things!</em> project, write Dockerfile for ndn-cxx and related software.</p>
mGuard - Task #5296 (New): bundle datapoints based on attributeshttps://redmine.named-data.net/issues/52962023-08-31T22:18:13ZAdam Thieme
<p>datapoints should not be added to the same packet unless they have the same attributes. this can be automated</p>
mGuard - Task #5295 (New): Application needs to be able to configure time granularityhttps://redmine.named-data.net/issues/52952023-08-31T22:17:44ZAdam Thieme
<p>time granularity is currently at the second level. we should allow an application to determine what granularity to use, eg. second, 5 seconds, 10 seconds, 30 seconds, 1 minute, 5 minute, etc.</p>
mGuard - Task #5294 (New): add time-based attributes to producer/publisher codehttps://redmine.named-data.net/issues/52942023-08-31T22:16:15ZAdam Thieme
<p>datapoints should be encrypted with the unix timestamp associated with the datapoint. add to publisher.cpp data-adapter.cpp</p>
mGuard - Task #5292 (New): Script to automate mGuard evaluationhttps://redmine.named-data.net/issues/52922023-08-31T21:20:46ZAdam Thieme
<p>we need scripts that parse mGuard logs to determine whether all data was received, if it is decrypted, along with how long it takes for data transfers, encryption, decryption, etc. times. see mguard paper for those metrics</p>
<p>most important right now is a check to see whether all data was transferred. next important is timing</p>
ndn-cxx - Feature #5291 (New): Deterministic ECDSA signinghttps://redmine.named-data.net/issues/52912023-08-23T23:45:50ZDavide Pesavento
<p>Looks like <a href="https://www.openssl.org/blog/blog/2023/09/07/ossl32a1/" class="external">OpenSSL 3.2</a> will finally support deterministic ECDSA signing.</p>
mGuard - Task #5290 (New): create valid trust schema for testbed experimentshttps://redmine.named-data.net/issues/52902023-08-19T00:58:08ZAdam Thieme
<p>there will be multiple trust anchors used in the testbed experiments, and this should be reflected in valid trust schemas</p>
mGuard - Task #5289 (New): test mGuard for possible testbed-related bugshttps://redmine.named-data.net/issues/52892023-08-19T00:57:14ZAdam Thieme
<p>we currently reset nfd, certs, repos, etc. before running tests. we should start removing these resets to see what needs to be changed before we can test on the testbed (where these resets might not be possible).</p>
NFD - Feature #5288 (New): DeadNonceList with hashtable bucketshttps://redmine.named-data.net/issues/52882023-08-13T22:14:27ZJunxiao Shi
<p><a href="https://named-data.net/wp-content/uploads/2023/08/ndn-tr-0077-1-DNL.pdf" class="external">Lessons Learned from Fixing the Dead Nonce List in NFD</a> technical report proposed a new data structure design.<br>
For a DNL with expected longevity <em>L</em>,</p>
<ul>
<li>The DNL consists of <em>M</em> buckets, where each bucket is a hashtable-like data structure.</li>
<li>To insert an entry, the entry is inserted into the bucket at index <em>M-1</em>.</li>
<li>To lookup an entry, the entry is searched among all <em>M</em> buckets.
If it is found in any of the buckets, it exists in the DNL.
Otherwise, it does not exist in the DNL.</li>
<li>Every <em>L/M</em> duration, the bucket at index 0 is discarded, the remaining buckets are shifted down (bucket <em>i+1</em> becomes bucket <em>i</em>), and an empty bucket is created at index <em>M-1</em>.</li>
</ul>
<p>This issue is to implement a baseline version of this data structure design.<br>
Each bucket shall use a regular hashtable data structure <code>std::unorderd_set</code>; exploration of non-deterministic data structures such as Bloom filters is out of scope.<br>
<em>L</em> and <em>M</em> parameters shall be configurable at runtime i.e. from <code>nfd.conf</code>; in case of parameter change during runtime configuration reload, it's OK to fully clear the DNL.</p>
mGuard - Task #5287 (New): Developer Guidehttps://redmine.named-data.net/issues/52872023-08-09T22:09:39ZAdam Thieme
<p>Create documentation to discuss how the mGuard system works, specifically how the individual modules work together. </p>
mGuard - Task #5284 (New): Documentation Updateshttps://redmine.named-data.net/issues/52842023-08-09T22:04:03ZAdam Thieme