View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001769 | GNUnet | DHT service | public | 2011-09-16 10:23 | 2011-09-18 14:59 |
| Reporter | Christian Grothoff | Assigned To | Christian Grothoff | ||
| Priority | high | Severity | minor | Reproducibility | N/A |
| Status | closed | Resolution | fixed | ||
| Product Version | Git master | ||||
| Summary | 0001769: DHT checks peer-bloomfilter for replies | ||||
| Description | But there is no good reason to do this. Nate writes: "I think it's probably just a bug. I cannot imagine any good reason for that check to be there..." | ||||
| Additional Information | We were doing some additions to the DHT code and we stumbled upon a piece of code that we hoped you could clarify. Specifically, when routing result messages, it still uses the bloomfilter before routing to the peer the request came from (gnunet-service-dht.c : 1990-2032) (code simplified): peer_info = find_peer_by_id (&pos->source); if ((GNUNET_NO == GNUNET_CONTAINER_bloomfilter_test (msg_ctx->bloom, &peer_info->id.hashPubKey))) { forward_result_message (msg, peer_info, msg_ctx); else { } Since the bloomfilter can return false positives, this behavior could potentially cause results to get lost. Also, since the bloomfilter is used during the query forwarding, one would expect the return route with "pos->source" to be fully constructed and not contain any loops and therefore the use of the bloomfilter would not be necessary at all in this case. There seems to be no reason for the bloomfilter to be used on the return route, and it could actually be harmful in some rare cases. | ||||
| Tags | No tags attached. | ||||
| related to | 0001776 | closed | Christian Grothoff | dht multipeer test fails SOMETIMES |
|
|
I've added a define "HAVE_REPLY_BLOOMFILTER" in dht.h which effectively comments out all of the reply-BF-related code. I've not completely removed it due to bug 0001776. |
|
|
Code has now been removed. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2011-09-16 10:23 | Christian Grothoff | New Issue | |
| 2011-09-16 10:23 | Christian Grothoff | Status | new => assigned |
| 2011-09-16 10:23 | Christian Grothoff | Assigned To | => Bart Polot |
| 2011-09-16 10:23 | Christian Grothoff | Priority | normal => high |
| 2011-09-16 19:52 | Christian Grothoff | Assigned To | Bart Polot => Christian Grothoff |
| 2011-09-16 20:50 | Christian Grothoff | Note Added: 0004591 | |
| 2011-09-16 20:50 | Christian Grothoff | Relationship added | related to 0001776 |
| 2011-09-16 20:52 | Christian Grothoff | Assigned To | Christian Grothoff => |
| 2011-09-16 20:52 | Christian Grothoff | Status | assigned => confirmed |
| 2011-09-17 22:38 | Christian Grothoff | Assigned To | => Christian Grothoff |
| 2011-09-17 22:38 | Christian Grothoff | Status | confirmed => assigned |
| 2011-09-17 23:05 | Christian Grothoff | Note Added: 0004596 | |
| 2011-09-17 23:05 | Christian Grothoff | Status | assigned => resolved |
| 2011-09-17 23:05 | Christian Grothoff | Resolution | open => fixed |
| 2011-09-18 14:59 | Christian Grothoff | Status | resolved => closed |