1 This file lists minor work items (also possibly called "known bugs").
2 We are not tracking them in Mantis yet since there are too many and no
3 sane end-user should care about this codebase yet anyway.
7 - only connect() sockets that are ready (select()) [Nils]
8 [On W32, we need to select after calling socket before
11 - transport_api: support forcing disconnects through low quotas!
12 (required for working F2F support!)
13 - API: consider having core provide deadline information for each message
14 (likely important for DV plugin which wants to loop back!)
15 - implement transport API to pretty-print transport address
16 + transport_api extension (API extension!)
17 + service-transport extension (protocol extension)
18 - add calls to statistics in various places
19 - implement gnunet-transport (transport configurator / tester)
20 - UPnP-based IP detection
21 (Note: build library always, build service when libxml2/etc. are available)
22 - instantly filter addresses from *other* peers that
23 are *equal* to our own address + port (i.e., localhost:2086). We
24 no longer filter those for outgoing (helps with loopback testing
25 and keeps the code clean), but we should filter strictly *impossible*
26 incoming addresses! This is for efficiency, not correctness.
27 - We currently are happy to take any address told to us in a WELCOME
28 to our set of addresses; we should have some minimal threshold-based
29 scheme, limiting both the total number of addresses that we accept
30 this way as well as requiring multiple confirmations; also, we
31 should possibly try to confirm that the given address works for
32 us ourselves (loopback-style) before adding it to the list
34 + we may be able to simplify WELCOME messages (no need to add
35 addresses there anymore, but may help to learn them there anyway...).
36 + we probably want some kind of voting/counting for learning IP addresses
37 (maybe including IP addresses in ads proportional to how often others
38 report them? we at least need some protection against >64k HELLOs!),
39 + provide a way to give the user a list of "learned" IP addresses and
40 a way to easily "veto" addresses off the list!
41 => If MiM attacker uses vetoed address, blacklist the specific IP for
42 the presumed neighbour!
43 - not sure current way of doing ACKs works well-enough
44 with unreliable transports where the ACK maybe lost;
45 the "is_new" check would then possibly prevent future
46 ACKs to be delivered, all while we're happily
47 receiving messages from that peer! Worse, the other
48 peer won't generate another ACK since it thinks we're
49 connected just fine...
51 + How necessary is ACKing in the first place? (alternatives?)
52 + Should we transmit ACKs in response to every HELLO? (would that
53 fully address the problem?)
54 - latency measurements implemented in the transport
55 plugins makes it only work for bi-di transports
56 and results in code replication
57 - should latency be included in the ReceiveCallback and
58 NotifyConnect or passed on request?
59 - FIXME's with latency being simply set to 0 in a few places
60 - [./transport/gnunet-service-transport.c:173]: (style) struct or union member 'TransportPlugin::rebuild' is never used
61 - [./transport/plugin_transport_tcp.c:391]: (style) struct or union member 'Plugin::address_update_task' is never used
64 - [./fs/gnunet-service-fs.c:208]: (style) struct or union member 'LocalGetContext::results_bf_size' is never used
65 - [./fs/gnunet-service-fs.c:501]: (style) struct or union member 'PendingRequest::used_pids_size' is never used
66 - [./fs/gnunet-service-fs.c:654]: (style) struct or union member 'ConnectedPeer::last_client_replies' is never used
67 - [./fs/gnunet-service-fs.c:669]: (style) struct or union member 'ConnectedPeer::avg_delay' is never used
68 - [./fs/gnunet-service-fs.c:675]: (style) struct or union member 'ConnectedPeer::avg_priority' is never used
69 - [./fs/gnunet-service-fs.c:688]: (style) struct or union member 'ConnectedPeer::pending_requests' is never used
70 - [./fs/gnunet-service-fs.c:694]: (style) struct or union member 'ConnectedPeer::last_p2p_replies_woff' is never used
71 - [./fs/gnunet-service-fs.c:700]: (style) struct or union member 'ConnectedPeer::last_client_replies_woff' is never used
74 - [./topology/gnunet-daemon-topology.c:94]: (style) struct or union member 'PeerList::last_hello_sent' is never used
77 - auto-generate "defaults.conf" using gnunet-setup from "config.scm"
78 - integrate all options into "config.scm"
79 - change config-file writing to exclude options set to default values
81 - implement exponential back-off for service restarts
82 - better tracking of which config changes actually need to cause process restarts by ARM.
83 - have way to specify dependencies between services (to manage ARM restarts better)
85 - test case (test_core_api) hangs for a while (some timeout task not killed somewhere?)
86 - [./core/gnunet-service-core.c:469]: (style) struct or union member 'Neighbour::message_queue_size' is never used
87 - [./core/test_core_api_start_only.c:50]: (style) struct or union member 'PeerContext::id' is never used
90 - Better SSL-support for MHD
93 - active reply route caching design & implementation of service,
96 - consider changing API for peer-group termination to
97 call continuation when done
100 - 'server' uses 'GNUNET_PEERINFO_iterate', should probably switch to notification API
101 (for more instant / up-to-date hostlists at lower cost) [OPTIMIZATION]