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 - connection.c:553 fails when "make check" is run!
8 (check if this could be memory corruption).
12 - container_bloomfilter: improve efficiency (see FIXME)
13 - server: inefficient memmove
14 - Windows: use events instead of pipes to signal select()s [Nils]
15 - only connect() sockets that are ready (select()) [Nils]
16 [On W32, we need to select after calling socket before
19 - transport_api: support forcing disconnects through low quotas!
20 (required for working F2F support!)
21 - API: consider having core provide deadline information for each message
22 (likely important for DV plugin which wants to loop back!)
23 - implement transport API to pretty-print transport address
24 + transport_api extension (API extension!)
25 + service-transport extension (protocol extension)
26 - add calls to statistics in various places
27 - implement gnunet-transport (transport configurator / tester)
28 - UPnP-based IP detection
29 (Note: build library always, build service when libxml2/etc. are available)
30 - instantly filter addresses from *other* peers that
31 are *equal* to our own address + port (i.e., localhost:2086). We
32 no longer filter those for outgoing (helps with loopback testing
33 and keeps the code clean), but we should filter strictly *impossible*
34 incoming addresses! This is for efficiency, not correctness.
35 - We currently are happy to take any address told to us in a WELCOME
36 to our set of addresses; we should have some minimal threshold-based
37 scheme, limiting both the total number of addresses that we accept
38 this way as well as requiring multiple confirmations; also, we
39 should possibly try to confirm that the given address works for
40 us ourselves (loopback-style) before adding it to the list
42 + we may be able to simplify WELCOME messages (no need to add
43 addresses there anymore, but may help to learn them there anyway...).
44 + we probably want some kind of voting/counting for learning IP addresses
45 (maybe including IP addresses in ads proportional to how often others
46 report them? we at least need some protection against >64k HELLOs!),
47 + provide a way to give the user a list of "learned" IP addresses and
48 a way to easily "veto" addresses off the list!
49 => If MiM attacker uses vetoed address, blacklist the specific IP for
50 the presumed neighbour!
51 - not sure current way of doing ACKs works well-enough
52 with unreliable transports where the ACK maybe lost;
53 the "is_new" check would then possibly prevent future
54 ACKs to be delivered, all while we're happily
55 receiving messages from that peer! Worse, the other
56 peer won't generate another ACK since it thinks we're
57 connected just fine...
59 + How necessary is ACKing in the first place? (alternatives?)
60 + Should we transmit ACKs in response to every HELLO? (would that
61 fully address the problem?)
62 - latency measurements implemented in the transport
63 plugins makes it only work for bi-di transports
64 and results in code replication
65 - should latency be included in the ReceiveCallback and
66 NotifyConnect or passed on request?
67 - FIXME's with latency being simply set to 0 in a few places
68 - Memory leak (running valgrind --trace-children=yes on test_transport_api:
69 ==28393== 16 bytes in 1 blocks are indirectly lost in loss record 1 of 5
70 ==28393== at 0x4C2260E: malloc (vg_replace_malloc.c:207)
71 ==28393== by 0x52343E3: GNUNET_xmalloc_unchecked_ (common_allocation.c:62)
72 ==28393== by 0x5234389: GNUNET_xmalloc_ (common_allocation.c:53)
73 ==28393== by 0x524458A: GNUNET_NETWORK_socket_create_from_accept (network.c:289)
74 ==28393== by 0x524B2DA: ??? (server.c:332)
75 ==28393== by 0x524A4C7: ??? (scheduler.c:425)
76 ==28393== by 0x524A73D: GNUNET_SCHEDULER_run (scheduler.c:510)
77 ==28393== by 0x524FF8C: GNUNET_SERVICE_run (service.c:1326)
78 ==28393== by 0x405500: main (gnunet-service-transport.c:2645)
80 ==28393== 65,744 (65,728 direct, 16 indirect) bytes in 1 blocks are definitely lost in loss record 5 of 5
81 ==28393== at 0x4C2260E: malloc (vg_replace_malloc.c:207)
82 ==28393== by 0x52343E3: GNUNET_xmalloc_unchecked_ (common_allocation.c:62)
83 ==28393== by 0x5234389: GNUNET_xmalloc_ (common_allocation.c:53)
84 ==28393== by 0x524473E: GNUNET_NETWORK_socket_create_from_accept (network.c:323)
85 (rest of trace identical)
87 - [./transport/gnunet-service-transport.c:173]: (style) struct or union member 'TransportPlugin::rebuild' is never used
88 - [./transport/plugin_transport_tcp.c:391]: (style) struct or union member 'Plugin::address_update_task' is never used
91 - [./fs/gnunet-service-fs.c:208]: (style) struct or union member 'LocalGetContext::results_bf_size' is never used
92 - [./fs/gnunet-service-fs.c:501]: (style) struct or union member 'PendingRequest::used_pids_size' is never used
93 - [./fs/gnunet-service-fs.c:654]: (style) struct or union member 'ConnectedPeer::last_client_replies' is never used
94 - [./fs/gnunet-service-fs.c:669]: (style) struct or union member 'ConnectedPeer::avg_delay' is never used
95 - [./fs/gnunet-service-fs.c:675]: (style) struct or union member 'ConnectedPeer::avg_priority' is never used
96 - [./fs/gnunet-service-fs.c:688]: (style) struct or union member 'ConnectedPeer::pending_requests' is never used
97 - [./fs/gnunet-service-fs.c:694]: (style) struct or union member 'ConnectedPeer::last_p2p_replies_woff' is never used
98 - [./fs/gnunet-service-fs.c:700]: (style) struct or union member 'ConnectedPeer::last_client_replies_woff' is never used
101 - [./topology/gnunet-daemon-topology.c:94]: (style) struct or union member 'PeerList::last_hello_sent' is never used
102 - while running the topology test with valgrind (--trace-children=yes), I get:
104 Nov 03 22:38:57 topology DEBUG I am peer `AJ5M'==4186== Syscall param socketcall.send(msg) points to uninitialised byte(s)
105 ==4186== at 0x4164BF1: send (socket.S:64)
106 ==4186== by 0x404CC1F: transmit_ready (connection.c:1393)
107 ==4186== by 0x4063C3B: run_ready (scheduler.c:451)
108 ==4186== by 0x40640AE: GNUNET_SCHEDULER_run (scheduler.c:575)
109 ==4186== by 0x406090A: GNUNET_PROGRAM_run (program.c:196)
110 ==4186== by 0x804B1CA: main (gnunet-daemon-topology.c:1250)
111 ==4186== Address 0x46e33b0 is 136 bytes inside a block of size 65,664 alloc'd
112 ==4186== at 0x4024C1C: malloc (vg_replace_malloc.c:195)
113 ==4186== by 0x40476F6: GNUNET_xmalloc_unchecked_ (common_allocation.c:61)
114 ==4186== by 0x404768E: GNUNET_xmalloc_ (common_allocation.c:52)
115 ==4186== by 0x404BB22: GNUNET_CONNECTION_create_from_connect (connection.c:887)
116 ==4186== by 0x40460B8: do_connect (client.c:233)
117 ==4186== by 0x404610C: GNUNET_CLIENT_connect (client.c:259)
118 ==4186== by 0x402C6D5: GNUNET_CORE_connect (core_api.c:857)
119 ==4186== by 0x804B118: run (gnunet-daemon-topology.c:1217)
120 ==4186== by 0x4060498: program_main (program.c:80)
121 ==4186== by 0x4063C3B: run_ready (scheduler.c:451)
122 ==4186== by 0x40640AE: GNUNET_SCHEDULER_run (scheduler.c:575)
123 ==4186== by 0x406090A: GNUNET_PROGRAM_run (program.c:196)
131 - auto-generate "defaults.conf" using gnunet-setup from "config.scm"
132 - integrate all options into "config.scm"
133 - change config-file writing to exclude options set to default values
135 - implement exponential back-off for service restarts
136 - better tracking of which config changes actually need to cause process restarts by ARM.
137 - have way to specify dependencies between services (to manage ARM restarts better)
139 - code currently notifies clients about "encrypted" connections being up well before
140 we get the encrypted PONG; sometimes this may be OK (for topology killing
141 unwanted connnections), but of course not in general. I suspect we want
142 to signal on PONG and have topology hook directly into transport to
143 kill plaintext connections before they have a chance to become encrypted
144 (may require minor hack in transport API)
145 - [./core/gnunet-service-core.c:469]: (style) struct or union member 'Neighbour::message_queue_size' is never used
146 - [./core/test_core_api_start_only.c:50]: (style) struct or union member 'PeerContext::id' is never used
149 - have gnunet-peerinfo print actual host addresses again
150 - add option to gnunet-peerinfo to modify trust value
152 - finish postgres implementation; simplify other SQLs using new stats
154 - Better SSL-support for MHD
157 - active reply route caching design & implementation of service,
160 - implement advertising of hostlist URL
161 - implement learning of hostlist URLs
163 - consider changing API for peer-group termination to
164 call continuation when done