2 * HOSTLIST: seems to have NO 'tcp' in it, so cannot have any addresses!? [CG]
3 * FS-acceptance testing [CG]
11 - Get IPv6 hooked up [AK, after April 12th]
12 - change DNS [CG, need DNS]
16 - trust: need *fast* way to check/update trust in peers
17 (async peerinfo would not be right; certainly not with the
20 - scheduler should change OS process priority based on task priority;
21 should make better use of task priorities in general
22 - only connect() sockets that are ready (select()) [Nils]
23 [On W32, we need to select after calling socket before
26 - design & implement new setup tool
28 - good to have for transport/DV evaluation!
30 - write DV API (need to move declarations from dv_api.c to gnunet_dv_service.h!)
31 - implement DV service
32 - implement DV library (looks done)
33 - implement DV transport plugin
35 - implement performance tests
37 - does not seem to work with timeouts (especially if service is not running)
39 - needs more testing (especially F2F topology)
40 - needs to re-try connecting after disconnect (currently, it
41 initially triggers a connection request, but if that connection
42 fails / goes down, it does not retry in a timely fashion;
43 cause seems to be the 'blacklist_after_attempt' being set to 1h,
44 which is rather long -- and should probably be adjusted based on
45 the number of connections / known peers)
46 - If the topology daemon crashes, peers that were put on the
47 blacklist with transport will never be removed from it (until
48 transport service dies); we should use the blacklist notification
49 API to learn about the exact set of blacklisted peers at all times
50 (FIXME: the transport_api implementation of blacklisting
51 also does not work nicely for this since it won't let us know about
52 disconnect-reconnect events and the implicit whitelisting
53 that might happen here; that's not so bad since we will
54 re-blacklist on pre-connect attempts anyway, so this is
55 a minor issue; OTOH, we might want to be more explicit about
56 allowing/forbidding connects on pre-connect to avoid
57 entering connect attempts to just be blacklisted shortly afterwards).
58 - the code uses the term 'blacklist' for both peers that are forbidden
59 to connect (i.e. F2F mode) as well as peers that we currently
60 won't try to actively connect to ourselves (since we just tried);
61 This is confusing. We need two distinct terms (greylist?).
62 - move code to use hash table instead of linked list
63 - instead of periodically discarding blacklisted entries,
64 simply add task that is triggered at the right time (earlier free,
66 - check if new HELLO learned is different from old HELLO
67 before resetting entire state!
69 - support recursive download even if filename is NULL and we hence
70 do not generate files on disk (use temp_filename)
71 - bound parallelism (# fs downloads)
72 - distinguish in performance tracking and event signalling between
73 downloads that are actually running and those that are merely in the queue
74 - persistence support (publish, unindex, search, download)
75 - active migration support (in fs or in datastore or new daemon?)
76 - gnunet-service-fs (hot-path routing, load-based routing, nitpicks)
77 - [gnunet-service-fs.c:208]: member 'LocalGetContext::results_bf_size' is never used
78 - [gnunet-service-fs.c:501]: member 'PendingRequest::used_pids_size' is never used
79 - [gnunet-service-fs.c:654]: member 'ConnectedPeer::last_client_replies' is never used
80 - [gnunet-service-fs.c:669]: member 'ConnectedPeer::avg_delay' is never used
81 - [gnunet-service-fs.c:675]: member 'ConnectedPeer::avg_priority' is never used
82 - [gnunet-service-fs.c:688]: member 'ConnectedPeer::pending_requests' is never used
83 - [gnunet-service-fs.c:694]: member 'ConnectedPeer::last_p2p_replies_woff' is never used
84 - [gnunet-service-fs.c:700]: member 'ConnectedPeer::last_client_replies_woff' is never used
86 + active reply route caching design & implementation of service; gap extension!
88 - API lacks cancellation methods (needed? or is disconnect enough?);
89 may also want to integrate request queuing here instead of gnunet-service-fs_drq.c
91 - use g_main_context_set_poll_func to integrate GTK with GNUnet Scheduler!? (YUCK!)
92 - OR: add scheduler API to enable integration with GTK main loop instead of doing our own select
93 - use g_main_context_pending, g_main_context_query / g_main_context_check / g_main_context_dispatch
94 and NEVER g_main_loop_run (can this be done? might be the clean way to do this! But how
95 to integrate this with "tak_main"? Docu says:
96 "It's OK to use the GLib main loop directly instead of gtk_main(), though it involves
97 slightly more typing. See GMainLoop in the GLib documentation."
98 => so maybe it "just works"?
102 - good to have for DV/DHT evaluation!
104 - implement DHT service
105 - implement DHT library
106 - implement testcases
107 - implement performance tests
111 - need to get rid of synchronous API for service starts (cause all kinds of problems)
112 [=> eliminate for need to tell ARM about service starts most of the time!] [Safey]
113 - better tracking of which config changes actually need to cause process restarts by ARM.
114 - listen for requests to discover dependencies between services (and avoid
115 having to explicitly program start requests)
116 - better crash management (attach debugging support, capture and analyze
117 debug output, detect random vs. deterministic crashes)
120 - datastore reservation (publishing)
121 - search: availability probes
122 - location URIs (publish, search, download)
123 - non-anonymous FS service (needs DHT)
124 + DHT integration for search
125 + CS-DHT-functions (DHT-put of LOC)
126 + P2P-functions (DHT-get)
127 - collection API & tests
128 + gnunet-pseudonym (collection support)
129 - implement FS performance tests
136 * Determine RC bugs and fix those!
138 - modify configuration to allow controlling connections for non-local starts
139 - testbed creation with topology (needs working F2F topology)
141 - implement testcases for distributed testing
142 - test basic peer re-configure
143 - test topology creation
144 - test churn generation
145 - consider changing API for peer-group termination to
146 call continuation when done
148 - finalize API design
151 - integration with transport service
152 * MYSQL database backends: [CG]
158 - expand bibliography
159 - convert documentation pages to books
160 - update books (especially for developers)
161 - create good Drupal theme for GNUnet
162 - make a NICE download page and figure out how to
163 enable developers to publish TGZs nicely
164 - port "contact" page
165 - add content type for "todo" items?
166 * POSTGRES database backends: [CG]
169 * Determine RC bugs and fix those!
173 - SMTP transport backend
174 - HTTPS transport backend
175 + improved HTTPS support in MHD
177 - Implement method of learning our external addresses from
178 other peers; need some kind of threshold-based
179 scheme, limiting both the total number of addresses that we accept
180 this way as well as requiring multiple confirmations; also, we
181 should possibly try to confirm that the given address works for
182 us ourselves (loopback-style) before adding it to the list
183 + we may be able to simplify WELCOME messages (no need to add
184 addresses there anymore, but may help to learn them there anyway...).
185 + we probably want some kind of voting/counting for learning IP addresses
186 (maybe including IP addresses in ads proportional to how often others
187 report them? we at least need some protection against >64k HELLOs!),
188 + provide a way to give the user a list of "learned" IP addresses and
189 a way to easily "veto" addresses off the list!
190 => If MiM attacker uses vetoed address, blacklist the specific IP for
191 the presumed neighbour!
192 - implement gnunet-transport (transport configurator / tester)
193 - UPnP-based IP detection
194 (Note: build library always, build service when libxml2/etc. are available)
196 - Remove KBlocks in gnunet-unindex (see discussion with Kenneth Almquist on gnunet-devs in 9/2009)
197 - Allow checking of presence of search results and/or content via command-line tools
198 (add options to gnunet-search / gnunet-download to limit search to local peer)
200 - expire 'ancient' HELLOs (those without valid addresses AND that
201 we have not 'used' (for their public keys) in a while; need a way
202 to track actual 'use')
203 - make sue we also trigger notifications whenever HELLOs expire
210 - should use hash map to look up sessions
212 - api creates many, many short-lived TCP connections; either some
213 clients should use the API differently or we need to change the
214 API to enable re-use of connections to the service
216 - should use BIO instead of mmap
218 - need to periodically probe latency/transport cost changes & possibly switch transport
219 - instantly filter addresses from *other* peers that
220 are *equal* to our own address + port (i.e., localhost:2086). We
221 no longer filter those for outgoing (helps with loopback testing
222 and keeps the code clean), but we should filter strictly *impossible*
223 incoming addresses! This is for efficiency, not correctness.
224 - should use hash map to look up Neighbours
226 - 'server' uses 'GNUNET_PEERINFO_iterate', should probably switch to notification API
227 (for more instant / up-to-date hostlists at lower cost) [OPTIMIZATION]
229 - check for duplicates on insertion (currently, same content is frequently
230 stored again [seen with KBLOCKS and SBLOCKS]!)
234 - repeatedly resolve hostname and look up interfaces to determine our own IP
235 - [./transport/plugin_transport_tcp.c:391]: (style) struct or union member 'Plugin::address_update_task' is never used (related to issue above)
237 - [./transport/gnunet-service-transport.c:173]: (style) struct or union member 'TransportPlugin::rebuild' is never used (related to TCP not refreshing external addresses?)
239 - add stats (# bytes available, # bytes used, # PUTs, # GETs, # GETs satisfied)
241 - support inline data in directories for recursive file downloads (fs_download)