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