1 PHASE #1: (Goal: settle key design questions)
4 * API review: should "bpm" and "last_activity" arguments be
5 included in the ClientEventHandler?
7 * Internal: advertising (propagate other peers' HELLOs, find new ones)
8 * Internal: bootstrapping
9 => bootstrap should use plugins, possible multiple at the same time!
12 * consider adding "get_time" to "configuration" API
15 For these functions, it would be nice if we had testcases ("make check")
16 that would cause them to be executed and check that they are working:
17 * gnunet-service-peerinfo:
18 - change_host_trust / flush_trust
20 - discard_hosts_helper / cron_clean_data_hosts
21 * gnunet-service-transport:
22 - try_unvalidated_addresses
23 - lookup_address_callback
24 - lookup_hello_callback
25 - plugin_env_lookup_address
26 - notify_clients_disconnect
27 - list_validated_addresses
29 - disconnect_neighbour
31 * plugin_transport_tcp.c:
33 - tcp_plugin_address_pretty_printer / append_port
34 - tcp_plugin_set_receive_quota
37 - GNUNET_TRANSPORT_set_qutoa / send_set_quota
41 - remove_from_any_list / remove_neighbour
42 - GNUNET_TRANSPORT_notify_transmit_ready_cancel
45 - solicit_traffic / copy_and_free
46 - GNUNET_CORE_peer_configure / produce_configure_message
47 * gnunet-service-core:
50 - handle_client_request_configure
53 - handle_transport_notify_disconnect
56 PHASE #2: (Goal: recover basic core functionality)
59 * implement sqlite-based sqstore/datastore service
60 + implement library (talks to service)
61 + implement service (datastore + talks to plugin)
62 + implement sqlite plugin (talks to DB)
63 * implement sqlite-based dstore services
64 + implement library (talks to service)
65 + implement service (talks to plugin)
66 + implement sqlite plugin (talks to DB)
69 * implement hostlist service (need to bootstrap!)
74 * mysql & postgres backend
77 * UDP backend (need LIBRARY to support (de)fragmentation!)
80 + Testcases for set_quota, timeouts, disconnects, transmit_ready_cancel
81 + Better coverage of gnunet-service-transport (hello validation)
82 + direct test of plugins compliance to plugin API
85 PHASE #3: (Goal: ready for pre-release) [completion-goal: end of 2009]
100 - transport_api: support forcing disconnects through low quotas!
101 - API: consider having core provide priority and possibly
102 deadline information for each message
103 (likely important for DV plugin which wants to loop back!)
104 - implement transport API to pretty-print transport address
105 + transport_api extension (API extension!)
106 + service-transport extension (protocol extension)
107 - add calls to statistics in various places
108 - implement gnunet-transport (transport configurator / tester)
109 - UPnP-based IP detection
110 (Note: build library always, build service when libxml2/etc. are available)
111 - instantly filter addresses from *other* peers that
112 are *equal* to our own address + port (i.e., localhost:2086). We
113 no longer filter those for outgoing (helps with loopback testing
114 and keeps the code clean), but we should filter strictly *impossible*
115 incoming addresses! This is for efficiency, not correctness.
116 - We currently are happy to take any address told to us in a WELCOME
117 to our set of addresses; we should have some minimal threshold-based
118 scheme, limiting both the total number of addresses that we accept
119 this way as well as requiring multiple confirmations; also, we
120 should possibly try to confirm that the given address works for
121 us ourselves (loopback-style) before adding it to the list
123 - not sure current way of doing ACKs works well-enough
124 with unreliable transports where the ACK maybe lost;
125 the "is_new" check would then possibly prevent future
126 ACKs to be delivered, all while we're happily
127 receiving messages from that peer! Worse, the other
128 peer won't generate another ACK since it thinks we're
129 connected just fine...
131 + How necessary is ACKing in the first place? (alternatives?)
132 + Should we transmit ACKs in response to every HELLO? (would that
133 fully address the problem?)
134 - latency measurements implemented in the transport
135 plugins makes it only work for bi-di transports
136 and results in code replication
137 - should latency be included in the ReceiveCallback and
138 NotifyConnect or passed on request?
140 - auto-generate "defaults.conf" using gnunet-setup from "config.scm"
141 - integrate all options into "config.scm"
142 - change config-file writing to exclude options set to default values
144 - implement exponential back-off for service restarts
145 - better tracking of which config changes actually need to cause process restarts by ARM.
146 - have way to specify dependencies between services (to manage ARM restarts better)
148 - have gnunet-peerinfo print actual host addresses again
149 - add option to gnunet-peerinfo to modify trust value
151 - finish postgres implementation; simplify other SQLs using new stats
156 - active reply route caching design & implementation of service,
161 PHASE #4: [completion-goal: mid 2010]
169 * Features eliminated from util:
170 - threading (goal: good riddance!)
171 - complex logging features [ectx-passing, target-kinds] (goal: good riddance!)
172 - complex configuration features [defaults, notifications] (goal: good riddance!)
173 - network traffic monitors (goal: eliminate)
174 - IPC semaphores (goal: d-bus? / eliminate?)
176 - DNS lookup (goal: have async service; issue: still need synchronous resolution in places, current code may not be portable)
177 => code shrunk from 61 files to 34, 22k LOC to 15k LOC, 470k to 330k (with symbols)
178 * New features in util:
180 - service and program boot-strap code
181 * Major changes in util:
182 - more expressive server (replaces selector)
184 - how to integrate scheduler with GTK event loop!
189 * UTIL: 75%, 4914 out of 6463