1 PHASE #1: (Goal: settle key design questions)
4 * hostlist server (MHD-side)
5 * hostlist client (CURL-side); monitoring of number of active connections (to establish need for bootstrapping)
6 * hostlist server URL advertising & learning via P2P
9 * Selecting peers from peerinfo for connects; blacklisting
10 * Managing connections, F2F configuration obedience, rejecting prohibited connections
11 * Forwarding of known HELLOs to neighbours (advertising)
12 [ Inbound HELLOs are processed by transport, right?
13 But what about inbound encrypted HELLOs? ]
16 * Make sure ARM supports daemons (non-service background processes, see hostlist/topology)
19 * improve disk API [Nils]
20 * Windows: use events instead of pipes to signal select()s [Nils]
21 * only connect() sockets that are ready (select())
24 For these functions, it would be nice if we had testcases ("make check")
25 that would cause them to be executed and check that they are working:
26 * gnunet-service-peerinfo:
27 - change_host_trust / flush_trust
29 - discard_hosts_helper / cron_clean_data_hosts
30 * gnunet-service-transport:
31 - try_unvalidated_addresses
32 - lookup_address_callback
33 - lookup_hello_callback
34 - plugin_env_lookup_address
35 - notify_clients_disconnect
36 - list_validated_addresses
38 - disconnect_neighbour
40 * plugin_transport_tcp.c:
42 - tcp_plugin_address_pretty_printer / append_port
43 - tcp_plugin_set_receive_quota
46 - GNUNET_TRANSPORT_set_qutoa / send_set_quota
50 - remove_from_any_list / remove_neighbour
51 - GNUNET_TRANSPORT_notify_transmit_ready_cancel
54 - solicit_traffic / copy_and_free
55 - GNUNET_CORE_peer_configure / produce_configure_message
56 * gnunet-service-core:
59 - handle_client_request_configure
62 - handle_transport_notify_disconnect
65 PHASE #2: (Goal: recover basic file-sharing functionality)
68 * implement sqlite-based sqstore/datastore service
69 + implement library (talks to service)
70 + implement service (datastore + talks to plugin)
71 + implement sqlite plugin (talks to DB)
72 * implement sqlite-based dstore services
73 + implement library (talks to service)
74 + implement service (talks to plugin)
75 + implement sqlite plugin (talks to DB)
78 * implement hostlist service (need to bootstrap!)
83 * have ONE backend working
86 PHASE #3: (Goal: ready for pre-release) [completion-goal: end of 2009]
96 * UDP backend (need LIBRARY to support (de)fragmentation!)
99 + Testcases for set_quota, timeouts, disconnects, transmit_ready_cancel
100 + Better coverage of gnunet-service-transport (hello validation)
101 + direct test of plugins compliance to plugin API
104 * sqlite, mysql & postgres backend
110 - inefficient memmove
112 - transport_api: support forcing disconnects through low quotas!
113 - API: consider having core provide deadline information for each message
114 (likely important for DV plugin which wants to loop back!)
115 - implement transport API to pretty-print transport address
116 + transport_api extension (API extension!)
117 + service-transport extension (protocol extension)
118 - add calls to statistics in various places
119 - implement gnunet-transport (transport configurator / tester)
120 - UPnP-based IP detection
121 (Note: build library always, build service when libxml2/etc. are available)
122 - instantly filter addresses from *other* peers that
123 are *equal* to our own address + port (i.e., localhost:2086). We
124 no longer filter those for outgoing (helps with loopback testing
125 and keeps the code clean), but we should filter strictly *impossible*
126 incoming addresses! This is for efficiency, not correctness.
127 - We currently are happy to take any address told to us in a WELCOME
128 to our set of addresses; we should have some minimal threshold-based
129 scheme, limiting both the total number of addresses that we accept
130 this way as well as requiring multiple confirmations; also, we
131 should possibly try to confirm that the given address works for
132 us ourselves (loopback-style) before adding it to the list
134 - not sure current way of doing ACKs works well-enough
135 with unreliable transports where the ACK maybe lost;
136 the "is_new" check would then possibly prevent future
137 ACKs to be delivered, all while we're happily
138 receiving messages from that peer! Worse, the other
139 peer won't generate another ACK since it thinks we're
140 connected just fine...
142 + How necessary is ACKing in the first place? (alternatives?)
143 + Should we transmit ACKs in response to every HELLO? (would that
144 fully address the problem?)
145 - latency measurements implemented in the transport
146 plugins makes it only work for bi-di transports
147 and results in code replication
148 - should latency be included in the ReceiveCallback and
149 NotifyConnect or passed on request?
151 - auto-generate "defaults.conf" using gnunet-setup from "config.scm"
152 - integrate all options into "config.scm"
153 - change config-file writing to exclude options set to default values
155 - implement exponential back-off for service restarts
156 - better tracking of which config changes actually need to cause process restarts by ARM.
157 - have way to specify dependencies between services (to manage ARM restarts better)
159 - have gnunet-peerinfo print actual host addresses again
160 - add option to gnunet-peerinfo to modify trust value
162 - finish postgres implementation; simplify other SQLs using new stats
167 - active reply route caching design & implementation of service,
172 PHASE #4: [completion-goal: mid 2010]
180 * Features eliminated from util:
181 - threading (goal: good riddance!)
182 - complex logging features [ectx-passing, target-kinds] (goal: good riddance!)
183 - complex configuration features [defaults, notifications] (goal: good riddance!)
184 - network traffic monitors (goal: eliminate)
185 - IPC semaphores (goal: d-bus? / eliminate?)
187 - DNS lookup (goal: have async service; issue: still need synchronous resolution in places, current code may not be portable)
188 => code shrunk from 61 files to 34, 22k LOC to 15k LOC, 470k to 330k (with symbols)
189 * New features in util:
191 - service and program boot-strap code
192 * Major changes in util:
193 - more expressive server (replaces selector)
195 - how to integrate scheduler with GTK event loop!
200 * UTIL: 75%, 4914 out of 6463