TODO 15 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322
  1. Legend:
  2. SPEC!! - Not specified
  3. SPEC - Spec not finalized
  4. NICK - nick claims
  5. ARMA - arma claims
  6. - Not done
  7. * Top priority
  8. . Partially done
  9. o Done
  10. D Deferred
  11. X Abandoned
  12. Bugs:
  13. - on solaris, need to build with
  14. LDFLAGS="-lsocket -lnsl" ./configure --with-ssl-dir=/usr/local/ssl
  15. - should maybe make clients exit(1) when bad things happen?
  16. e.g. clock skew.
  17. o client-side dns cache doesn't appear to be getting populated
  18. by 'connected' cells. In fact, the 'connected' cells don't even
  19. include the IP.
  20. o When it can't resolve any dirservers, it is useless from then on.
  21. We should make it reload the RouterFile if it has no dirservers.
  22. o Sometimes it picks a middleman node as the exit for a circuit.
  23. o if you specify a non-dirserver as exitnode or entrynode, when it
  24. makes the first few circuits it hasn't yet fetched the directory,
  25. so it warns that it doesn't know the node.
  26. o make 'make test' exit(1) if a test fails.
  27. . fix buffer unit test so it passes
  28. Short-term:
  29. o put IP into the descriptor, so clients don't need to resolve things
  30. o when you hup, rewrite the router.desc file (and maybe others)
  31. - consider handling broken socks4 implementations
  32. - improve how it behaves when i remove a line from the approved-routers files
  33. - Make tls connections tls_close intentionally
  34. - Add '[...truncated]' or similar to truncated log entries (like the directory
  35. in connection_dir_process_inbuf()).
  36. - Make logs handle it better when writing to them fails.
  37. - leave server descriptor out of directory if it's too old
  38. o Rename ACI to circID
  39. . integrate rep_ok functions, see what breaks
  40. - update tor faq
  41. o obey SocksBindAddress, ORBindAddress
  42. o warn if we're running as root
  43. o make connection_flush_buf() more obviously obsolete
  44. o let hup reread the config file, eg so we can get new exit
  45. policies without restarting
  46. o Put recommended_versions in a config entry
  47. X use times(2) rather than gettimeofday to measure how long it
  48. takes to process a cell
  49. o Separate trying to rebuild a circuit because you have none from trying
  50. to rebuild a circuit because the current one is stale
  51. X Continue reading from socks port even while waiting for connect.
  52. o Exit policies
  53. o Spec how to write the exit policies
  54. o Path selection algorithms
  55. o Choose path more incrementally
  56. o Let user request first/last node
  57. o And disallow certain nodes
  58. D Choose path by jurisdiction, etc?
  59. o Make relay end cells have failure status and payload attached
  60. X let non-approved routers handshake.
  61. X Dirserver shouldn't put you in running-routers list if you haven't
  62. uploaded a descriptor recently
  63. X migrate to using nickname rather than addr:port for routers
  64. - migrate to using IPv6 sizes everywhere
  65. o Move from onions to ephemeral DH
  66. o incremental path building
  67. o transition circuit-level sendmes to hop-level sendmes
  68. o implement truncate, truncated
  69. o move from 192byte DH to 128byte DH, so it isn't so damn slow
  70. X exiting from not-last hop
  71. X OP logic to decide to extend/truncate a path
  72. X make sure exiting from the not-last hop works
  73. X logic to find last *open* hop, not last hop, in cpath
  74. o Remember address and port when beginning.
  75. - Extend by nickname/hostname/something, not by IP.
  76. - Need a relay teardown cell, separate from one-way ends.
  77. X remove per-connection rate limiting
  78. - Make it harder to circumvent bandwidth caps: look at number of bytes
  79. sent across sockets, not number sent inside TLS stream.
  80. - Audit users of connnection_remove and connection_free outside of
  81. main.c; many should use mark_for_close instead.
  82. On-going
  83. . Better comments for functions!
  84. . Go through log messages, reduce confusing error messages.
  85. . make the logs include more info (fd, etc)
  86. . Unit tests
  87. . Update the spec so it matches the code
  88. Mid-term:
  89. - Rotate tls-level connections -- make new ones, expire old ones.
  90. So we get actual key rotation, not just symmetric key rotation
  91. o Are there anonymity issues with sequential streamIDs? Sequential
  92. circIDs? Eg an attacker can learn how many there have been.
  93. The fix is to initialize them randomly rather than at 1.
  94. - Look at having smallcells and largecells
  95. . Redo scheduler
  96. o fix SSL_read bug for buffered records
  97. - make round-robining more fair
  98. - What happens when a circuit's length is 1? What breaks?
  99. . streams / circuits
  100. o Implement streams
  101. o Rotate circuits after N minutes?
  102. X Circuits should expire when circuit->expire triggers
  103. NICK . Handle half-open connections
  104. o openssh is an application that uses half-open connections
  105. o Figure out what causes connections to close, standardize
  106. when we mark a connection vs when we tear it down
  107. o Look at what ssl does to keep from mutating data streams
  108. o Put CPU workers in separate processes
  109. o Handle multiple cpu workers (one for each cpu, plus one)
  110. o Queue for pending tasks if all workers full
  111. o Support the 'process this onion' task
  112. D Merge dnsworkers and cpuworkers to some extent
  113. o Handle cpuworkers dying
  114. . Scrubbing proxies
  115. - Find an smtp proxy?
  116. - Check the old smtp proxy code
  117. o Find an ftp proxy? wget --passive
  118. D Wait until there are packet redirectors for Linux
  119. . Get socks4a support into Mozilla
  120. . Develop rendezvous points
  121. X Handle socks commands other than connect, eg, bind?
  122. o Design
  123. - Spec
  124. - Implement
  125. . Tests
  126. o Testing harness/infrastructure
  127. D System tests (how?)
  128. - Performance tests, so we know when we've improved
  129. . webload infrastructure (Bruce)
  130. . httperf infrastructure (easy to set up)
  131. . oprofile (installed in RH >8.0)
  132. NICK . Daemonize and package
  133. o Teach it to fork and background
  134. . Red Hat spec file
  135. o Debian spec file equivalent
  136. . Portability
  137. . Which .h files are we actually using?
  138. . Port to:
  139. o Linux
  140. o BSD
  141. . Solaris
  142. o Cygwin
  143. . Win32
  144. o OS X
  145. - deal with pollhup / reached_eof on all platforms
  146. o openssl randomness
  147. o inet_ntoa
  148. o stdint.h
  149. - Make a script to set up a local network on your machine
  150. o More flexibility in node addressing
  151. D Support IPv6 rather than just 4
  152. o Handle multihomed servers (config variable to set IP)
  153. In the distant future:
  154. D Load balancing between router twins
  155. D Keep track of load over links/nodes, to
  156. know who's hosed
  157. SPEC!! D Non-clique topologies
  158. D Implement our own memory management, at least for common structs
  159. (Not ever necessary?)
  160. D Advanced directory servers
  161. D Automated reputation management
  162. SPEC!! D Figure out how to do threshold directory servers
  163. D jurisdiction info in dirserver entries? other info?
  164. Older (done) todo stuff:
  165. For 0.0.2pre17:
  166. o Put a H(K | handshake) into the onionskin response
  167. o Make cells 512 bytes
  168. o Reduce streamid footprint from 7 bytes to 2 bytes
  169. X Check for collisions in streamid (now possible with
  170. just 2 bytes), and back up & replace with padding if so
  171. o Use the 4 reserved bytes in each cell header to keep 1/5
  172. of a sha1 of the ongoing relay payload (move into stream header)
  173. o Move length into the stream header too
  174. o Make length 2 bytes
  175. D increase DH key length
  176. D increase RSA key length
  177. D Spec the stream_id stuff. Clarify that nobody on the backward
  178. stream should look at stream_id.
  179. Cell:
  180. ACI (anonymous circuit identifier) [2 bytes]
  181. Command [1 byte]
  182. Payload (padded with 0 bytes) [509 bytes]
  183. Relay payload:
  184. Relay command [1 byte]
  185. Stream ID [7 bytes]
  186. Partial SHA-1 [4 bytes]
  187. Length [2 bytes]
  188. Relay payload [495 bytes]
  189. For 0.0.2pre15:
  190. o don't pick exit nodes which will certainly reject all things.
  191. o don't pick nodes that the directory says are down
  192. o choose randomly from running dirservers, not just first one
  193. o install the man page
  194. o warn when client-side tries an address/port which no router in the dir accepts.
  195. For 0.0.2pre14:
  196. o More flexible exit policies (18.*, 18.0.0.0/8)
  197. o Work to succeed in the precense of exit policy violation
  198. o Replace desired_path_len with opaque path-selection specifier
  199. o Client-side DNS caching
  200. o Add entries to client DNS cache based on END cells
  201. o Remove port from END_REASON_EXITPOLICY cells
  202. o Start building new circuits when we get an exit-policy
  203. failure. (Defer exiting from the middle of existing
  204. circuits or extending existing circuits for later.)
  205. o Implement function to check whether a routerinfo_t
  206. supports a given exit addr.
  207. o Choose the exit node of an in-progress circuit based on
  208. pending AP connections.
  209. o Choose the exit node _first_, then beginning, then
  210. middle nodes.
  211. Previous:
  212. o Get tor to act like a socks server
  213. o socks4, socks4a
  214. o socks5
  215. o routers have identity key, link key, onion key.
  216. o link key certs are
  217. D signed by identity key
  218. D not in descriptor
  219. o not in config
  220. D not on disk
  221. o identity and onion keys are in descriptor (and disk)
  222. o upon boot, if it doesn't find identity key, generate it and write it.
  223. o also write a file with the identity key fingerprint in it
  224. o router generates descriptor: flesh out router_get_my_descriptor()
  225. o Routers sign descriptors with identity key
  226. o routers put version number in descriptor
  227. o routers should maybe have `uname -a` in descriptor?
  228. o Give nicknames to routers
  229. o in config
  230. o in descriptors
  231. o router posts descriptor
  232. o when it boots
  233. o every DirFetchPostPeriod seconds
  234. D when it changes
  235. o change tls stuff so certs don't get written to disk, or read from disk
  236. o make directory.c 'thread'safe
  237. o dirserver parses descriptor
  238. o dirserver checks signature
  239. D client checks signature?
  240. o dirserver writes directory to file
  241. o reads that file upon boot
  242. o directory includes all routers, up and down
  243. o add "up" line to directory, listing nicknames
  244. o instruments ORs to report stats
  245. o average cell fullness
  246. o average bandwidth used
  247. o configure log files. separate log file, separate severities.
  248. o what assumptions break if we fclose(0) when we daemonize?
  249. o make buffer struct elements opaque outside buffers.c
  250. o add log convention to the HACKING file
  251. o make 'make install' do the right thing
  252. o change binary name to tor
  253. o change config files so you look at commandline, else look in
  254. /etc/torrc. no cascading.
  255. o have an absolute datadir with fixed names for files, and fixed-name
  256. keydir under that with fixed names
  257. o Move (most of) the router/directory code out of main.c
  258. o Simple directory servers
  259. o Include key in source; sign directories
  260. o Signed directory backend
  261. o Document
  262. o Integrate
  263. o Add versions to code
  264. o Have directories list recommended-versions
  265. o Include line in directories
  266. o Check for presence of line.
  267. o Quit if running the wrong version
  268. o Command-line option to override quit
  269. o Add more information to directory server entries
  270. o Exit policies
  271. o Clearer bandwidth management
  272. o Do we want to remove bandwidth from OR handshakes?
  273. o What about OP handshakes?
  274. X Move away from openssl
  275. o Abstract out crypto calls
  276. X Look at nss, others? Just include code?
  277. o Use a stronger cipher
  278. o aes now, by including the code ourselves
  279. X On the fly compression of each stream
  280. o Clean up the event loop (optimize and sanitize)
  281. o Remove that awful concept of 'roles'
  282. o Terminology
  283. o Circuits, topics, cells stay named that
  284. o 'Connection' gets divided, or renamed, or something?
  285. o DNS farm
  286. o Distribute queries onto the farm, get answers
  287. o Preemptively grow a new worker before he's needed
  288. o Prune workers when too many are idle
  289. o DNS cache
  290. o Clear DNS cache over time
  291. D Honor DNS TTL info (how??)
  292. o Have strategy when all workers are busy
  293. o Keep track of which connections are in dns_wait
  294. o Need to cache positives/negatives on the tor side
  295. o Keep track of which queries have been asked
  296. o Better error handling when
  297. o An address doesn't resolve
  298. o We have max workers running
  299. o Consider taking the master out of the loop?
  300. X Implement reply onions
  301. o Total rate limiting
  302. o Look at OR handshake in more detail
  303. o Spec it
  304. o Merge OR and OP handshakes
  305. o rearrange connection_or so it doesn't suck so much to read
  306. D Periodic link key rotation. Spec?
  307. o wrap malloc with something that explodes when it fails
  308. o Clean up the number of places that get to look at prkey