TODO 18 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370
  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. o we call signal(), but we should be calling sigaction()
  14. o send socks rejects when things go bad ?
  15. o on solaris, need to build with
  16. LDFLAGS="-lsocket -lnsl" ./configure
  17. o on solaris, we HAVE_UNAME but the uname() call fails?
  18. . should maybe make clients exit(1) when bad things happen?
  19. e.g. clock skew.
  20. o client-side dns cache doesn't appear to be getting populated
  21. by 'connected' cells. In fact, the 'connected' cells don't even
  22. include the IP.
  23. o When it can't resolve any dirservers, it is useless from then on.
  24. We should make it reload the RouterFile if it has no dirservers.
  25. o Sometimes it picks a middleman node as the exit for a circuit.
  26. o if you specify a non-dirserver as exitnode or entrynode, when it
  27. makes the first few circuits it hasn't yet fetched the directory,
  28. so it warns that it doesn't know the node.
  29. o make 'make test' exit(1) if a test fails.
  30. . fix buffer unit test so it passes
  31. Short-term:
  32. - add in 'notice' log level
  33. - make recommendedversions different for clients and servers.
  34. e.g. C0.0.3 vs S0.0.3?
  35. o put IP into the descriptor, so clients don't need to resolve things
  36. o when you hup, rewrite the router.desc file (and maybe others)
  37. - consider handling broken socks4 implementations
  38. - improve how it behaves when i remove a line from the approved-routers files
  39. - Make tls connections tls_close intentionally
  40. - Add '[...truncated]' or similar to truncated log entries (like the directory
  41. in connection_dir_process_inbuf()).
  42. . Make logs handle it better when writing to them fails.
  43. o leave server descriptor out of directory if it's too old
  44. o Rename ACI to circID
  45. o integrate rep_ok functions, see what breaks
  46. - update tor faq
  47. o obey SocksBindAddress, ORBindAddress
  48. o warn if we're running as root
  49. o make connection_flush_buf() more obviously obsolete
  50. o let hup reread the config file, eg so we can get new exit
  51. policies without restarting
  52. o Put recommended_versions in a config entry
  53. X use times(2) rather than gettimeofday to measure how long it
  54. takes to process a cell
  55. o Separate trying to rebuild a circuit because you have none from trying
  56. to rebuild a circuit because the current one is stale
  57. X Continue reading from socks port even while waiting for connect.
  58. o Exit policies
  59. o Spec how to write the exit policies
  60. o Path selection algorithms
  61. o Choose path more incrementally
  62. o Let user request first/last node
  63. o And disallow certain nodes
  64. D Choose path by jurisdiction, etc?
  65. o Make relay end cells have failure status and payload attached
  66. X let non-approved routers handshake.
  67. X Dirserver shouldn't put you in running-routers list if you haven't
  68. uploaded a descriptor recently
  69. X migrate to using nickname rather than addr:port for routers
  70. - migrate to using IPv6 sizes everywhere
  71. o Move from onions to ephemeral DH
  72. o incremental path building
  73. o transition circuit-level sendmes to hop-level sendmes
  74. o implement truncate, truncated
  75. o move from 192byte DH to 128byte DH, so it isn't so damn slow
  76. X exiting from not-last hop
  77. X OP logic to decide to extend/truncate a path
  78. X make sure exiting from the not-last hop works
  79. X logic to find last *open* hop, not last hop, in cpath
  80. o Remember address and port when beginning.
  81. - Extend by nickname/hostname/something, not by IP.
  82. - Need a relay teardown cell, separate from one-way ends.
  83. X remove per-connection rate limiting
  84. - Make it harder to circumvent bandwidth caps: look at number of bytes
  85. sent across sockets, not number sent inside TLS stream.
  86. o Audit users of connnection_remove and connection_free outside of
  87. main.c; many should use mark_for_close instead.
  88. Rendezvous service:
  89. o Design and specify protocol
  90. o Possible preliminary refactoring:
  91. o Should we break circuits up into "circuit-with-cpath" and
  92. "circuit-without-cpath"?
  93. o We need a way to tag circuits as special-purpose circuits for:
  94. o Connecting from Bob's OP to the introduction point
  95. o Sending introduction requests from the IPoint to Bob
  96. o Connecting from Alice to the rendezvous point for Bob
  97. o Connecting from Bob to the rendezvous point for Alice
  98. o Waiting at a rendezvous point to be joined
  99. o Joined to another circuit at the rendezvous point.
  100. (We should also enumerate all the states that these operations
  101. can be in.) [NM]
  102. o Add circuit metadata [NM]
  103. o Code to configure hidden services [NM] 4 hours
  104. - a section in the README (or its own README): how to configure
  105. hidden services
  106. o Service descriptors
  107. o OPs need to maintain identity keys for hidden services [NM]
  108. o Code to generate and parse service descriptors [NM]
  109. . Advertisement
  110. o Generate y.onion hostnames [NM]
  111. o Store y.onion hostnames to disk. [NM]
  112. o Code to do an HTTP connection over Tor from within Tor [RD]
  113. . Publish service descriptors to directory [RD]
  114. o Directory accepts and remembers service descriptors, and
  115. delivers them as requested
  116. o Frontend [RD]
  117. o Backend [NM]
  118. . Code for OPs to retrieve (and cache?) service descriptors [RD]
  119. - Rendezvous
  120. - Code as needed to generate and parse all rendezvous-related
  121. cell types, and do all handshaking [NM]
  122. - ORs implement introduction points
  123. - OPs with hidden services establish introduction points
  124. - ORs implement rendezvous points
  125. - OPs notice y.onion URLs, and:
  126. - Retrieve service descriptors
  127. - Establish rendezvous points
  128. - Send introduction requests to introduction points
  129. - Communication
  130. - OPs remember which circuits are used for which rendezvous
  131. points, and can look up circuits by location-hidden service
  132. - OPs send/handle BEGIN cells for location-hidden services
  133. - End-to-end communication for location-hidden services
  134. On-going
  135. . Better comments for functions!
  136. . Go through log messages, reduce confusing error messages.
  137. . make the logs include more info (fd, etc)
  138. . Unit tests
  139. . Update the spec so it matches the code
  140. Mid-term:
  141. - Rotate tls-level connections -- make new ones, expire old ones.
  142. So we get actual key rotation, not just symmetric key rotation
  143. o Are there anonymity issues with sequential streamIDs? Sequential
  144. circIDs? Eg an attacker can learn how many there have been.
  145. The fix is to initialize them randomly rather than at 1.
  146. - Look at having smallcells and largecells
  147. . Redo scheduler
  148. o fix SSL_read bug for buffered records
  149. - make round-robining more fair
  150. - What happens when a circuit's length is 1? What breaks?
  151. . streams / circuits
  152. o Implement streams
  153. o Rotate circuits after N minutes?
  154. X Circuits should expire when circuit->expire triggers
  155. NICK . Handle half-open connections
  156. o openssh is an application that uses half-open connections
  157. o Figure out what causes connections to close, standardize
  158. when we mark a connection vs when we tear it down
  159. o Look at what ssl does to keep from mutating data streams
  160. o Put CPU workers in separate processes
  161. o Handle multiple cpu workers (one for each cpu, plus one)
  162. o Queue for pending tasks if all workers full
  163. o Support the 'process this onion' task
  164. D Merge dnsworkers and cpuworkers to some extent
  165. o Handle cpuworkers dying
  166. . Scrubbing proxies
  167. - Find an smtp proxy?
  168. - Check the old smtp proxy code
  169. o Find an ftp proxy? wget --passive
  170. D Wait until there are packet redirectors for Linux
  171. . Get socks4a support into Mozilla
  172. . Tests
  173. o Testing harness/infrastructure
  174. D System tests (how?)
  175. - Performance tests, so we know when we've improved
  176. . webload infrastructure (Bruce)
  177. . httperf infrastructure (easy to set up)
  178. . oprofile (installed in RH >8.0)
  179. NICK . Daemonize and package
  180. o Teach it to fork and background
  181. . Red Hat spec file
  182. o Debian spec file equivalent
  183. . Portability
  184. . Which .h files are we actually using?
  185. . Port to:
  186. o Linux
  187. o BSD
  188. . Solaris
  189. o Cygwin
  190. . Win32
  191. o OS X
  192. - deal with pollhup / reached_eof on all platforms
  193. o openssl randomness
  194. o inet_ntoa
  195. o stdint.h
  196. - Make a script to set up a local network on your machine
  197. o More flexibility in node addressing
  198. D Support IPv6 rather than just 4
  199. o Handle multihomed servers (config variable to set IP)
  200. In the distant future:
  201. D Load balancing between router twins
  202. D Keep track of load over links/nodes, to
  203. know who's hosed
  204. SPEC!! D Non-clique topologies
  205. D Implement our own memory management, at least for common structs
  206. (Not ever necessary?)
  207. D Advanced directory servers
  208. D Automated reputation management
  209. SPEC!! D Figure out how to do threshold directory servers
  210. D jurisdiction info in dirserver entries? other info?
  211. Older (done) todo stuff:
  212. For 0.0.2pre17:
  213. o Put a H(K | handshake) into the onionskin response
  214. o Make cells 512 bytes
  215. o Reduce streamid footprint from 7 bytes to 2 bytes
  216. X Check for collisions in streamid (now possible with
  217. just 2 bytes), and back up & replace with padding if so
  218. o Use the 4 reserved bytes in each cell header to keep 1/5
  219. of a sha1 of the ongoing relay payload (move into stream header)
  220. o Move length into the stream header too
  221. o Make length 2 bytes
  222. D increase DH key length
  223. D increase RSA key length
  224. D Spec the stream_id stuff. Clarify that nobody on the backward
  225. stream should look at stream_id.
  226. Cell:
  227. ACI (anonymous circuit identifier) [2 bytes]
  228. Command [1 byte]
  229. Payload (padded with 0 bytes) [509 bytes]
  230. Relay payload:
  231. Relay command [1 byte]
  232. Stream ID [7 bytes]
  233. Partial SHA-1 [4 bytes]
  234. Length [2 bytes]
  235. Relay payload [495 bytes]
  236. For 0.0.2pre15:
  237. o don't pick exit nodes which will certainly reject all things.
  238. o don't pick nodes that the directory says are down
  239. o choose randomly from running dirservers, not just first one
  240. o install the man page
  241. o warn when client-side tries an address/port which no router in the dir accepts.
  242. For 0.0.2pre14:
  243. o More flexible exit policies (18.*, 18.0.0.0/8)
  244. o Work to succeed in the precense of exit policy violation
  245. o Replace desired_path_len with opaque path-selection specifier
  246. o Client-side DNS caching
  247. o Add entries to client DNS cache based on END cells
  248. o Remove port from END_REASON_EXITPOLICY cells
  249. o Start building new circuits when we get an exit-policy
  250. failure. (Defer exiting from the middle of existing
  251. circuits or extending existing circuits for later.)
  252. o Implement function to check whether a routerinfo_t
  253. supports a given exit addr.
  254. o Choose the exit node of an in-progress circuit based on
  255. pending AP connections.
  256. o Choose the exit node _first_, then beginning, then
  257. middle nodes.
  258. Previous:
  259. o Get tor to act like a socks server
  260. o socks4, socks4a
  261. o socks5
  262. o routers have identity key, link key, onion key.
  263. o link key certs are
  264. D signed by identity key
  265. D not in descriptor
  266. o not in config
  267. D not on disk
  268. o identity and onion keys are in descriptor (and disk)
  269. o upon boot, if it doesn't find identity key, generate it and write it.
  270. o also write a file with the identity key fingerprint in it
  271. o router generates descriptor: flesh out router_get_my_descriptor()
  272. o Routers sign descriptors with identity key
  273. o routers put version number in descriptor
  274. o routers should maybe have `uname -a` in descriptor?
  275. o Give nicknames to routers
  276. o in config
  277. o in descriptors
  278. o router posts descriptor
  279. o when it boots
  280. o every DirFetchPostPeriod seconds
  281. D when it changes
  282. o change tls stuff so certs don't get written to disk, or read from disk
  283. o make directory.c 'thread'safe
  284. o dirserver parses descriptor
  285. o dirserver checks signature
  286. D client checks signature?
  287. o dirserver writes directory to file
  288. o reads that file upon boot
  289. o directory includes all routers, up and down
  290. o add "up" line to directory, listing nicknames
  291. o instruments ORs to report stats
  292. o average cell fullness
  293. o average bandwidth used
  294. o configure log files. separate log file, separate severities.
  295. o what assumptions break if we fclose(0) when we daemonize?
  296. o make buffer struct elements opaque outside buffers.c
  297. o add log convention to the HACKING file
  298. o make 'make install' do the right thing
  299. o change binary name to tor
  300. o change config files so you look at commandline, else look in
  301. /etc/torrc. no cascading.
  302. o have an absolute datadir with fixed names for files, and fixed-name
  303. keydir under that with fixed names
  304. o Move (most of) the router/directory code out of main.c
  305. o Simple directory servers
  306. o Include key in source; sign directories
  307. o Signed directory backend
  308. o Document
  309. o Integrate
  310. o Add versions to code
  311. o Have directories list recommended-versions
  312. o Include line in directories
  313. o Check for presence of line.
  314. o Quit if running the wrong version
  315. o Command-line option to override quit
  316. o Add more information to directory server entries
  317. o Exit policies
  318. o Clearer bandwidth management
  319. o Do we want to remove bandwidth from OR handshakes?
  320. o What about OP handshakes?
  321. X Move away from openssl
  322. o Abstract out crypto calls
  323. X Look at nss, others? Just include code?
  324. o Use a stronger cipher
  325. o aes now, by including the code ourselves
  326. X On the fly compression of each stream
  327. o Clean up the event loop (optimize and sanitize)
  328. o Remove that awful concept of 'roles'
  329. o Terminology
  330. o Circuits, topics, cells stay named that
  331. o 'Connection' gets divided, or renamed, or something?
  332. o DNS farm
  333. o Distribute queries onto the farm, get answers
  334. o Preemptively grow a new worker before he's needed
  335. o Prune workers when too many are idle
  336. o DNS cache
  337. o Clear DNS cache over time
  338. D Honor DNS TTL info (how??)
  339. o Have strategy when all workers are busy
  340. o Keep track of which connections are in dns_wait
  341. o Need to cache positives/negatives on the tor side
  342. o Keep track of which queries have been asked
  343. o Better error handling when
  344. o An address doesn't resolve
  345. o We have max workers running
  346. o Consider taking the master out of the loop?
  347. X Implement reply onions
  348. o Total rate limiting
  349. o Look at OR handshake in more detail
  350. o Spec it
  351. o Merge OR and OP handshakes
  352. o rearrange connection_or so it doesn't suck so much to read
  353. D Periodic link key rotation. Spec?
  354. o wrap malloc with something that explodes when it fails
  355. o Clean up the number of places that get to look at prkey