Commit Graph

379 Commits

Author SHA1 Message Date
Roger Dingledine
a4753283dd better interface for connection_ap_handshake_socks_reply()
make --list-fingerprint print the fingerprint again


svn:r2668
2004-11-03 23:13:28 +00:00
Nick Mathewson
451f8b5045 - Implement all of control interface except authentication, setconfig,
and actually making the sockets.
- Make sure that identity-based nicknames start with $.
- Use new string_join interface.


svn:r2661
2004-11-03 18:33:07 +00:00
Nick Mathewson
4fdaa5de51 Streamline the two redundant "Has the second rolled over?" checks in prepare_for_poll and connection_bucket_refill; also, generate BANDWIDTH control events
svn:r2657
2004-11-03 16:38:04 +00:00
Roger Dingledine
12e92eb82f connection_read_bucket_decrement() has a side-effect that
we need to get even if we just read 0 bytes


svn:r2656
2004-11-03 10:18:31 +00:00
Nick Mathewson
347d3f9d62 Start implementing control interface.
svn:r2652
2004-11-03 01:32:26 +00:00
Nick Mathewson
8db0abb687 First attempt at hibernation code. It needs more work around the XXXXs, but first I want arma to review the basic approach while I meditate on how to fix them.
svn:r2649
2004-11-02 23:47:32 +00:00
Nick Mathewson
ad4dc74482 Use a stricter set of warnings; make them all pass.
svn:r2645
2004-11-02 03:02:17 +00:00
Roger Dingledine
984ffec105 implement first piece of hibernation
still need to track bandwidth, and make decisions based on bandwidth


svn:r2630
2004-10-31 20:28:41 +00:00
Nick Mathewson
2fbf31533b Tricksy compiler warnings! We hates them, hates them forever, my precious!
svn:r2615
2004-10-27 21:14:11 +00:00
Roger Dingledine
0113a643a2 quick-and-dirty dir policy since the dirservers are getting hammered
nick, could you abstract this sometime so we don't repeat the
sockspolicy code twice?


svn:r2589
2004-10-25 06:16:26 +00:00
Nick Mathewson
ef4ac65f5e Document the swiss-army-knife that retry_listeners has become
svn:r2584
2004-10-24 01:50:33 +00:00
Nick Mathewson
0df2d383d1 Many non-native speakers use Tor. It is our duty to ensure that our log messages set an example of good English spelling.
svn:r2583
2004-10-24 01:29:01 +00:00
Nick Mathewson
57f09573f5 Every 60 seconds, check whether the listeners are still alive, and relaunch the dead ones.
svn:r2581
2004-10-24 01:22:40 +00:00
Nick Mathewson
5d53828c57 Handle more errnos from accept() without closing the connection. This may fix a bug that could close OR listeners when (a) TCP connections were hung up before accept() could be called, or (b) during FD exhaustion.
svn:r2579
2004-10-24 00:55:18 +00:00
Roger Dingledine
de65052312 don't assert multiple things in the same tor_assert()
svn:r2544
2004-10-16 22:14:52 +00:00
Nick Mathewson
25a046df74 Fix unit tests, make it compile
svn:r2543
2004-10-16 21:57:24 +00:00
Nick Mathewson
47dc024f7e Change interface of parse_addr_port() to return address in host order, since most users seem to want that.
svn:r2542
2004-10-16 21:53:30 +00:00
Nick Mathewson
0f144d6a04 fix warning on win32
svn:r2518
2004-10-14 15:41:41 +00:00
Roger Dingledine
cf9664151d bugfix on a really old bug:
if read() fails on a stream, send back an end cell before you
close it down.


svn:r2509
2004-10-14 09:28:31 +00:00
Nick Mathewson
1c9426d6e0 Build without warnings on mac gcc 3.3
svn:r2487
2004-10-14 03:18:14 +00:00
Roger Dingledine
aebc3a03ba more int to size_t conversions, fixing one or more amd64 bugs
plus a whitespace patch on config.c from vicman


svn:r2482
2004-10-14 02:47:09 +00:00
Nick Mathewson
0d5a847f12 Remove need for dirservers file: now, we note trusted dirservers in configuration options, and only need to remember addr:port and key digest for each one.
svn:r2479
2004-10-14 01:44:32 +00:00
Roger Dingledine
2b7b86f5e6 stop trying to print a NULL
plus make router_nickname_is_approved() more obviously unused


svn:r2469
2004-10-13 20:17:45 +00:00
Nick Mathewson
e0cce8fba8 Unify tests for "did I originate this nonopen OR connection?"
svn:r2468
2004-10-13 20:05:57 +00:00
Nick Mathewson
508970f85d fix typo
svn:r2466
2004-10-13 19:55:40 +00:00
Nick Mathewson
149f57bdef answer question about rep_hist_note_connect_failed
svn:r2464
2004-10-13 19:42:36 +00:00
Nick Mathewson
0383afac9a Use parse_addr_port() function instead of ad-hoc variants
svn:r2441
2004-10-12 15:52:09 +00:00
Roger Dingledine
136d4e5739 prefer tor_free to free
plus complain more loudly when we fail to parse a dir we just fetched


svn:r2401
2004-09-29 06:52:36 +00:00
Nick Mathewson
c66e4c4870 Flush more changes from sandbox
- make clients cache directories and use them to seed their router lists
   at startup.  This means clients have a datadir again.
 - Introduce a global_write_bucket.  We need to respond better to exhausting
   it.
 - Remove the last vestiges of LinkPadding and TrafficShaping.
 - Configuration infrastructure support for warning on obsolete options.
 - Refactor directory header parsing to use smartlist_split_string.
 - Respond to content-encoding headers by trying to uncompress as appropriate.
 - Reply with a deflated directory when a client asks for "dir.z".
   (We could use allow-encodings instead, but allow-encodings isn't
   specified in HTTP 1.0.)


svn:r2335
2004-09-08 06:52:33 +00:00
Nick Mathewson
9836ccda8c Move a paren 4 characters to the left; fix a bug in OutboundBindAddress
svn:r2330
2004-09-02 23:25:23 +00:00
Roger Dingledine
2d8a1f208a if they ask for exactly one entrynode, don't pick it as an exitnode.
also, fix part of the zero-identity-router-in-stats bug


svn:r2252
2004-08-17 07:09:02 +00:00
Roger Dingledine
0a6e27208b o Handle servers with dynamic IP addresses: don't replace
options->Address with the resolved one at startup.
  o detect our address right before we make a routerinfo each time.
o external IP vs bind-IP. Already done, just use options->Address.
o OutboundBindAddress config option, to bind to a specific
  IP address for outgoing connect()s.


svn:r2241
2004-08-16 11:43:18 +00:00
Roger Dingledine
699a9d8b46 commit some minor fixes from this sandbox so i can commit the next one
svn:r2229
2004-08-15 05:28:09 +00:00
Roger Dingledine
7506cce649 stop wasting cpu time on authdirservers
we were making a new directory only when the old one was dirty, but
every time we made one, we'd dirty it by testing it right then.


svn:r2199
2004-08-08 11:15:38 +00:00
Roger Dingledine
b2c7b5adfb fix a race condition in 008pre2: don't try to extend onto a connection
that's still handshaking.

for servers in clique mode, require the conn to be open before you'll
choose it for your path.


svn:r2198
2004-08-08 10:32:36 +00:00
Roger Dingledine
276d953b22 fix an assert: check the sockspolicy before we make/add the connection,
else we close a connection without assigning it a state, which is bad
because it fails assert_conn_ok()


svn:r2156
2004-08-06 09:56:36 +00:00
Roger Dingledine
0da256ef97 when we get a sigint, don't accept new connections/circuits,
but delay 30 seconds until exiting.
if we get a second sigint, exit immediately.


svn:r2070
2004-07-20 23:31:00 +00:00
Roger Dingledine
dc7951a5a7 a whole swath of fixes
introduce an authdir_mode() macro to match the others.
don't initialize uptime to the number of seconds since 1970.
non-authoritative dirservers don't cache their directory on disk.
make only authdirservers use clique_mode.
only read approved-routers file if you're an authdirserver.
even authdirservers fetch a new directory in do_hup.
retry_all_connections() is now called retry_all_listeners().
router_parse_list_from_string() correctly reports the router number
  it's working on.
only call dirserv_add_own_fingerprint() and
   dirserv_add_descriptor() on startup if we're an authdirserver.
if AuthDir and !ORPort then fail.
if AuthDir and ClientOnly then fail.


svn:r2061
2004-07-20 10:17:43 +00:00
Roger Dingledine
a5ff0527e6 it turns out we weren't looking at the result from getsockopt().
now we do.
but i'm not sure it matters, since we also poll for reads, and if
there's an error with the connecting socket, poll is supposed to
return readable, so we should notice it then.

who knows.


svn:r2057
2004-07-19 23:26:21 +00:00
Nick Mathewson
54c129d8dc abstract ORPort/SocksPort checks into server_mode(), proxy_mode(), clique_mode(), etc. Dont change underlying comments.
svn:r2054
2004-07-18 21:47:04 +00:00
Nick Mathewson
c2103eb63a Finish most pre2 items: make running-routers list work right; rename secret key files; make even more lookup-by-nickname use lookup-by-id; default nicknames to hostname.
svn:r2043
2004-07-13 18:23:40 +00:00
Roger Dingledine
d37850bb98 only count bytes transmitted to/from non-local IPs
svn:r2041
2004-07-13 16:58:01 +00:00
Roger Dingledine
86e0ede07e checkpoint: start working on bandwidth tracking and
letting clients evaluate whether they're suitable servers


svn:r2037
2004-07-13 07:42:20 +00:00
Roger Dingledine
2325050b1b tor now tolerates down dirservers better
svn:r2031
2004-07-12 20:39:40 +00:00
Roger Dingledine
538655cd80 if connecting to an OR fails immediately, mark it as down
svn:r2029
2004-07-12 18:19:55 +00:00
Roger Dingledine
cbf73b2bc5 touchups all over
put uptime in descriptor


svn:r2011
2004-07-04 22:48:11 +00:00
Nick Mathewson
7d8de8cd10 More digest/nickname fixes
svn:r2000
2004-07-02 23:40:03 +00:00
Nick Mathewson
541add90a1 Track routers by hash of identity key; use hex hash of identity key in place of nickname; accept (and use) hash of identity key in EXTEND cells.
svn:r1994
2004-07-01 01:16:59 +00:00
Nick Mathewson
3708886939 Implement RESOLVE/RESOLVED cells and socks resolve code
svn:r1978
2004-06-17 18:13:09 +00:00
Nick Mathewson
17b5b3685f Make tor build on win32 again; handle locking for server
svn:r1948
2004-06-05 01:50:35 +00:00
Roger Dingledine
69931106f8 be sure to detach streams from the circuit linked list before freeing them
also, don't bother marking a conn for close if you're about to free it


svn:r1935
2004-06-02 18:32:24 +00:00
Roger Dingledine
f51061439b when the dns resolve is cancelled, or fails, be sure to remove
conn from circ->resolving_streams

otherwise it gets freed and stays there, causing seg faults.


svn:r1915
2004-05-21 12:25:15 +00:00
Roger Dingledine
b6faca2268 enable checking the socks policy
svn:r1906
2004-05-20 04:16:43 +00:00
Nick Mathewson
80365b9897 Allow multiple exit policy lines; mostly add support for AP policies
svn:r1905
2004-05-20 02:42:50 +00:00
Roger Dingledine
b37450ce55 do all the heavy lifting in connection_about_to_close_connection,
not in _connection_mark_for_close

this will hopefully clean up the huge cyclical function mess


svn:r1903
2004-05-19 23:32:20 +00:00
Nick Mathewson
d95f347b14 Add Port to *bindAddress.
svn:r1901
2004-05-19 20:25:44 +00:00
Roger Dingledine
630e930799 don't flush the buf from inside connection_write_to_buf
svn:r1859
2004-05-12 22:56:26 +00:00
Roger Dingledine
1cf1836990 break connection_edge_end out of connection_mark_for_close
svn:r1858
2004-05-12 21:12:33 +00:00
Nick Mathewson
537fb82cbb Make some functions static
svn:r1857
2004-05-12 20:58:27 +00:00
Roger Dingledine
32c42a0ee2 break rend_client_desc_fetched out of connection_mark_for_close
svn:r1856
2004-05-12 20:36:44 +00:00
Roger Dingledine
7ee6194f3a break part of circuit.c into relay.c
svn:r1854
2004-05-12 19:49:48 +00:00
Nick Mathewson
f1bc7af9f5 Make "connected" a different case from "finished_flushing"; always close_immediate whhen connect() fails.
svn:r1852
2004-05-12 19:17:09 +00:00
Roger Dingledine
3cdf2d67da it's amazing what a bit of punctuation can do for appearances
svn:r1843
2004-05-10 10:27:54 +00:00
Roger Dingledine
341d6f2cab more cleanup, including fleshing out or.h more
svn:r1839
2004-05-10 07:37:10 +00:00
Nick Mathewson
81922a125c More doxygenation.
svn:r1832
2004-05-10 04:34:48 +00:00
Roger Dingledine
397466a23d give connection.c a skeletal file/brief header
svn:r1831
2004-05-10 03:56:58 +00:00
Roger Dingledine
8f6aa6688f function header comments for connection.c, including doxygen markup
svn:r1825
2004-05-10 01:32:57 +00:00
Roger Dingledine
b384c5a18d bugfix: do directory_send_command only after we've established a socket
also many comments, cleaning, etc


svn:r1823
2004-05-09 16:33:04 +00:00
Roger Dingledine
9bf9ca4d09 bugfix: add a circ->resolving_streams field, and make dns resolving streams not actually in the connection_array until they're done resolving.
bugfix: actually complain if we duplicate mark-for-close a circuit

add more logging for relay ends that claim dns resolve failed, so we can
find out why they're not being retried.


svn:r1798
2004-05-06 11:08:04 +00:00
Nick Mathewson
2ba0776b02 Only connection_add connections once they have conn->s sett; refactor code around this. Should make stuff more bulletproof.
svn:r1788
2004-05-05 01:26:57 +00:00
Nick Mathewson
af08c4f878 Working strerror for windows socket errors, plus some snide comments.
svn:r1775
2004-05-02 20:18:21 +00:00
Roger Dingledine
25909c2c29 patches on patches
svn:r1763
2004-05-02 03:15:55 +00:00
Nick Mathewson
f6fe336ad4 Documenmt buffers.c; remove function that nobody ever calls.
svn:r1760
2004-05-01 22:08:43 +00:00
Nick Mathewson
6fa2ded74c Fix windows compile
svn:r1757
2004-05-01 20:55:31 +00:00
Nick Mathewson
908ccb9dcd Handle windows socket errors correctly; comment most of common.
svn:r1756
2004-05-01 20:46:28 +00:00
Nick Mathewson
9322641710 Use socketclose on windows as appropriate; end pid files with newline
svn:r1745
2004-04-28 21:14:56 +00:00
Roger Dingledine
dd335d9bb2 use tor_lookup_hostname for binding local addresses too
svn:r1744
2004-04-28 20:57:03 +00:00
Roger Dingledine
80d0b634a4 add a debugging entry, to find the wants-to-read-but-can't bug
svn:r1719
2004-04-26 21:15:06 +00:00
Roger Dingledine
fb2279b90b since we don't support truncateds much, don't bother sending them;
just close the circ.

(this wasn't relevant before, because we were mis-handling destroys.)


svn:r1711
2004-04-26 04:32:01 +00:00
Roger Dingledine
10f3b74cb9 tell us the nickname of the OR that hung up on us
svn:r1706
2004-04-26 01:50:08 +00:00
Roger Dingledine
482d7e7cd3 resolve a double-mark-for-close when things die inside connection_handle_write
svn:r1702
2004-04-25 22:23:54 +00:00
Roger Dingledine
25d54257fb use tor_assert
remove obsolete BUF_OK macro


svn:r1697
2004-04-25 20:37:37 +00:00
Nick Mathewson
0fca143ea1 Remove onion_pkey from connection, since onion keys can change more often than connections. Also add more log messages
svn:r1693
2004-04-25 19:04:11 +00:00
Roger Dingledine
e355ed0e15 log correctly if decoding onion failed
svn:r1690
2004-04-25 04:49:11 +00:00
Nick Mathewson
c44016e86e Merge flagday into main branch.
svn:r1683
2004-04-24 22:17:50 +00:00
Roger Dingledine
8e686d0c72 we added more AP conns, oops
svn:r1569
2004-04-08 07:25:54 +00:00
Roger Dingledine
e79a578fec if the rend desc cache entry was fetched more than 15 mins ago,
then try to fetch a new one. but if the new fetch fails, use the
old one.


svn:r1540
2004-04-07 22:41:00 +00:00
Nick Mathewson
3d60cd1d61 Delete trailing whitespace and expand tabs; remove unused aes decrypt code
svn:r1522
2004-04-07 19:57:40 +00:00
Roger Dingledine
7793078dff alice can now look up bob's service descriptor,
choose an intro point, connect to it,
choose a rend point, connect to it and establish a cookie,
get an ack from the rendezvous point,
and know when both circs are ready for her.

APConns don't use conn->purpose anymore

don't initiate a renddesc lookup if one is already in progress

also fix a buffer overflow in nickname parsing (only exploitable
by the operator though)


svn:r1471
2004-04-05 00:47:48 +00:00
Roger Dingledine
bbc9484957 fix some bugs; more remain
svn:r1450
2004-04-03 00:55:53 +00:00
Roger Dingledine
2d3ac08633 Refactor directory servers
* read all the time (before we would ignore eof sometimes, oops)
* we can handle different urls now
* send back 404 for an un-handled url
* commands initiated by the client can handle payloads now
* introduce conn->purpose to avoid exponential state-space explosion


svn:r1400
2004-03-30 22:57:49 +00:00
Roger Dingledine
bcda3ebaef bugfix: we were closing socks-request connections immediately, rather
than flushing them, if they were timing out before we marked them


svn:r1366
2004-03-29 20:04:09 +00:00
Roger Dingledine
e6a2e03524 hold socks reply open until flushed (if possible)
svn:r1356
2004-03-28 09:48:21 +00:00
Roger Dingledine
56dfc3151f work around a tsocks bug
when an AP connection dies early, be sure to do a socks reject
for it. if you just hang up, tsocks goes into an infinite loop.


svn:r1352
2004-03-27 05:45:52 +00:00
Roger Dingledine
9ab5f3c067 fix an assert trigger where an OP would fail to handshake, and we'd
expect it to have a nickname.


svn:r1351
2004-03-27 01:28:14 +00:00
Roger Dingledine
4d747cd954 a missing comma could wreak all sorts of havoc.
svn:r1335
2004-03-26 18:29:05 +00:00
Nick Mathewson
2ba2f02181 Integrate the new "rephist" [rep(utation) hist(ory)] module to trace
successful/failed connections, successful/failed extends, and
connection uptimes.

It's still not done: more tests are needed, and not everything calls
connection/circuit_mark_for_close properly.  This skews the results.

Also, there needs to be a 'testing' mode for non-OP ORs, where they
periodically build circuits just to test whether extends work.


svn:r1313
2004-03-20 04:59:29 +00:00
Roger Dingledine
74ef3a63ae bugfix: decrement OR connections from global_bucket too
bugfix: don't return immediately for has_pending_tls_data unless
        the conn is allowed to read


svn:r1270
2004-03-14 17:06:29 +00:00
Roger Dingledine
703b2d3cf8 refactor bandwidth-control token buckets
this is a checkpoint commit; there still remain some
bugs, er, somewhere.


svn:r1269
2004-03-14 16:00:52 +00:00
Roger Dingledine
87797bcde5 get rid of those pesky tabs
svn:r1264
2004-03-12 13:00:34 +00:00
Nick Mathewson
9b4203266e Make OP work on windows! (Also misc logging tweaks)
svn:r1258
2004-03-11 06:19:08 +00:00
Nick Mathewson
0e5a134709 Fix DOS line endings; autoconfize checks for strptime and inet_aton
svn:r1248
2004-03-09 22:09:13 +00:00
Nick Mathewson
2da54de968 Make tor build on windows again. More work still needed
svn:r1247
2004-03-09 22:01:17 +00:00
Roger Dingledine
6c9d78b319 stop asserting that computers always go forward in time
it's simply not true


svn:r1236
2004-03-06 01:43:37 +00:00
Nick Mathewson
b7633e2e67 Try to find out early if buffers get trashed or double-freed.
svn:r1225
2004-03-03 22:49:15 +00:00
Roger Dingledine
419a672222 holding until flush was borked
we were never writing anything when hold_open_until_flushed was set,
since conn_write returns early if marked_for_conn is set.

seems a bit better now.


svn:r1214
2004-03-03 08:46:18 +00:00
Roger Dingledine
d8e5f3a3a2 close immediately when write() fails too, else things will turn bad
svn:r1212
2004-03-03 07:26:34 +00:00
Roger Dingledine
3c71d09402 use conn->hold_open_until_flushed for streams
and also note/fix a variety of other stream-based bugs


svn:r1210
2004-03-03 06:26:34 +00:00
Nick Mathewson
f4e4dac801 Implement hold_open_until_flushed. I may have missed something important.
svn:r1209
2004-03-03 05:08:01 +00:00
Nick Mathewson
6379dbbe2a warn more when losing data on closed/marked connections.
svn:r1203
2004-03-03 02:07:57 +00:00
Roger Dingledine
49281424ae bugfix: close_immediate called buf_clear but listeners don't have bufs
svn:r1202
2004-03-03 01:58:45 +00:00
Nick Mathewson
a1e4c6dc85 Fix assertion failure spotted by arma.
Apparently, when a DNS failure was already cached, then when we tried
to mark the exit connection as closed, we'd try to remove it from the
pending queue anyway, and hit an assert.  Now, we put failed-resolve
connections in a separate state so that mark_for_close does the right
thing.


svn:r1196
2004-03-02 07:24:11 +00:00
Roger Dingledine
195ee8c845 create a separate connection_edge_destroy() function
svn:r1176
2004-02-29 00:11:37 +00:00
Roger Dingledine
ae67c48805 bugfix: in some cases, connection_mark_for_close wasn't marking the conn
svn:r1174
2004-02-28 23:52:56 +00:00
Nick Mathewson
0a673f4491 Split dns_cancel_pending_resolve into dns_cancel_pending_resolve (kill a
resolve in response to a DNS worker dying) and connection_dns_remove (remove
a pending connection from the resolve structure.)


svn:r1168
2004-02-28 22:23:44 +00:00
Nick Mathewson
f039eca658 Stop trying to flush on broken sockets marked for close.
svn:r1163
2004-02-28 19:14:11 +00:00
Roger Dingledine
80410fa8dc some comments to point nick at the latest bugs
svn:r1162
2004-02-28 11:48:22 +00:00
Roger Dingledine
cec2eee533 enable assert_connection_ok()
we weren't actually asserting much. oops.


svn:r1160
2004-02-28 07:51:30 +00:00
Roger Dingledine
70c43e152a note bug in assert_connection_ok
svn:r1159
2004-02-28 07:48:28 +00:00
Roger Dingledine
dd632e697e clean up some log messages and severities
still plenty more left to clean


svn:r1158
2004-02-28 07:01:22 +00:00
Roger Dingledine
e59f19ca03 some more small fixes
svn:r1152
2004-02-28 04:44:48 +00:00
Nick Mathewson
4e89c6661a Push responsibility for connection marking down as far as possible; have only a close path; add some missing end cells; change return conventions a little.
svn:r1149
2004-02-28 04:11:53 +00:00
Nick Mathewson
88e222ff76 Make has_sent_end irrelevant: only try to send end if reason is nonzero
svn:r1148
2004-02-28 03:06:31 +00:00
Roger Dingledine
d0ad9b147a and another one
svn:r1147
2004-02-27 23:30:27 +00:00
Roger Dingledine
611f117ce1 patches/answers to nick's commit
svn:r1146
2004-02-27 23:23:33 +00:00
Nick Mathewson
c5bbb571d2 Refactor mark_for_close, connection_edge_end and friends. Now, everybody
who wants to shut down a connection calls connection_mark_for_close instead
of setting marked_for_close to 1.  This automatically removes the connection
from the DNS cache if needed, sends a RELAY END cell if appropriate, and can
be changed to do whatever else is needed.

Still to do:
  - The same for circuits, maybe.
  - Add some kind of hold_connection_open_until_flushed flag, maybe.
  - Change stuff that closes connections with return -1 to use mark_for_close,
    maybe.


svn:r1145
2004-02-27 22:00:26 +00:00
Roger Dingledine
195dfd35a8 Fix a bug where you might flush some data on a tls connection, and then
add some more data to be flushed but never turn POLLOUT on. not sure
how commonly this bug was hit, but it would be a doozy.

Also add some asserts to see if it happens elsewhere.


svn:r1142
2004-02-27 04:42:14 +00:00
Nick Mathewson
496e414e52 Basic RAM poisoning and magic-checking to notice connection and circuit
corruption faster; also, check for corruption in dns.c so we can fail fast
for the bug that's nailing Lucky and moria3.


svn:r1123
2004-02-25 07:31:46 +00:00
Roger Dingledine
1b25794a56 bugfix: if the exit stream is pending on the resolve, and a destroy
arrives, then the stream wasn't getting removed from the pending list.

this may have been the lucky-bug.
this commit may also not actually fix the bug. it's darn hard to
reproduce.


svn:r1122
2004-02-25 06:57:57 +00:00
Roger Dingledine
7985f525ec fix some bugs in assert_connection_ok
svn:r1118
2004-02-24 22:33:30 +00:00
Roger Dingledine
f00dc6fe1e last minute fix: don't early-flush non-tls streams
e.g., this breaks directory service


svn:r1033
2004-01-31 00:36:00 +00:00
Roger Dingledine
3516ae1a26 add an ap_conn connecting state; report connecting delay
svn:r1011
2004-01-20 09:21:46 +00:00
Roger Dingledine
f5ebf4c712 push out a tls record's worth of data once you've got it
svn:r1009
2004-01-20 02:14:51 +00:00
Roger Dingledine
c485725c5a Fix the dns bug: children weren't dying
We were telling a child to die by closing the parent's file descriptor
to him. But newer children were inheriting the open file descriptor from
the parent, and since they weren't closing them, the socket never closed,
so the child never read eof, so he never knew to exit.

As a side effect to this bug, we were probably failing to properly close
connections to remote hosts, ORs, and OPs, after a dns child was born.

I'm surprised Tor worked at all.


svn:r974
2004-01-06 07:53:40 +00:00
Roger Dingledine
f3fdbadfdf randomize the initial circ_id and stream_id, so an adversary who
breaks in part-way through can't learn how many circs/streams have
been made


svn:r960
2003-12-28 04:46:09 +00:00
Roger Dingledine
a5e6ec244f rename circ_id_t to uint16_t for code clarity
change message when using non-recommended tor version


svn:r954
2003-12-19 19:55:02 +00:00
Roger Dingledine
45a66c6955 make a relay_header_t struct and pack/unpack funcs
split 7-byte stream_id string into 2-byte recognized and 2-byte stream_id
fix two seg faults in fetch_from_buf_http
fix several lurking seg faults in handling unexpected relay cells

still need to
 * clean up relay_crypt
 * use relay dummies if there's going to be a conflict with rh.recognized
 * check for a conflict when generating stream_ids


svn:r953
2003-12-19 05:09:51 +00:00
Roger Dingledine
6a19e64066 remove trailing whitespace
svn:r951
2003-12-17 21:09:31 +00:00
Roger Dingledine
21cc01299b infrastructure for integrity-checks in relay cells
make circuit_consider_sending_sendme use connection_edge_send_command
fix endian bug in relay length handling (maybe)


svn:r946
2003-12-16 22:56:50 +00:00
Roger Dingledine
08534a6271 start tracking the 'It appears I've already sent the end' warning
svn:r932
2003-12-14 08:32:14 +00:00
Roger Dingledine
36ea39ec9c on hup, close and rebind listener ports too (in case their config has changed)
svn:r926
2003-12-14 06:03:46 +00:00
Roger Dingledine
48e42e9346 back out the broken change from yesterday
svn:r891
2003-12-09 01:04:11 +00:00
Roger Dingledine
8bd7c94bf6 general cleanups
svn:r889
2003-12-07 22:03:47 +00:00
Roger Dingledine
d172fdd300 fix vicious bug in connection_ap_attach_pending that caused it to
never work.
fix vicious bug in choose_good_exit_server that caused it to *skip over*
  pending circuits, and look only at *non-pending circuits*, when choosing
  a good exit node for the new circuit.
bugfix: remove incorrect asserts in circuit_get_newest()


svn:r876
2003-12-03 09:50:02 +00:00
Roger Dingledine
02ec2cdb34 bugfix: we were writing to stderr even when it's closed
svn:r868
2003-11-30 09:51:35 +00:00
Roger Dingledine
3818776f62 remove a conn_close_if_marked warning
Nov 27 15:45:19.003 [warn] connection_write_to_buf(): write_to_buf failed. Closing connection (fd 42).
Nov 27 15:45:19.010 [warn] conn_close_if_marked(): Conn (socket 42) still wants to flush. Losing 0 bytes!


svn:r866
2003-11-30 09:35:26 +00:00
Roger Dingledine
00de055f43 bugfix: an offline OP would leak all its sockets and stop working
svn:r865
2003-11-29 02:50:12 +00:00
Nick Mathewson
1155ae9b03 Bring assert_connection_ok up to date
svn:r863
2003-11-23 18:14:19 +00:00
Roger Dingledine
c00953d2a1 make more sure we can't end up with two connections to the same OR
fix leaked pk in connection_tls_finish_handshake


svn:r839
2003-11-18 10:17:52 +00:00
Roger Dingledine
ac56486bf6 use the tor_malloc_zero wrapper
svn:r837
2003-11-18 08:20:19 +00:00