Commit graph

4005 commits

Author SHA1 Message Date
Roger Dingledine
ab68e312bd actually, i think this fix was closest to what i originally
intended. this way we don't call resolve-my-address, which
yells and screams if we're using an internal address (mere
clients should not have to care).


svn:r8998
2006-11-26 20:01:45 +00:00
Nick Mathewson
116ae04c2d r9399@totoro: nickm | 2006-11-26 14:22:17 -0500
Actually, clients *should* try to guess their addresses.


svn:r8997
2006-11-26 19:22:25 +00:00
Nick Mathewson
68142b18cd r9396@totoro: nickm | 2006-11-26 13:36:29 -0500
Fix router_new_address_suggestion() so it only changes last_guessed_ip() when it actually calls resolve_my_address(). (bug found by dev-girl)


svn:r8996
2006-11-26 18:36:51 +00:00
Roger Dingledine
69ae0a2986 we were leaking all the fingerprints for !reject, !invalid,
and !badexit lines in the approved-routers file.


svn:r8974
2006-11-20 16:36:44 +00:00
Nick Mathewson
b1dc014fc6 r9347@totoro: nickm | 2006-11-20 11:24:46 -0500
11:23 < armadev> typo
 11:23 < nickm> ?
 11:23 < armadev> did you get the coffee yet? :)
 11:23 < nickm> no
 11:23 < armadev> i'm at an advantage..i haven't slept yet


svn:r8972
2006-11-20 16:26:52 +00:00
Nick Mathewson
b4ea6136a4 r9345@totoro: nickm | 2006-11-20 11:22:10 -0500
Fix a small and unlikely memory leak in dirserv.c. I think this is not the one arma is looking for.


svn:r8971
2006-11-20 16:22:46 +00:00
Nick Mathewson
aef64c1277 r9577@Kushana: nickm | 2006-11-20 11:07:28 -0500
Oops; fix bug in last fix. Time to get some coffee.


svn:r8970
2006-11-20 16:07:33 +00:00
Nick Mathewson
a667f2e514 r9567@Kushana: nickm | 2006-11-20 11:05:59 -0500
Fix leak in choose_good_exit_server_general()


svn:r8969
2006-11-20 16:06:09 +00:00
Nick Mathewson
e2abc727e5 r9561@Kushana: nickm | 2006-11-16 22:32:54 -0500
Tweaks to test-connection patch:  use ".noconnect" instead of ".test" (since there are lots of ways to test things).  Use a regular sequence of STREAM events (NEW followed by CLOSED) instead of a new event type.  Make the function that checks the address be static and use const and strcasecmpend properly.
 


svn:r8959
2006-11-17 03:34:58 +00:00
Nick Mathewson
d125c61e02 r9560@Kushana: nickm | 2006-11-16 22:09:12 -0500
Check in an implementation of "test" connections from Scott Squires:
 these connections immediately close upon reaching Tor.  They're useful
 for apps that want to check whether they're talking to the same Tor as
 a given controller.  (I'll be tweaking this a bit before I push.)
 


svn:r8958
2006-11-17 03:34:44 +00:00
Nick Mathewson
b052b810e2 r9319@totoro: nickm | 2006-11-14 13:04:50 -0500
Patch from Zajcev Evgeny to make natd work properly (we needed to allow extra characters after the port number).


svn:r8952
2006-11-14 18:05:01 +00:00
Nick Mathewson
ad1f019110 r9317@totoro: nickm | 2006-11-13 22:45:33 -0500
Fix some more mipspro warnings.  Nothing should remain excecpt "conversion from pointer to same-sized integral type."


svn:r8950
2006-11-14 03:45:48 +00:00
Nick Mathewson
9243e54177 r9313@totoro: nickm | 2006-11-13 20:07:41 -0500
Try to compile with fewer warnings on irix64's MIPSpro compiler /
 environment, which apparently believes that:
   - off_t can be bigger than size_t.
   - only mean kids assign things they do not subsequently inspect.
 
 I don't try to fix the "error" that makes it say:
 
 cc-3970 cc: WARNING File = main.c, Line = 1277
   conversion from pointer to same-sized integral type (potential portability
           problem)
 
     uintptr_t sig = (uintptr_t)arg;
 
 Because really, what can you do about a compiler that claims to be c99
 but doesn't understand that void* x = NULL; uintptr_t y = (uintptr_t) x;
 is safe?
 


svn:r8948
2006-11-14 01:07:52 +00:00
Nick Mathewson
0f6402f17b r9309@totoro: nickm | 2006-11-13 19:05:41 -0500
Whitespace fixes, and clean up code from last natd patch.


svn:r8947
2006-11-14 00:06:45 +00:00
Nick Mathewson
1913cb915e r9308@totoro: nickm | 2006-11-13 18:41:23 -0500
Add support for (Free?)BSD's natd, which was an old way to let you
 have your firewall automatically redirect traffic.  (Original patch
 from Zajcev Evgeny, updated for 0.1.2.x by tup.)
 


svn:r8946
2006-11-14 00:06:31 +00:00
Nick Mathewson
fa6fbbc150 r9307@totoro: nickm | 2006-11-13 18:25:56 -0500
Patch from tup based on patch from Zajcev Evgeny: Make TransPort work even when the server wants to talk before the client.


svn:r8945
2006-11-14 00:06:02 +00:00
Nick Mathewson
0ad2fd1129 r9467@Kushana: nickm | 2006-11-01 18:51:16 -0500
Allow larger amounts of drift in uptime to be "cosmetic".


svn:r8891
2006-11-01 23:53:15 +00:00
Nick Mathewson
5d7ca19b44 r9465@auth3-479: nickm | 2006-11-01 13:16:21 -0500
Add some temporary code to count  how many *distinct* router digests we verify.  (I have a sneaking suspicion that the numbers look way way high.)


svn:r8890
2006-11-01 18:16:44 +00:00
Peter Palfrader
c195b8f806 r10475@danube: weasel | 2006-11-01 14:33:35 +0100
Fix copy/paste error


svn:r8889
2006-11-01 13:33:49 +00:00
Nick Mathewson
31e274ddd3 r9461@Kushana: nickm | 2006-10-31 20:30:12 -0500
Oops. Fix compile error in last build.


svn:r8888
2006-11-01 01:30:25 +00:00
Nick Mathewson
9e13a6f31c r9457@Kushana: nickm | 2006-10-31 18:35:17 -0500
Do not warn when an OR gives us a new circuit end reason. (This will prevent bug 351 from recurring.)


svn:r8887
2006-10-31 23:35:50 +00:00
Nick Mathewson
1bf1f9d2fc r9449@Kushana: nickm | 2006-10-31 00:12:02 -0500
Dump breakdown of PK operations when we get a USR2 signal.  This should help us figure out of we are doing some of them for stupid reasons.


svn:r8881
2006-10-31 19:17:07 +00:00
Roger Dingledine
8a781f7f34 general cleanups from looking through or-cvs
svn:r8855
2006-10-29 07:41:25 +00:00
Roger Dingledine
8f7940348f Avoid an infinite loop on evil controlport input. Fixes bug 350,
unless it doesn't. (Nick, please confirm.)

Backport candidate.


svn:r8844
2006-10-28 08:17:02 +00:00
Roger Dingledine
e473ca2427 give a nicer warning message when windows people try to
configure syslog in their torrc


svn:r8841
2006-10-27 19:35:12 +00:00
Nick Mathewson
5b72dc77ad r9395@Kushana: nickm | 2006-10-26 22:06:51 -0400
Fix Bug 349: Have GETINFO network-status return even old routers, and use long nicknames where appropriate.  Document this.


svn:r8834
2006-10-27 02:07:04 +00:00
Nick Mathewson
3fbb292bff r9389@Kushana: nickm | 2006-10-25 17:46:16 -0400
Add a CLEARDNSCACHE signal to clear the client-side DNS cache.


svn:r8829
2006-10-25 21:46:21 +00:00
Nick Mathewson
5c670a186c r9386@Kushana: nickm | 2006-10-25 17:35:38 -0400
Resolve bug 347: translate v0 controller signal numbers into internal signal numbers; leave v1 signal numbers alone.


svn:r8828
2006-10-25 21:39:42 +00:00
Nick Mathewson
4df90e455c r9164@totoro: nickm | 2006-10-24 17:50:43 -0400
Oops: 17:50 < armadev> today is opposite day?


svn:r8822
2006-10-24 21:51:06 +00:00
Nick Mathewson
6b7d1c94da r9162@totoro: nickm | 2006-10-24 17:41:35 -0400
Make "boolean !=" logic more obvious.


svn:r8821
2006-10-24 21:41:48 +00:00
Nick Mathewson
a5bf6df14a r9160@totoro: nickm | 2006-10-24 17:38:07 -0400
More dirserver oops: actually set the digest for a given name in the case when the name is not yet recognized.


svn:r8820
2006-10-24 21:38:31 +00:00
Nick Mathewson
ba1b04c736 r9158@totoro: nickm | 2006-10-24 16:50:18 -0400
Oops on last authority patch: handle routers with unrecognized fingerprints correctly


svn:r8819
2006-10-24 20:51:28 +00:00
Roger Dingledine
6877a7e1ee and update the pile of getinfo strings that's impossible to remember
to update


svn:r8813
2006-10-24 04:55:02 +00:00
Nick Mathewson
6c7bf2f830 r9364@Kushana: nickm | 2006-10-23 22:57:48 -0400
Remove reference to NS_FULL event; add two GETINFO targets.


svn:r8812
2006-10-24 02:58:03 +00:00
Roger Dingledine
6ecf74af60 and blow away what appears to be some redundant logic
svn:r8806
2006-10-23 10:17:52 +00:00
Roger Dingledine
77c8c0ee51 spec and implement some more status events. start to build
some conventions.


svn:r8805
2006-10-23 10:16:43 +00:00
Roger Dingledine
2a1c294717 implement a few status events, so we can make sure they work,
and so vidalia can start handling them if it wants.


svn:r8802
2006-10-23 05:51:46 +00:00
Roger Dingledine
c804eea39d build on 64-bit again
svn:r8801
2006-10-23 05:36:56 +00:00
Roger Dingledine
d48828dbe1 good thing we didn't have to track this bug down the old-fashioned way
svn:r8800
2006-10-23 05:29:40 +00:00
Nick Mathewson
4401c47f63 r9347@Kushana: nickm | 2006-10-23 01:24:48 -0400
Implement status event backend logic.


svn:r8799
2006-10-23 05:24:57 +00:00
Nick Mathewson
27a3f0da2b r9320@Kushana: nickm | 2006-10-22 15:56:27 -0400
fix unittests; add more tests for smartlist stuff; make tests cache generated RSA keys (to speed up unittest run time by 30% and encourage people to run them more).


svn:r8796
2006-10-23 03:48:58 +00:00
Nick Mathewson
ed69c10b3f r9319@Kushana: nickm | 2006-10-22 15:25:18 -0400
oops; include or.h changes for last patch.


svn:r8795
2006-10-23 03:48:52 +00:00
Nick Mathewson
42bab1c6d3 r9318@Kushana: nickm | 2006-10-22 15:22:57 -0400
Let directory authorities set the BadExit flag if they like.  Also, refactor directory authority code so we can believe multiple things about a single router, and do fewer linear searches.


svn:r8794
2006-10-23 03:48:42 +00:00
Roger Dingledine
07cfeb0198 fix a minor memory leak every time we rebuild the router store,
fix a rare memory leak if something goes wrong while rebuilding it,
and clean up some code. nick, please confirm.

reported by "fookoowa" in flyspray 346 (yay!)


svn:r8789
2006-10-22 08:08:10 +00:00
Roger Dingledine
0f33a548c2 fix a string, pointed out by matt edman.
also fix a bug next to it that i noticed.


svn:r8784
2006-10-21 02:51:35 +00:00
Roger Dingledine
a0ec758f9e stop writing arbitrary memory out on the controller port.
it's an ugly patch, but at least this way we can see that
it's fixed. maybe clean up the whole "@" syntax later on.


svn:r8783
2006-10-20 23:23:19 +00:00
Nick Mathewson
c9a01dea53 r9313@Kushana: nickm | 2006-10-20 17:04:11 -0400
Oops. Conform to spec.


svn:r8782
2006-10-20 21:04:44 +00:00
Nick Mathewson
01bc03eff4 r9312@Kushana: nickm | 2006-10-20 14:45:22 -0400
Add an event type to tell the controller when our opinion of a router status has changed.  I might have missed some cases here.


svn:r8781
2006-10-20 21:04:39 +00:00
Roger Dingledine
0459db2c0d checkpoint some changes as i read diffs
svn:r8780
2006-10-20 19:11:12 +00:00
Nick Mathewson
b27cb38dee r9307@Kushana: nickm | 2006-10-20 13:53:54 -0400
Have connection_about_to_close use an end_reason field in edge_connection_t to tell what reason to tell the controller for closing the stream.  Set end_reason in connection_edge_end, connection_mark_unattached_ap, and everwhere we set edge_has_sent_end.  Add a changelog entry.


svn:r8779
2006-10-20 17:54:48 +00:00