Commit Graph

168 Commits

Author SHA1 Message Date
Nick Mathewson
d073d7d4eb Consistency issues in load_windows_system_library patch. Thanks Sebastian 2010-09-24 14:16:55 -04:00
Nick Mathewson
aa7f55c45f Use load_windows_system_library in place of LoadLibrary 2010-09-21 14:40:10 -04:00
Sebastian Hahn
ff5ffd1776 Attempt to fix compilation on Windows
Our attempt to make compilation work on old versions of Windows
again while keeping wince compatibility broke the build for Win2k+.
helix reports this patch fixes the issue for WinXP. Bugfix on
0.2.2.15-alpha; related to bug 1797.
2010-09-19 13:39:30 +02:00
Nick Mathewson
c0c7868250 Make the windows build succeed with or without -DUNICODE enabled.
This should keep WinCE working (unicode always-on) and get Win98
working again (unicode never-on).

There are two places where we explicitly use ASCII-only APIs, still:
in ntmain.c and in the unit tests.

This patch also fixes a bug in windoes tor_listdir that would cause
the first file to be listed an arbitrary number of times that was
also introduced with WinCE support.

Should fix bug 1797.
2010-08-20 13:40:01 -04:00
valerino
8d31141ccb Port Tor to work on Windows CE
Most of the changes here are switches to use APIs available on Windows
CE.  The most pervasive change is that Windows CE only provides the
wide-character ("FooW") variants of most of the windows function, and
doesn't support the older ASCII verions at all.

This patch will require use of the wcecompat library to get working
versions of the posix-style fd-based file IO functions.

[commit message by nickm]
2010-05-24 11:46:45 -04:00
Nick Mathewson
2590d733f4 Try to untangle the logic in server_port_flush
It's a bit confusing to have a loop where another function,
confusingly named "*_free", is responsible for advancing the loop
variable (or rather, for altering a structure so that the next time
the loop variable's initializer is evaluated it evaluates to something
different.)

Not only has this confused people: it's also confused coverity scan.
Let's fix that.
2010-01-25 14:09:58 -05:00
Nick Mathewson
d4354b506b Don't use OutboundBindAddress to connect to localhost
The OutboundBindAddress option is useful for making sure that all of
your outbond connections use a given interface.  But when connecting
to 127.0.0.1 (or ::1 even) it's important to actually have the
connection come _from_ localhost, since lots of programs running on
localhost use the source address to authenticate that the connection
is really coming from the same host.

Our old code always bound to OutboundBindAddress, whether connecting
to localhost or not.  This would potentially break DNS servers on
localhost, and socks proxies on localhost.  This patch changes the
behavior so that we only look at OutboundBindAddress when connecting
to a non-loopback address.
2010-01-20 13:09:10 -05:00
Nathan Freitas
8c585cce39 Include util.h and log.h as relative paths.
This shouldn't be necessary, but apparently the Android cross-compiler
doesn't respect -I as well as it should.  (-I is supposed to add to the
*front* of the search path.  Android's gcc wrapper apparently likes to add to
the end.  This is broken, but we need to work around it.)
2009-09-29 00:52:52 -04:00
Nick Mathewson
1cda6f3e75 Merge commit 'origin/maint-0.2.1' 2009-09-01 15:59:40 -04:00
Sebastian Hahn
aea9cf1011 Fix compile warnings on Snow Leopard
Big thanks to nickm and arma for helping me with this!
2009-09-01 18:36:27 +02:00
Mike Perry
9e1fe29beb Switch over to tor_strtok_r instead of strtok_r. 2009-08-09 18:42:29 -07:00
Nick Mathewson
9f28cfe86a Merge commit 'origin/maint-0.2.1' 2009-06-30 10:13:51 -04:00
Nick Mathewson
7ce767181f Have eventdns set the "truncated" bit correctly.
Fixed bug 1022; This isn't actually a live bug in Tor, since in Tor
we never generate large DNS replies.
2009-06-30 10:05:42 -04:00
Nick Mathewson
e5b88dc83f Update Tor to use Libevent 2.0 APIs when available.
This patch adds a new compat_libevent.[ch] set of files, and moves our
Libevent compatibility and utilitity functions there.  We build them
into a separate .a so that nothing else in src/commmon depends on
Libevent (partially fixing bug 507).

Also, do not use our own built-in evdns copy when we have Libevent
2.0, whose evdns is finally good enough (thus fixing Bug 920).
2009-06-12 14:27:52 -04:00
Nick Mathewson
b805e5debf Merge commit 'origin/maint-0.2.1' into master.
Conflicts:
	ChangeLog
	configure.in
	contrib/tor-mingw.nsi.in
	src/win32/orconfig.h
2009-05-25 16:43:52 -04:00
Nick Mathewson
99c315a473 Use tor_socket_strerror in eventdns.c. Fixes bug 987. 2009-05-25 16:38:50 -04:00
Nick Mathewson
a3fadddd4a Improved bug-957 fix for 0.2.2.
Really, our idiocy was that we were calling event_set() on the same
event more than once, which sometimes led to us calling event_set() on
an event that was already inserted, thus making it look uninserted.
With this patch, we just initialize the timeout events when we create
the requests and nameservers, and we don't need to worry about
double-add and double-del cases at all.
2009-05-22 14:06:39 -04:00
Nick Mathewson
1d002a25a2 Finish up the 0.2.1 version of the bug 957 fix.
Basically, all this means is downgrading our warning messages to debug
messages, since the bug workaround code here is adequate to stop the
bug.
2009-05-22 11:47:09 -04:00
Nick Mathewson
34030a3d23 Possible fix for crash bug related to event timeouts. [Bug 957]
If we ever add an event, then set it, then add it again, there will be
now two pointers to the event in the event base.  If we delete one and
free it, the first pointer will still be there, and possibly cause a
crash later.

This patch adds detection for this case to the code paths in
eventdns.c, and works around it.  If the warning message ever
displays, then a cleverer fix is in order.

{I am not too confident that this *is* the fix, since bug 957 is very
tricky.  If it is, it is a bugfix on 0.2.0.}
2009-05-18 16:12:39 -04:00
Nick Mathewson
27c3b43817 Fix one case of bug 929.
svn:r18683
2009-02-23 20:04:51 +00:00
Nick Mathewson
c7f3416a9e Add some debugging code to try to catch the likely cause of bug 929
svn:r18682
2009-02-23 18:57:21 +00:00
Nick Mathewson
65dc835773 Revert an erroneous part of the non-fix to bug 326, and add comments to explain why it was erroneous.
svn:r18494
2009-02-11 17:23:11 +00:00
Nick Mathewson
6e4afe6f13 I worry that the CLEAR() macro in eventdns.c is hiding bugs. That is sad, because it was meant to make them more detectable. Change it to change stuff to garbage rather than to 0. If no bugs turn up, we can remove it in 0.2.2.x
svn:r18493
2009-02-11 17:22:34 +00:00
Nick Mathewson
8d9aadf57c Port some mostly cosmetic eventdns changes from libevent
svn:r18492
2009-02-11 17:22:29 +00:00
Nick Mathewson
fe987d3a17 Remove some deadcode and use tor_inet_aton uniformly.
svn:r18422
2009-02-09 03:13:05 +00:00
Nick Mathewson
2ebdf91a52 Fix a remote-crash bug. This will need a patch release.
svn:r18421
2009-02-09 03:13:02 +00:00
Nick Mathewson
f78793879d Fix a race condition on nameserver reconfiguration.
This resolves bug 526, wherein we would crash if the following
events occurred in this order:
  A: We're an OR, and one of our nameservers goes down.
  B: We launch a probe to it to see if it's up again. (We do this hourly
     in steady-state.)
  C: Before the probe finishes, we reconfigure our nameservers,
     usually because we got a SIGHUP and the resolve.conf file changed.
  D: The probe reply comes back, or times out. (There is a five-second
     window for this, after B has happens).

IOW, if one of our nameservers is down and our nameserver
configuration has changed, there were 5 seconds per hour where HUPing
the server was unsafe.

Bugfix on 0.1.2.1-alpha.  Too obscure to backport.

svn:r18306
2009-01-28 18:26:20 +00:00
Nick Mathewson
2aaf71826d Fix a compile warning on win32.
svn:r18099
2009-01-14 04:03:24 +00:00
Nick Mathewson
462f64b6b9 Make outgoing DNS requests respect OutboundBindAddress.
Fixes the bug part of bug 789.

svn:r17983
2009-01-06 20:50:55 +00:00
Nick Mathewson
c4b8fef362 Remove svn $Id$s from our source, and remove tor --version --version.
The subversion $Id$ fields made every commit force a rebuild of
whatever file got committed.  They were not actually useful for
telling the version of Tor files in the wild.

svn:r17867
2009-01-04 00:35:51 +00:00
Nick Mathewson
9c20441bcb Only set sin_len/sin6_len when they exist.
svn:r17851
2009-01-02 20:57:10 +00:00
Nick Mathewson
48f2ce298b Try harder to make sure we zero-out the extraneous sockaddr fields and that we set sockaddr_len. Conceivably a backport candidate, though nothing has yet been sen to break.
svn:r17849
2009-01-02 20:39:38 +00:00
Nick Mathewson
ffe22fab9b Indeed, arma was right. There was one ntohl too many in debug_ntop in eventdns.c
svn:r17826
2008-12-30 04:08:45 +00:00
Roger Dingledine
c190842a34 point out the bug location to nick. unless i'm wrong.
svn:r17825
2008-12-30 02:36:11 +00:00
Nick Mathewson
2c4097e99d More compilation tweaks on Android: fix two clear errors in our code that apparently the compiler cares about there.
svn:r17808
2008-12-29 04:01:58 +00:00
Nick Mathewson
029be5ad02 Move in-addr.arpa parsing and generation into address.c, and simplify the code that does it elsewhere. Incidentally, this lets exit servers answer requests for ip6.arpa addresses.
svn:r17707
2008-12-19 18:52:00 +00:00
Nick Mathewson
148e4e8837 Do not leak the DNS server port structure.
svn:r17680
2008-12-18 15:00:18 +00:00
Nick Mathewson
6693f32530 Resolve many DOCDOCs.
svn:r17662
2008-12-17 22:58:20 +00:00
Nick Mathewson
d8027aa689 Fix a logic error that would automatically reject all but the first configured DNS server. Bugfix on 0.2.1.5-alpha. Possible fix for part of 813/868. Spotted by coderman
svn:r17569
2008-12-10 22:57:41 +00:00
Nick Mathewson
c7910c3bc9 Consider GetNetworkParams() nameserver parsing to have suceeded if even one nameserver can be added. Log more useful info about what exactly is failing when we fail to add a nameserver.
svn:r17368
2008-11-22 21:46:34 +00:00
Nick Mathewson
0c9dfffe5a Implement the 0x20-hack to make DNS poisoning harder against us, especially when resolving large names. Add a cfg option to disable it, since apparently 3/10 of a percent of servers get it wrong.
svn:r17171
2008-10-29 19:20:02 +00:00
Roger Dingledine
bb1f488dbe fix typo in log entry
svn:r17050
2008-10-08 10:47:23 +00:00
Nick Mathewson
b0c48d8e45 Oops; we need to make sure that DNS request names are matched in the _questions_ section of the replies. Rejecting answers whether the _answers_ section did not match made us reject A records waiting at the end of a CNAME record. Bug 823.
svn:r16933
2008-09-22 20:06:25 +00:00
Nick Mathewson
c33dde4ac1 Fix mingw build with --enable-gcc-warnings set.
svn:r16759
2008-09-04 21:58:09 +00:00
Nick Mathewson
6bf1dec945 Oops; set address families on nameservers.
svn:r16674
2008-08-27 14:12:10 +00:00
Nick Mathewson
239cbe99ba Oops. Linux has no sa_len field.
svn:r16627
2008-08-22 18:00:26 +00:00
Nick Mathewson
fc52d85b7c r17846@tombo: nickm | 2008-08-22 11:54:00 -0400
Make dns resolver code more robust: handle nameservers with IPv6 addresses, make sure names in replies match requested names, make sure origin address of reply matches the address we asked.


svn:r16621
2008-08-22 16:24:43 +00:00
Nick Mathewson
e0de72dd87 r14410@tombo: nickm | 2008-02-23 16:51:46 -0500
Fix the last of the -Wshorten-64-to-32 warnings.


svn:r13696
2008-02-24 22:10:08 +00:00
Nick Mathewson
daefbfe691 r14371@tombo: nickm | 2008-02-21 16:13:18 -0500
Fix all -Wshorten-64-to-32 warnings that appear on my macbook.


svn:r13662
2008-02-21 21:15:31 +00:00
Nick Mathewson
5dd9d16b94 r18263@catbus: nickm | 2008-02-20 12:45:15 -0500
Oops. not supposed to use tor_assert() in eventdns.c


svn:r13622
2008-02-20 17:48:37 +00:00