|
another todo item we ought to do
r6363@totoro (orig r6362): weasel | 2006-04-10 06:03:47 -0400
New upstream version
r6364@totoro (orig r6363): phobos | 2006-04-10 10:12:29 -0400
Remove a dependency on dist and assume a tarball is in ".." for
dist-rpm.
r6365@totoro (orig r6364): weasel | 2006-04-10 15:39:26 -0400
Remove redundant includes. They are all hanled in torint.h which we already do include.
r6370@totoro (orig r6369): arma | 2006-04-10 16:00:31 -0400
be willing to add our own routerinfo into the routerlist.
this means authorities will include themselves in their
directories and network-statuses.
r6373@totoro (orig r6372): arma | 2006-04-10 16:08:12 -0400
and forward-port the man page change
r6374@totoro (orig r6373): arma | 2006-04-10 16:16:46 -0400
make DirFetchPeriod and StatusFetchPeriod truly obsolete.
r6376@totoro (orig r6375): arma | 2006-04-10 16:21:55 -0400
0.1.2.0-alpha-cvs!
r6377@totoro (orig r6376): nickm | 2006-04-10 17:23:00 -0400
Remove DER64 functions in trunk: they will never be used again unless the directory authorities switch back to 0.0.9tooearly.
r6378@totoro (orig r6377): weasel | 2006-04-10 17:29:29 -0400
Stop assuming every authority is a v1 authority
r6381@totoro (orig r6380): weasel | 2006-04-10 17:37:20 -0400
[forward port] Update the list of documentation files in tor.nsi and package_nsis-weasel.sh
r6382@totoro (orig r6381): arma | 2006-04-10 17:40:43 -0400
drop moria1 from the list of authorities.
r6383@totoro (orig r6382): arma | 2006-04-11 10:12:04 -0400
ok, put moria1 back in, since it's v1 and thus we need to send
our rendezvous descriptors to it.
eventually we might make a 'v1only' tag that explains it's only
for rendezvous descriptors.
r6384@totoro (orig r6383): phobos | 2006-04-12 21:46:27 -0400
Temporary hacks to ensure make dist-rpm works from cvs until a better
solution can be found.
r6386@totoro (orig r6385): arma | 2006-04-14 16:19:33 -0400
forward-port the n_named log severity downgrade.
r6388@totoro (orig r6387): arma | 2006-04-15 03:15:23 -0400
if the bottom eighth of the servers by bandwidth is really crummy,
try the bottom quartile instead.
r6389@totoro (orig r6388): arma | 2006-04-15 19:53:58 -0400
update spec to reflect the downgraded loglevel for naming complaints.
r6390@totoro (orig r6389): arma | 2006-04-16 18:34:00 -0400
no need to escape the address for our connections -- they are
always IP addresses.
r6391@totoro (orig r6390): arma | 2006-04-16 18:44:08 -0400
better error checking for torify, contributed by jacob appelbaum.
r6392@totoro (orig r6391): arma | 2006-04-16 18:48:41 -0400
and remove the 'debugging' flag for torify
r6393@totoro (orig r6392): weasel | 2006-04-16 22:46:14 -0400
Add an XXX to torify.in - "which" is evil,
quote a few variables
do -h and --help
add (c) and license ("Same as tor")
print an error message if exec falls through
r6394@totoro (orig r6393): arma | 2006-04-17 02:43:27 -0400
slightly clearer log message when you use a nickname rather than
a key for an unnamed server.
r6395@totoro (orig r6394): arma | 2006-04-17 02:46:20 -0400
note another bug: we complain that a router doesn't exist,
because it's down so we never fetched it so it doesn't exist.
r6397@totoro (orig r6396): nickm | 2006-04-17 23:07:24 -0400
forward-port: "Resolve" all XXX011 items, mostly by marking them non-011.
r6399@totoro (orig r6398): nickm | 2006-04-17 23:36:28 -0400
[forward-port] Implement an option, VirtualAddrMask, to set which addresses get handed out in response to mapaddress requests. Needs testing and docs!
r6401@totoro (orig r6400): nickm | 2006-04-17 23:51:18 -0400
[Forward-port ]Test and document last patch.
r6402@totoro (orig r6401): nickm | 2006-04-17 23:58:42 -0400
mainline branch. Remove some more dead XXXs.
r6403@totoro (orig r6402): nickm | 2006-04-18 00:57:07 -0400
require at least 2**16 assignable virtual addresses
r6405@totoro (orig r6404): arma | 2006-04-18 01:03:09 -0400
class B really means /16 here
r6407@totoro (orig r6406): arma | 2006-04-18 15:48:06 -0400
Raise the timeout for complaining about wedged cpuworkers.
This value is high because some servers with low memory/cpu
sometimes spend an hour or more swapping, and Tor starves.
r6408@totoro (orig r6407): weasel | 2006-04-21 11:52:49 -0400
Add add-tor helper script to contrib
r6409@totoro (orig r6408): weasel | 2006-04-21 11:57:09 -0400
Remove test values from add-tor
r6410@totoro (orig r6409): nickm | 2006-04-23 17:36:52 -0400
Prioritize items for 0.1.2
r6411@totoro (orig r6410): nickm | 2006-04-23 17:40:15 -0400
And another 0.1.2 item
r6412@totoro (orig r6411): arma | 2006-04-23 19:05:34 -0400
make more hibernate log messages use local time. we should audit
to see what other log messages keep switching back and forth
between GMT and local.
r6413@totoro (orig r6412): arma | 2006-04-23 19:09:03 -0400
Regenerate our local descriptor if it's dirty and some local
function asks for it. This may resolve bug 286.
r6414@totoro (orig r6413): nickm | 2006-04-24 12:29:06 -0400
Add a stub of a "path-spec", containing only the helper-node emails from arma. whee.
r6415@totoro (orig r6414): nickm | 2006-04-24 13:51:31 -0400
Possible partial fix for 285; needs review
r6416@totoro (orig r6415): arma | 2006-04-24 17:11:56 -0400
resolve typos in add-tor contrib script
r6417@totoro (orig r6416): nickm | 2006-04-25 01:42:09 -0400
Add some bullet points to write up
r6418@totoro (orig r6417): nickm | 2006-04-25 01:59:31 -0400
fix a segfault in last bug-285-related commit.
r6419@totoro (orig r6418): nickm | 2006-04-25 02:02:46 -0400
you wanted it spelled properly too? And me not even funded!
r6420@totoro (orig r6419): arma | 2006-04-25 02:16:38 -0400
list some more items to remember in path-building
r6421@totoro (orig r6420): nickm | 2006-04-25 02:20:47 -0400
Only warn about a down node once
r6422@totoro (orig r6421): arma | 2006-04-25 03:00:04 -0400
petty cleanups
r6423@totoro (orig r6422): arma | 2006-04-25 03:06:48 -0400
put one of the XXX's back in
r6427@totoro (orig r6426): nickm | 2006-04-29 13:44:31 -0400
shorten some too-wide lines
r6428@totoro (orig r6427): nickm | 2006-04-29 14:42:26 -0400
Start remembering *where* we are storing routerdescs. This will make us easier to move from a RAM-mirrors-disk model to a RAM-caches-disk model, and save maybe around 10MB on a directory server.
r6429@totoro (orig r6428): nickm | 2006-04-29 14:43:05 -0400
Note some subtasks and difficulties involved with reducing RAM usage on dirservers
r6431@totoro (orig r6430): arma | 2006-05-03 14:29:44 -0400
forward-port the ORPort==0 patch
r6434@totoro (orig r6433): arma | 2006-05-03 14:32:15 -0400
forward-port the changelog
r6438@totoro (orig r6437): phobos | 2006-05-03 20:34:51 -0400
Reworked dist-rpm in order to duplicate what dist used to do, but don't actually require dist.
r6444@totoro (orig r6443): weasel | 2006-05-05 11:40:54 -0400
Merge differences between debian_version_0_1_1_18-rc-1 and debian_version_0_1_1_19-rc-1 from tor-0_1_1-patches into head
r6445@totoro (orig r6444): weasel | 2006-05-05 12:58:38 -0400
Handle website/* in tor.nsi
r6448@totoro (orig r6447): arma | 2006-05-08 00:28:49 -0400
stop telling people that "tor -h" will help them in any way.
r6449@totoro (orig r6448): arma | 2006-05-09 05:47:47 -0400
Tor servers are also giving spurious "you're invalid" warnings.
This is because we get a lot of network statuses that don't list
us at all, and we conclude that they all think we're invalid.
The long-term fix is to get better logic, and the short-term fix
is to downgrade the log severity.
r6451@totoro (orig r6450): arma | 2006-05-10 03:35:03 -0400
another piece of doing tor over udp that i am concerned about.
r6452@totoro (orig r6451): arma | 2006-05-10 03:35:33 -0400
a few more tweaks to the faq.
r6453@totoro (orig r6452): weasel | 2006-05-10 06:24:17 -0400
All these headers we get via torint.h
r6455@totoro (orig r6454): phobos | 2006-05-16 01:48:08 -0400
Fixed //Library/Tor in Tor.loc for osx
r6456@totoro (orig r6455): nickm | 2006-05-16 22:18:35 -0400
finally write some comments on tor-spec-udp.txt
r6457@totoro (orig r6456): nickm | 2006-05-21 16:01:13 -0400
apply control-spec patch from Matt Edman: Circuit status only has a path when it has been extended one or more hops.
r6458@totoro (orig r6457): arma | 2006-05-22 00:44:57 -0400
add a few items it would be smart todo
r6459@totoro (orig r6458): arma | 2006-05-22 15:56:32 -0400
remove all the interim changelog stuff for 0.1.1.x
r6460@totoro (orig r6459): arma | 2006-05-22 16:00:12 -0400
my current notes on a 0.1.1.20 changelog
r6461@totoro (orig r6460): arma | 2006-05-22 16:16:18 -0400
other todo tweaks
r6462@totoro (orig r6461): arma | 2006-05-22 16:26:30 -0400
shuffle the todo items some more
r6464@totoro (orig r6463): nickm | 2006-05-22 19:24:06 -0400
Remove string size limit on NEWDESC messages; solve bug 291.
r6466@totoro (orig r6465): arma | 2006-05-22 23:08:30 -0400
add a few more debugging lines to help mikec track down his
11 minute jump into the future.
r6467@totoro (orig r6466): arma | 2006-05-22 23:27:39 -0400
two more todo items that need to be solved during the wsaenobufs quest.
r6468@totoro (orig r6467): arma | 2006-05-23 00:05:45 -0400
claim a few of the todo items.
i guess that means i'm hoping nick does the rest. ;)
r6469@totoro (orig r6468): arma | 2006-05-23 02:20:35 -0400
continue messing with the changelog. it's getting better now.
r6470@totoro (orig r6469): arma | 2006-05-23 03:03:30 -0400
document that runasdaemon has no effect on windows.
r6472@totoro (orig r6471): arma | 2006-05-23 03:04:55 -0400
ignore RunAsDaemon more thoroughly when we're running on windows.
r6474@totoro (orig r6473): nickm | 2006-05-23 04:23:03 -0400
Throw out this UNALIGNED_INT_ACCESS_OK nonsense. Even where it works, it is often way way slower than doing the right thing. Backport candidate.
r6475@totoro (orig r6474): nickm | 2006-05-23 04:38:18 -0400
Patch from Michael Mohr to fix cross-compilation. Backport candidate. Tweaked to use sensible defaults for NULL_REP_IS_ZERO_BYTES and TIME_T_IS_SIGNED.
r6476@totoro (orig r6475): nickm | 2006-05-23 04:50:39 -0400
Add cross.sh cross-compilation script from Michael Mohr. Trivial backport candidate, since adding a new script cannot possibly break anything.
r6477@totoro (orig r6476): nickm | 2006-05-23 04:54:26 -0400
Mark cross-compilation as solved in TODO.
r6478@totoro (orig r6477): arma | 2006-05-23 11:06:05 -0400
trim out the parts of cross.sh that don't make sense now that we've
applied cross-path directly.
also, now we can run cross.sh from the tarball, not just from cvs.
r6479@totoro (orig r6478): arma | 2006-05-23 11:26:51 -0400
another todo item that will make goodell happy
r6480@totoro (orig r6479): arma | 2006-05-23 13:00:49 -0400
final changelog for 0.1.1.20. it is done.
r6487@totoro (orig r6486): weasel | 2006-05-23 14:19:36 -0400
Forward port changelog
r6488@totoro (orig r6487): arma | 2006-05-23 16:15:51 -0400
fix spelling of VirtualAddrNetwork in man page (thanks tup)
r6490@totoro (orig r6489): arma | 2006-05-23 20:21:55 -0400
Claim a commonname of Tor, rather than TOR, in tls handshakes.
Maybe this will help us win the war of names.
r6491@totoro (orig r6490): arma | 2006-05-23 20:37:38 -0400
Stop initializing the hardware accelerator engines simply because
we overloaded the meaning of the argument to crypto_global_init().
r6492@totoro (orig r6491): phobos | 2006-05-24 01:01:29 -0400
Add in the key CFBundleIdentifier required by XCode 2.x and beyond.
r6493@totoro (orig r6492): arma | 2006-05-24 07:13:03 -0400
make options->RedirectExit work again; resolve bug 293.
r6495@totoro (orig r6494): arma | 2006-05-24 19:03:28 -0400
make cookie authentication for the controller work again, maybe.
it sure doesn't now.
r6497@totoro (orig r6496): arma | 2006-05-25 16:06:09 -0400
Stop being picky about what the arguments to mapaddress look like.
we were refusing names that had $ in them, which people who specify
$key.exit will be sad about. There are likely other examples.
If people can think of reasons why we should be picky, let me know.
r6500@totoro (orig r6499): phobos | 2006-05-26 09:22:20 -0400
SUSEisms to enable "make dist-rpm" functionality in SuSe
r6501@totoro (orig r6500): phobos | 2006-05-26 09:32:56 -0400
Use macros in place of details.
r6502@totoro (orig r6501): phobos | 2006-05-26 09:42:28 -0400
Let AC_OUTPUT know about contrib/suse/tor.sh and we only need tor.sh
from contrib/suse
r6503@totoro (orig r6502): arma | 2006-05-26 09:51:20 -0400
build the Makefile in contrib/suse/ too
r6504@totoro (orig r6503): arma | 2006-05-26 09:51:45 -0400
resolve an unused variable
r6506@totoro (orig r6505): arma | 2006-05-26 12:29:20 -0400
correct a false log message, since we actually reset all our
downloading stats every hour, and sometimes more often.
r6507@totoro (orig r6506): arma | 2006-05-26 12:29:33 -0400
be more verbose about testing reachability of our ORPort.
r6508@totoro (orig r6507): arma | 2006-05-26 12:32:16 -0400
if we're a server and some peer has a broken tls certificate, don't
shout about it unless we want to hear about protocol violations.
r6509@totoro (orig r6508): arma | 2006-05-28 12:07:44 -0400
clean up a comment
r6510@totoro (orig r6509): arma | 2006-05-28 12:14:26 -0400
directory authorities should be more tolerant of failed reachability
tests before crying foul to the server operator.
r6511@totoro (orig r6510): nickm | 2006-05-28 12:54:39 -0400
Add a basic mmap function, with a "fake-it" wrapper to do read_file_from_str instead. Based on code from Michael Mohr.
r6512@totoro (orig r6511): arma | 2006-05-30 01:05:50 -0400
remove a few things from the 0.1.2 todo, and add one
r6513@totoro (orig r6512): arma | 2006-05-30 01:29:03 -0400
simplify a log message
r6514@totoro (orig r6513): arma | 2006-05-30 02:11:36 -0400
tentative change: if you have your dirport set, you are a directory
mirror, whether or not your orport is set.
r6515@totoro (orig r6514): arma | 2006-05-30 02:11:46 -0400
and clarify the spec to say this too.
r6516@totoro (orig r6515): arma | 2006-05-30 02:17:28 -0400
END_CIRC_REASON_OR_IDENTITY apparently means that we were told to
connect to a different OR than lives on the addr:port we connected
to. we don't actually remember whether that was the case, currently.
so call it END_CIRC_REASON_OR_CONN_CLOSED as a compromise.
r6517@totoro (orig r6516): arma | 2006-05-30 02:19:06 -0400
stop fetching descriptors if we're not a dir mirror and we
haven't tried to establish any circuits lately.
r6518@totoro (orig r6517): arma | 2006-05-30 02:19:48 -0400
and get grammar right
r6519@totoro (orig r6518): arma | 2006-05-30 02:23:44 -0400
connection_t kept the identity_pkey but all it did was store it and
free it. perhaps we don't need it after all?
r6520@totoro (orig r6519): arma | 2006-05-30 02:36:32 -0400
mark off a todo item. i'll put it back if it turns out it
doesn't work.
r6521@totoro (orig r6520): nickm | 2006-05-30 16:41:22 -0400
Rearrange TODO.
r6522@totoro (orig r6521): arma | 2006-06-01 04:43:56 -0400
update the explanation for deprecating v0 control spec.
r6523@totoro (orig r6522): arma | 2006-06-02 22:56:44 -0400
don't stop fetching server descriptors if we're a server and
haven't found ourselves reachable yet.
r6524@totoro (orig r6523): nickm | 2006-06-03 14:52:31 -0400
Patch based on post by Mike C to or-dev; special-case based on use of MSVC, rather than on MS_WINDOWS, so that mingw builds.
r6525@totoro (orig r6524): nickm | 2006-06-03 15:49:42 -0400
Add async dns code from Adam Langley, tweaked to build on OSX. Long-term, we may want to switch to libevnet/c-ares, if they ever handle 10k fd situations properly. This one still needs work too, but at least it is small. This code is disabled by default, and not integrated with dns.c.
r6526@totoro (orig r6525): nickm | 2006-06-03 16:52:24 -0400
Make dns.c use eventdns.c -- but only when you pass the --enable-eventdns argument to configure.in. This will *so* not work on Windows yet.
r6527@totoro (orig r6526): nickm | 2006-06-03 17:41:14 -0400
More DNS fixes. Send meaningful TTLs back to the client when
possible. Cache at the server side independently from the TTL, to
prevent attackers from probing the server to see who has been asking
for what hostnames. (Hi, Dan Kaminski!)
Also, clean some whitespace.
r6528@totoro (orig r6527): nickm | 2006-06-03 17:47:26 -0400
Oops. When we dont get a TTL, we should default to the default, not to the minimum.
r6529@totoro (orig r6528): nickm | 2006-06-03 18:05:23 -0400
Make eventdns.[ch] into good C90; remove signed/unsigned comparisons.
r6530@totoro (orig r6529): arma | 2006-06-04 02:16:20 -0400
punctuation and spelling
r6531@totoro (orig r6530): arma | 2006-06-04 02:17:32 -0400
if we insist on printing pointer values, at least make it stop
complaining on (my particular) 64 bit platform.
r6533@totoro (orig r6532): nickm | 2006-06-04 18:42:13 -0400
Add a new warning to our "warn a lot" list: unused parameters. This means we have to explicitly "use" unuseds, but it can catch bugs. (It caught two coding mistakes so far.)
r6534@totoro (orig r6533): nickm | 2006-06-04 19:23:53 -0400
Some eventdns.c fixes for windows correctness. More will doubtless be needed, especially around the #includes.
r6535@totoro (orig r6534): nickm | 2006-06-04 20:32:31 -0400
Hm. Where did we put that ntohl the last time we were juggling it? (hoop-lah).
r6536@totoro (orig r6535): nickm | 2006-06-04 21:59:12 -0400
More eventdns.c patches: use HAVE_ALLOCA_H; print IP addrs as dotted quads.
r6537@totoro (orig r6536): arma | 2006-06-05 00:29:03 -0400
bandaid for bug 299. this is still a bug, since we don't
initialize for hardware acceleration in certain configurations;
but not critical until that is supported.
r6539@totoro (orig r6538): arma | 2006-06-05 03:27:48 -0400
Note a bug that causes servers to sometimes never send the pending
create cell. Nick, is this a bug? If so, is my fix right?
r6540@totoro (orig r6539): arma | 2006-06-05 04:02:04 -0400
remove some unused code (i think)
r6541@totoro (orig r6540): arma | 2006-06-05 04:25:02 -0400
simplify some code, since circuit_build_failed() is only called
on non-open circuits.
r6542@totoro (orig r6541): arma | 2006-06-05 04:58:18 -0400
bugfix: if we are making our first ever connection to any entry
guard, then don't mark it down at first. we had this implemented
but it was disabled due to a bug.
r6544@totoro (orig r6543): arma | 2006-06-05 05:08:10 -0400
simplify code now that libevent considers all sockets pollable.
what we really mean now is ">= 0", which is clearer to test for.
r6545@totoro (orig r6544): arma | 2006-06-05 05:47:19 -0400
scream louder if you've got a pending circuit for a given
addr/port but the intended n_conn digest is wrong.
r6546@totoro (orig r6545): arma | 2006-06-05 05:51:29 -0400
whoops, add a man page entry for ProtocolWarnings
r6547@totoro (orig r6546): arma | 2006-06-05 06:01:52 -0400
don't tell people that the testing circuit failed if we already
consider ourselves reachable. this just confuses them.
r6548@totoro (orig r6547): nickm | 2006-06-05 19:01:22 -0400
Try to log useful messages at info and debug about what we are resolving and what answers we are getting wrt eventdns.
r6549@totoro (orig r6548): arma | 2006-06-05 20:04:52 -0400
ship the event*.h files too. perhaps this will make my 'make dist'
produce a tarball that i can build.
r6550@totoro (orig r6549): arma | 2006-06-05 20:05:39 -0400
fix typo
r6551@totoro (orig r6550): arma | 2006-06-05 20:06:52 -0400
We got an obscure report of an assert error on a windows Tor server with
connection_add being called with socket = -1. The only places I can see
where that happen would be if our tor_socketpair succeeds but it hands
back negative values for some of its file descriptors. Perhaps this will
produce more useful assert errors next time.
r6552@totoro (orig r6551): nickm | 2006-06-05 20:12:22 -0400
Also, add a temporary hack to make sure eventdns.c is distributd.
r6553@totoro (orig r6552): arma | 2006-06-05 23:33:24 -0400
fix the bug where we sometimes would fail to send some create cells
once we'd connected to a(nother) tor server.
r6557@totoro (orig r6556): arma | 2006-06-06 22:57:23 -0400
looks like we missed a piece of the 0.1.1.9 paranoia code.
hopefully this change is a no-op.
r6558@totoro (orig r6557): arma | 2006-06-07 02:10:54 -0400
simplify the tortls api: we only support being a "server", that
is, even tor clients do the same sort of handshake.
this has been true for years, so it's best to get rid of the
stale code.
r6559@totoro (orig r6558): arma | 2006-06-07 02:21:11 -0400
and now the exciting part: there is now no such thing as doing
a client-only tls, that is, one with no certs.
r6560@totoro (orig r6559): arma | 2006-06-07 02:53:43 -0400
the CookieAuthentication section in our spec seems to assume we're
still using the v0 control protocol.
r6561@totoro (orig r6560): arma | 2006-06-07 03:11:42 -0400
make connection_or_nonopen_was_started_here() based on something less
voodooey. it turns out we already do keep a flag like that around.
r6562@totoro (orig r6561): arma | 2006-06-07 04:42:24 -0400
put a bandaid in place so servers will have an easier time
believing that they're reachable. this may help resolve the
servers-on-dynamic-ip-addresses problem.
r6564@totoro (orig r6563): arma | 2006-06-07 05:18:53 -0400
re-enable per-connection rate limiting. get rid of the "OP bandwidth"
concept. lay groundwork for "bandwidth classes" -- separate global
buckets that apply depending on what sort of conn it is.
r6566@totoro (orig r6565): arma | 2006-06-08 05:20:58 -0400
ah, that explains why we weren't going dormant with
respect to descriptor fetches. maybe now it will work.
r6567@totoro (orig r6566): arma | 2006-06-08 05:35:20 -0400
remove a bit more obsolete code
r6569@totoro (orig r6568): arma | 2006-06-08 18:36:13 -0400
ok, ok, maybe *this* time my rep_hist_circbuilding_dormant() will work.
r6570@totoro (orig r6569): arma | 2006-06-08 22:20:42 -0400
try a better string at the top of torrc's autogenerated torrc.
r6571@totoro (orig r6570): arma | 2006-06-08 22:45:39 -0400
fix a bootstrapping check we ignored that prevents us from running
with only one dir authority.
r6572@totoro (orig r6571): arma | 2006-06-09 02:35:45 -0400
Bandaid for a seg fault i just got in 0.1.1.20.
More generally, i reopened bug 222. Whee.
r6574@totoro (orig r6573): arma | 2006-06-09 02:52:49 -0400
and forward-port too.
r6576@totoro (orig r6575): arma | 2006-06-09 05:02:32 -0400
when only one router is labelled as a guard, and we've
already picked him, we would cycle endlessly picking him
again, being unhappy about it, and so forth.
now we specifically exclude guards when picking a new guard.
r6577@totoro (orig r6576): arma | 2006-06-09 05:07:59 -0400
actually, don't fix it that far. we should still do some
error checking.
r6579@totoro (orig r6578): nickm | 2006-06-09 11:57:58 -0400
Override our notion of printability for esc_for_log. 127 and up are never printable. Take that, locales.
r6580@totoro (orig r6579): nickm | 2006-06-09 13:07:22 -0400
Another escape() fix, for picky sprintfs.
r6582@totoro (orig r6581): arma | 2006-06-09 20:26:39 -0400
Add a new config option TestVia, that lets you specify preferred middle
hops to use for testing circuits. Perhaps this will let me debug the
reachability problem better.
r6583@totoro (orig r6582): arma | 2006-06-09 20:30:49 -0400
clean up formatting in the man page
r6584@totoro (orig r6583): arma | 2006-06-09 20:32:14 -0400
add TestVia to the man page
r6586@totoro (orig r6585): arma | 2006-06-09 20:57:12 -0400
take out the reachability bandaid in 0.1.2.x as well.
maybe we will actually be able to fix it, instead.
r6596@totoro (orig r6595): phobos | 2006-06-10 01:37:17 -0400
Remove echo -n to make start script slightly more readable on boot.
r6600@totoro (orig r6599): arma | 2006-06-10 21:41:30 -0400
interim changelog for 0.1.2.1-alpha
r6601@totoro (orig r6600): arma | 2006-06-10 21:42:21 -0400
forward-port the 0.1.1.21 changelog.
r6604@totoro (orig r6603): weasel | 2006-06-11 20:49:07 -0400
Forward port changelog
r6605@totoro (orig r6604): arma | 2006-06-12 02:03:15 -0400
allow people to start their tor with runasdaemon set but
with no logs set at all.
r6606@totoro (orig r6605): arma | 2006-06-12 06:44:00 -0400
typo, whitespace, and a clarification
r6607@totoro (orig r6606): arma | 2006-06-12 07:59:19 -0400
Finally solve the "closing wedged cpuworkers" bug. Woo.
This happened when we got two create cells in a row from the same TLS
connection. It would hand one to the cpuworker, and then immediately
handle the second one -- after it had registered that the first one was
busy, but before it had updated the timestamp that we use to decide how
*long* it's been busy.
r6609@totoro (orig r6608): weasel | 2006-06-12 18:03:25 -0400
Make the Exit tag in status documents actually work in head too
r6610@totoro (orig r6609): arma | 2006-06-12 22:48:06 -0400
reintroduce the logic to exit_policy_is_general_exit() to count
how many ports are allowd. require two ports open, not just one.
r6611@totoro (orig r6610): arma | 2006-06-13 01:36:35 -0400
Fix the bug that was causing servers to not find themselves
reachable if they changed IP addresses. This happened because
middle servers knew the old descriptor, and kept swapping the
addr/port we asked for with the one they thought was right.
So the create cell never got sent, because it was asking for a
different addr/port than we believed we had connected to.
r6614@totoro (orig r6613): arma | 2006-06-13 01:50:24 -0400
harmless typo
r6615@totoro (orig r6614): arma | 2006-06-13 01:51:28 -0400
Defense in depth: fix the reachability bug a second way too.
Now if we establish a connection with the right digest, regardless
of what the addr/port is, and we have pending create cells, use it.
r6616@totoro (orig r6615): arma | 2006-06-13 05:16:09 -0400
export the default exit policy via the control port, so controllers
don't need to guess what it is / will be later.
r6617@totoro (orig r6616): arma | 2006-06-13 06:25:22 -0400
first cut at a workaround for the reachability bug: explicitly find a
server running the right version, if we can, and ask for that one.
r6618@totoro (orig r6617): arma | 2006-06-13 06:48:26 -0400
bugfix in exit_policy_is_general_exit() that weasel found.
this time for sure!
r6619@totoro (orig r6618): arma | 2006-06-13 07:11:19 -0400
now we can tell dirserv_dump_directory_to_string() whether we want
it to include down/invalid descriptors or not.
r6620@totoro (orig r6619): arma | 2006-06-13 08:05:59 -0400
be more lax about recognizing valid hexdigests.
r6621@totoro (orig r6620): arma | 2006-06-13 08:57:19 -0400
back off and add the $ at the beginning of the preferrednodes list
we generate.
r6622@totoro (orig r6621): nickm | 2006-06-13 17:49:56 -0400
eventdns: Apply a couple of patches from AGL; start working on windows compat; note some TODOs.
r6623@totoro (orig r6622): arma | 2006-06-14 07:06:43 -0400
bugfix: discourage picking directory authorities as our TestVia
hops, even if they're running the right versions, since we probably
already have a connection established to them.
r6624@totoro (orig r6623): arma | 2006-06-14 18:28:16 -0400
upgrade the severity of the 'clock jump' warn, and ask people
to report if it occurs.
r6625@totoro (orig r6624): arma | 2006-06-14 19:21:22 -0400
start checking for limits.h too. we should resume compiling
on irix64 and other weird platforms now.
r6626@totoro (orig r6625): arma | 2006-06-15 05:03:15 -0400
lower the number of seconds before we yell about clock jump.
and make the yelling only happen if you're a server.
r6627@totoro (orig r6626): arma | 2006-06-15 18:32:00 -0400
fix recommended url in torrc.sample for server sign-up
r6628@totoro (orig r6627): weasel | 2006-06-15 18:52:56 -0400
Add a /tor/dir-all-weaselhack directory resource so I do not have to update my scripts
r6629@totoro (orig r6628): arma | 2006-06-15 18:59:07 -0400
fix spacing
r6630@totoro (orig r6629): weasel | 2006-06-15 19:14:01 -0400
And a minor bugfix to the weaselhack
r6631@totoro (orig r6630): weasel | 2006-06-15 19:20:50 -0400
Forward port 07_log_to_file_by_default.dpatch
r6632@totoro (orig r6631): arma | 2006-06-15 20:04:46 -0400
clean up man page. expand on contactinfo a bit.
r6633@totoro (orig r6632): weasel | 2006-06-15 22:04:04 -0400
Fix configure.in to not produce broken configure files with more recent
versions of autoconf. Thanks to Clint for his auto* voodoo.
r6634@totoro (orig r6633): nickm | 2006-06-16 11:40:57 -0400
Clarify mmap and memory-use hacks.
r6635@totoro (orig r6634): nickm | 2006-06-18 03:21:35 -0400
Add smartlist_reverse and smartlist_pop_last.
r6636@totoro (orig r6635): nickm | 2006-06-18 03:22:36 -0400
Add a memdup function to util
r6637@totoro (orig r6636): nickm | 2006-06-18 03:24:29 -0400
Add some incremental encryption wrappers to torgzip code
r6638@totoro (orig r6637): nickm | 2006-06-18 03:27:47 -0400
Part of incremental encryption logic for buffers: there is a subtle yucky point documented in a comment.
r6639@totoro (orig r6638): nickm | 2006-06-18 03:32:31 -0400
perhaps the reason I rail against cut-and-paste programming so vehemently is that I am so bad at it.
r6640@totoro (orig r6639): nickm | 2006-06-18 03:35:10 -0400
Add tests for several of the more recently committed functions.
r6641@totoro (orig r6640): nickm | 2006-06-18 03:37:21 -0400
remove non-germane comment
r6642@totoro (orig r6641): nickm | 2006-06-18 03:38:55 -0400
Instead of adding servers and v1 directories to buffers en masse, directory servers add them on the fly as their outbufs are depleted. This will save ram on busy dirservers.
r6643@totoro (orig r6642): nickm | 2006-06-18 03:55:04 -0400
Oops. conn->requested_resource is client only.
r6644@totoro (orig r6643): nickm | 2006-06-18 03:57:47 -0400
add coverage for a default case
r6645@totoro (orig r6644): nickm | 2006-06-18 04:07:16 -0400
Fix a couple of bugs in last patch.
r6646@totoro (orig r6645): nickm | 2006-06-18 04:13:45 -0400
That dir_refresh_src fix will only work if I enable it.
r6647@totoro (orig r6646): nickm | 2006-06-18 04:16:05 -0400
And actually check the url when it exists. that might work better.
r6648@totoro (orig r6647): nickm | 2006-06-18 04:19:35 -0400
Stop trying to refresh when we are out of data.
r6649@totoro (orig r6648): nickm | 2006-06-18 04:21:27 -0400
make zlib buffer function set buf_highwater properly
r6650@totoro (orig r6649): nickm | 2006-06-18 04:44:34 -0400
Temporarily disable sentinels on buffers
r6651@totoro (orig r6650): nickm | 2006-06-18 04:46:55 -0400
write_to_buf != connection_write_to_buf. Also, add a connection_write_to_buf_zlib wrapper that sucks.
r6652@totoro (orig r6651): nickm | 2006-06-18 04:53:09 -0400
Make connection_write_to_buf_zlib set outbuf_flushlen right.
r6653@totoro (orig r6652): nickm | 2006-06-18 05:03:48 -0400
Another _zlib fix.
r6654@totoro (orig r6653): nickm | 2006-06-18 11:53:54 -0400
Re-enable buffer RAM guard values.
r6655@totoro (orig r6654): nickm | 2006-06-18 12:05:54 -0400
Backport candidate: implement the "is this uptime change cosmetic" test properly.
r6657@totoro (orig r6656): nickm | 2006-06-18 12:20:38 -0400
another write_to_buf_zlib fix.
r6658@totoro (orig r6657): nickm | 2006-06-18 12:39:26 -0400
Ah. That seems to work.
r6659@totoro (orig r6658): nickm | 2006-06-18 16:39:46 -0400
Resolve control flow warning.
r6660@totoro (orig r6659): nickm | 2006-06-18 16:58:27 -0400
Become capable of noticing that we are done sending a directory.
r6661@totoro (orig r6660): arma | 2006-06-18 17:07:45 -0400
correct a function comment in compute_preferred_testing_list()
r6663@totoro (orig r6662): nickm | 2006-06-18 17:15:01 -0400
Fix an assert that still isnt the assert we are hunting.
r6664@totoro (orig r6663): nickm | 2006-06-18 17:30:03 -0400
Fix a tricky crash: making the_directory heap-allocated (so we could refcount it and have multiple instances as neeeded) means that calls to dirserv_regenerate_directory could invalidate the auth_dir value passed to dirserv_pick_cached_dir_obj. Big fun.
r6665@totoro (orig r6664): nickm | 2006-06-19 20:48:23 -0400
Start spooling v2 networkstatus docs as well.
r6666@totoro (orig r6665): phobos | 2006-06-20 00:16:46 -0400
First crack at version checking for OSX installer.
r6667@totoro (orig r6666): phobos | 2006-06-20 00:18:47 -0400
Minor fix to pre-instllation version check for OSX
r6668@totoro (orig r6667): nickm | 2006-06-20 02:27:13 -0400
Ah. We need a new zlib_state for each networkstatus we spool out.
r6669@totoro (orig r6668): arma | 2006-06-20 09:14:07 -0400
add a note for nick to fix
r6670@totoro (orig r6669): nickm | 2006-06-20 12:48:32 -0400
Fishy, but harmless.
r6671@totoro (orig r6670): nickm | 2006-06-20 19:06:52 -0400
Fix bug in networkstatus spooling: spool more than the first networkstatus.
r6672@totoro (orig r6671): weasel | 2006-06-20 19:11:15 -0400
<nickm> ooh, that log shouldn't be there. can you take it out?
r6673@totoro (orig r6672): phobos | 2006-06-20 22:27:18 -0400
Remove the osx version requirements on install due to a messy Installer
situation between pre-panther, tiger and beyond, and metapackages.
r6674@totoro (orig r6673): nickm | 2006-06-21 00:57:12 -0400
When requesting or serving resources via fingerprint/digest, request and respond in-order, removing duplicates.
r6675@totoro (orig r6674): weasel | 2006-06-21 18:13:03 -0400
Fix 07_log_to_file_by_default in debian head
r6676@totoro (orig r6675): nickm | 2006-06-22 03:01:54 -0400
Next batch of memory miserdom: mmap cached-routers file. This is sure to break somewhere.
r6677@totoro (orig r6676): nickm | 2006-06-22 03:10:37 -0400
#if out test that was failing because of an extra newline.
r6678@totoro (orig r6677): nickm | 2006-06-22 03:19:28 -0400
Set offset properly when parsing cache.
r6679@totoro (orig r6678): nickm | 2006-06-22 03:25:15 -0400
Fix a bunch of spaces.
r6680@totoro (orig r6679): nickm | 2006-06-22 03:29:14 -0400
Ooh, that could have been bad. Sort digests as digests, not strings.
r6681@totoro (orig r6680): nickm | 2006-06-22 03:34:04 -0400
Make some more verbose gcc warnings go away.
r6682@totoro (orig r6681): nickm | 2006-06-22 03:49:41 -0400
Mark some more TODO items done.
r6683@totoro (orig r6682): nickm | 2006-06-23 22:06:52 -0400
Apparently, zlib sometimes reports Z_BUF_ERROR on input exhaustion as well as on running out of output space. This could well fix the assert bug reported by weasel and arma.
r6684@totoro (orig r6683): nickm | 2006-06-23 22:10:21 -0400
Turn a while into a do/while; save a redundant test
r6685@totoro (orig r6684): arma | 2006-06-24 00:57:59 -0400
refuse to write an iso_time which we can't parse, when dumping
bandwidth state. this fixes the particular incident in bug 308,
but the general issue remains.
r6686@totoro (orig r6685): phobos | 2006-06-25 00:02:43 -0400
Create binary osx un-installer, update perms on install so anyone can run the
uninstaller.
r6689@totoro (orig r6688): phobos | 2006-06-25 00:07:24 -0400
Fix the messed up commit.
r6690@totoro (orig r6689): phobos | 2006-06-25 00:08:21 -0400
And, commit it correctly.
r6693@totoro (orig r6692): arma | 2006-06-27 07:23:10 -0400
specify the dir spec better (suggested by lexi)
r6694@totoro (orig r6693): nickm | 2006-06-27 11:52:51 -0400
Likely fix for bug 309: when we calculate offsets after rebuilding the descriptor cache, do not reset the offset pointer half-way through.
r6695@totoro (orig r6694): nickm | 2006-06-28 04:54:32 -0400
Actually enable mmap. That should improve matters.
r6696@totoro (orig r6695): nickm | 2006-06-28 04:55:53 -0400
Make sure that our calculated offsets for routers is correct; again. This time bug 309 may be gone gone gone.
r6697@totoro (orig r6696): nickm | 2006-06-28 04:57:41 -0400
Aaand re-disable the bogus test in get_body()
r6698@totoro (orig r6697): nickm | 2006-06-28 07:03:34 -0400
Add a check to try to make cache rebuild fail fast if it is going to fail
r6699@totoro (orig r6698): nickm | 2006-06-28 11:36:28 -0400
Fix another idiot bug causing symptom 309. Why cant I program?
r6700@totoro (orig r6699): arma | 2006-06-28 11:39:02 -0400
tab-man strikes again
r6701@totoro (orig r6700): nickm | 2006-06-29 07:04:42 -0400
complete_only == !allow_partial. This enables useful use of partial desc downloads. Backport candidate
r6702@totoro (orig r6701): nickm | 2006-06-29 07:17:36 -0400
Harmless: Z_OK is not an acceptable answer to Z_FINISH.
r6703@totoro (orig r6702): nickm | 2006-06-29 07:19:52 -0400
Apparent 311 fix: apparently passing Z_FINISH an empty string is problematic.
r6704@totoro (orig r6703): arma | 2006-06-29 09:10:08 -0400
remove the word 'middleman' from the sample torrc
r6705@totoro (orig r6704): arma | 2006-06-29 09:11:23 -0400
also remove word 'middleman' from a log notice
r6706@totoro (orig r6705): nickm | 2006-06-30 06:50:43 -0400
Unify HTTP response code into one place so it is easier to add headers. Add an X-You-Are header, which we should probably rename. NOTE that we should not use this field for things where it matters if dirs lie.
r6707@totoro (orig r6706): nickm | 2006-06-30 06:52:12 -0400
Add a .cvsignore file for contrib/suse/
r6708@totoro (orig r6707): nickm | 2006-07-01 17:51:21 -0400
Oops. Headers work better when named right.
r6710@totoro (orig r6709): arma | 2006-07-03 23:19:59 -0400
minor fixes
r6711@totoro (orig r6710): arma | 2006-07-03 23:25:07 -0400
minor fixes
r6712@totoro (orig r6711): arma | 2006-07-03 23:27:09 -0400
No longer permit create cells to have the wrong circ_id_type. No
running Tors should still have this bug.
r6713@totoro (orig r6712): arma | 2006-07-03 23:31:27 -0400
Get rid of the router_retry_connections notion. Now routers no longer
try to rebuild long-term connections to directory authorities, and
directory authorities no longer try to rebuild long-term connections to
all servers.
We still don't hang up connections in these two cases though -- we need
to look at it more carefully to avoid flapping, and we likely need to
wait til 0.1.1.x is obsolete.
r6714@totoro (orig r6713): arma | 2006-07-03 23:33:17 -0400
touch up the TODO and HACKING files
r6715@totoro (orig r6714): arma | 2006-07-03 23:39:01 -0400
mention in the tor dmg instructions that you may need to remove your
old shared library libevent, lest your linker get confused.
r6716@totoro (orig r6715): arma | 2006-07-03 23:40:45 -0400
a first attempt at specifying HELLO cells. plus general cleanup on tor-spec.
r6717@totoro (orig r6716): arma | 2006-07-04 11:51:59 -0400
if we're the server-side of the tls and there are problems,
don't yell as loudly.
r6718@totoro (orig r6717): arma | 2006-07-04 11:52:22 -0400
fix a misleading function comment
r6720@totoro (orig r6719): arma | 2006-07-04 12:07:49 -0400
name the HELLO version the "link version"
r6721@totoro (orig r6720): arma | 2006-07-04 12:11:35 -0400
ok, i'm not allowed to say that there. oh well.
r6723@totoro (orig r6722): arma | 2006-07-04 14:18:08 -0400
Make the X-You-Are header more accurate when there's a proxy in the middle.
r6727@totoro (orig r6726): arma | 2006-07-04 16:25:17 -0400
oops, we were ignoring options->ExcludeNodes when picking entry guards.
it is still the case that we ignore it with respect to entry guards
that we've already picked.
r6728@totoro (orig r6727): arma | 2006-07-05 14:19:42 -0400
actually, that excludenodes fix was redundant. take it out.
r6729@totoro (orig r6728): nickm | 2006-07-05 17:28:37 -0400
Add plausile logging support to eventdns; stop putting stuff onto stdout.
r6730@totoro (orig r6729): nickm | 2006-07-05 17:33:46 -0400
Spelling fix.
r6731@totoro (orig r6730): nickm | 2006-07-05 17:42:18 -0400
Clean up eventdns messages.
r6734@totoro (orig r6733): arma | 2006-07-05 22:44:07 -0400
when an exit node gets a malformed begin cell, don't complain to
the node operator, since he can't do anything about it.
r6735@totoro (orig r6734): arma | 2006-07-05 22:45:46 -0400
whitespace/tab fixes
r6736@totoro (orig r6735): arma | 2006-07-05 23:05:01 -0400
rename X-You-Are to something slightly better.
r6737@totoro (orig r6736): phobos | 2006-07-06 11:51:07 -0400
Update osx binary un-installer naming, remove invalid osx binary
un-installer app
r6739@totoro (orig r6738): arma | 2006-07-06 12:19:00 -0400
forward-port the 0.1.1.22 changelog, minus the line about
the osx uninstaller.
r6740@totoro (orig r6739): phobos | 2006-07-06 12:28:19 -0400
Remove osx binary uninstaller changes. Tiger and Panther won't play
nice in the same way.
r6742@totoro (orig r6741): phobos | 2006-07-06 20:54:39 -0400
Binary OSX un-installer tarball
r6743@totoro (orig r6742): phobos | 2006-07-06 20:55:12 -0400
Updated OSX binary un-installer for testing.
r6744@totoro (orig r6743): nickm | 2006-07-07 13:31:56 -0400
eventdns: check for malloc() failures.
r6745@totoro (orig r6744): nickm | 2006-07-07 13:33:30 -0400
When using eventdns: suppress logging of addresses when SafeLogging is active, and make set of nameservers configurable from torrc.
r6746@totoro (orig r6745): nickm | 2006-07-07 15:08:44 -0400
Spellcheck and remove spurious include in eventdns.c
r6747@totoro (orig r6746): arma | 2006-07-08 13:38:46 -0400
Fix a crash if you enable FascistFirewall but not FirewallPorts.
Reported by Frediano Ziglio.
r6749@totoro (orig r6748): nickm | 2006-07-09 18:28:12 -0400
First part of making mmap-based stuff work on win32: save descriptors as "binary" (no LF->CRLF tanslation) so that we can mmap them properly later. Patch from Frediano Ziglio.
r6750@totoro (orig r6749): nickm | 2006-07-09 18:29:12 -0400
Make compilation work on old MSVCs without GetVertsionEx magic. Patch from Frediano Ziglio.
r6751@totoro (orig r6750): nickm | 2006-07-09 18:33:21 -0400
Fix project file for MSVC6 (!). Patch from Frediano Ziglio.
r6752@totoro (orig r6751): nickm | 2006-07-10 14:38:57 -0400
Add a const; fix a (probably harmless) bug when storing a resolve we forgot we asked for.
r6753@totoro (orig r6752): nickm | 2006-07-10 23:33:16 -0400
OR_CONN_EVENT_NEW: we should probably handle that , should we not? Especially since 23:26 < phobos> nickm: grep -c "Unrecognized status code 4" tor.log
r6754@totoro (orig r6753): phobos | 2006-07-11 00:27:12 -0400
First crack at launchd plist for Tor
r6755@totoro (orig r6754): phobos | 2006-07-11 00:37:05 -0400
Add in start parameters. Perhaps we should ship these set correct in
the default torrc.
r6756@totoro (orig r6755): phobos | 2006-07-11 16:51:06 -0400
Move cmd line parameters into the config file, remove chroot, and
satisfy the requirement of a ProgramArgument array
r6757@totoro (orig r6756): nickm | 2006-07-11 16:51:58 -0400
Add some debugging asserts to dns.c; these are too expensive to leave in permanently.
r6758@totoro (orig r6757): phobos | 2006-07-12 18:09:21 -0400
Update config to redirect output to the tor log file for now.
r6759@totoro (orig r6758): arma | 2006-07-13 23:14:02 -0400
Avoid an integer underflow when the dir authority decides whether a
router is stable: we might wrongly label it stable, and compute a slightly
wrong median stability, when a descriptor is published later than now.
Inspired by Matt's Vidalia checkin:
http://trac.vidalia-project.net/changeset/1074
r6761@totoro (orig r6760): arma | 2006-07-15 01:49:57 -0400
minor tweak on the dir spec
r6762@totoro (orig r6761): arma | 2006-07-15 01:50:22 -0400
mention the existence of dir-spec in tor-spec, and note that we need
to update it.
r6763@totoro (orig r6762): arma | 2006-07-15 01:53:41 -0400
add a whole lot more work to the todo.
r6764@totoro (orig r6763): arma | 2006-07-15 15:21:30 -0400
stick to nick's nul/null convention
r6765@totoro (orig r6764): arma | 2006-07-15 16:26:05 -0400
parameterize the loudness of resolve_my_address(), and call things
IP addresses, not IPs.
r6766@totoro (orig r6765): arma | 2006-07-16 01:57:11 -0400
whitespace/etc cleanups
r6767@totoro (orig r6766): phobos | 2006-07-16 09:58:10 -0400
Change the way Tor starts on OSX 10.4 vs pre-10.4. 10.4 Tor now uses
launchd for current and forward compatibility.
r6768@totoro (orig r6767): phobos | 2006-07-16 14:18:40 -0400
Remove a merged if-then from line 85
r6769@totoro (orig r6768): nickm | 2006-07-16 20:39:05 -0400
MSVC6 is apparently terrified of unnatural cross-breeding between uint64_t and double, and needs more persuasion than usual to cast one to the other. Issue identified by Frediano Ziglio; patch revised for minimal impact on non-MSVC6 compilers.
r6770@totoro (orig r6769): phobos | 2006-07-16 23:57:21 -0400
Added net.freehaven.tor.plist to AC_OUTPUT
r6771@totoro (orig r6770): phobos | 2006-07-17 00:31:22 -0400
Minor fixed for launchd xml plist
r6772@totoro (orig r6771): arma | 2006-07-17 01:12:54 -0400
nick suggests that the hello cell should have both server IP and
client IP. he's right.
r6773@totoro (orig r6772): arma | 2006-07-17 02:20:09 -0400
fix wordo
r6774@totoro (orig r6773): arma | 2006-07-17 02:26:19 -0400
we are constrained more than we realized, on what g^x values we can
accept or refuse.
r6775@totoro (orig r6774): arma | 2006-07-17 02:35:06 -0400
Allow servers with no hostname or IP address to learn their IP address
by asking the directory authorities. This code only kicks in when you
would normally have exited with a "no address" error.
This design is flawed, though, since the X-Your-Address-Is header is not
authenticated, and doing it this way introduces too many new attacks. The
right answer is to give IP address hints inside the HELLO cell; much of
this code can be reused when we switch.
r6776@totoro (orig r6775): arma | 2006-07-17 02:54:28 -0400
fix some more places where we shouldn't crash if we can't build
our own descriptor yet.
r6778@totoro (orig r6777): arma | 2006-07-17 02:59:56 -0400
and don't try to build the descriptor every second, if it's dirty
but we don't have a known address.
r6779@totoro (orig r6778): arma | 2006-07-17 04:11:27 -0400
huge bugfix: we weren't ever writing an http header when sending
out network statuses! so clients were downloading the whole thing,
and then discarding them because they're malformed.
r6780@totoro (orig r6779): arma | 2006-07-17 04:17:51 -0400
Make a louder statement the first time we learn a guessed
IP address.
r6781@totoro (orig r6780): arma | 2006-07-17 15:33:54 -0400
parameterize the loudness of log_addr_has_changed(), since it's the
only place where we inform the user of a new IP address, if we're
guessing it from external sources.
r6782@totoro (orig r6781): arma | 2006-07-17 15:42:22 -0400
shuffle todo items
r6783@totoro (orig r6782): arma | 2006-07-17 20:01:12 -0400
pick a log domain; resolve an xxxx
r6784@totoro (orig r6783): arma | 2006-07-17 20:59:46 -0400
If we are using an exit enclave and we can't connect, e.g. because
its webserver is misconfigured to not listen on localhost, then back
off and try connecting from somewhere else before we fail.
r6785@totoro (orig r6784): nickm | 2006-07-17 22:01:32 -0400
Hm. We probably should define INT64_MAX if we really want it. (Especially since we only want it on one platform, where, coincidentally, it is not defined.)
r6786@totoro (orig r6785): nickm | 2006-07-17 22:24:01 -0400
Oh. And apparently, msvc6 doesnt think very much of doing u64-and-double arithmetic either.
r6787@totoro (orig r6786): arma | 2006-07-17 23:06:12 -0400
Start publishing one minute or so after we find our ORPort
to be reachable. This will help reduce the number of descriptors
we have for ourselves floating around, since it's quite likely
other things (e.g. DirPort) will change during that minute too.
r6788@totoro (orig r6787): arma | 2006-07-17 23:06:55 -0400
when we find our dirport to be reachable, mark our descriptor
dirty so we'll tell the world. (fixes bug 306 reported by pnx)
r6789@totoro (orig r6788): phobos | 2006-07-18 00:37:43 -0400
Add the output of OSX arch into the package name in preparation for ppc
vs x86 packages.
r6790@totoro (orig r6789): phobos | 2006-07-18 00:40:02 -0400
Load and start tor in launchd at end of installation
r6791@totoro (orig r6790): arma | 2006-07-18 00:42:32 -0400
Define a schedule for how long to wait between retrying
application connections. Rather than waiting a fixed amount of
time between each retry, we wait only 5 seconds for the first,
10 seconds for the second, and 15 seconds for each retry after
that. Hopefully this will improve the expected experience.
Addresses bug 297.
r6792@totoro (orig r6791): arma | 2006-07-18 00:48:59 -0400
don't squeal if the first few retries fail.
r6793@totoro (orig r6792): nickm | 2006-07-20 12:47:35 -0400
Fork off v0 of the protocol spec; we are going to add versioning soon so we can make backward-incompatible changes without breaking the whole network. Also, fork the v0 directory protocol into its own document, and turn dir-spec.txt into the present tense.
r6794@totoro (orig r6793): nickm | 2006-07-20 12:48:02 -0400
Add a few more paragraphs to path-spec.txt
r6795@totoro (orig r6794): nickm | 2006-07-20 13:35:54 -0400
Document HELLO cells and proposed connection protocol versioning scheme. NOTE: This will not work as documented; see notes.
r6796@totoro (orig r6795): arma | 2006-07-20 19:33:11 -0400
tweak
r6797@totoro (orig r6796): arma | 2006-07-20 19:45:26 -0400
tweak
r6798@totoro (orig r6797): phobos | 2006-07-21 00:30:19 -0400
Grammar fixes for clarity.
r6799@totoro (orig r6798): arma | 2006-07-21 03:06:18 -0400
an entry guard that is "unlisted", as well as not known to be "down", is
not therefore "up".
r6800@totoro (orig r6799): arma | 2006-07-21 03:53:21 -0400
tweak
r6801@totoro (orig r6800): arma | 2006-07-21 03:55:35 -0400
bugfix: if you find yourself reachable, then don't ever make any client
requests (so you stop predicting circuits), then hup, then later your
IP changes, you won't think circuits are working so you won't try to
test reachability, so you won't publish.
r6803@totoro (orig r6802): nickm | 2006-07-21 10:53:23 -0400
Another MSVC6 fix. Grnk.
r6804@totoro (orig r6803): nickm | 2006-07-21 18:02:58 -0400
These asserts will either cause spurious crashes or help debug the pend->conn->s == -1 issue.
r6805@totoro (orig r6804): arma | 2006-07-22 01:29:31 -0400
more bulletproof reachability testing
r6806@totoro (orig r6805): arma | 2006-07-22 03:15:34 -0400
i lied, that won't work at all. maybe this will.
r6807@totoro (orig r6806): arma | 2006-07-22 03:19:11 -0400
think harder about my logic
r6809@totoro (orig r6808): arma | 2006-07-23 01:18:29 -0400
whitespace and docs
r6810@totoro (orig r6809): arma | 2006-07-23 01:19:31 -0400
more todo items
r6811@totoro (orig r6810): nickm | 2006-07-23 01:32:35 -0400
Add a mem_is_zero function (I think we will need this) and a STRUCT_OFFSET macro (we already need this).
r6812@totoro (orig r6811): nickm | 2006-07-23 01:33:10 -0400
Remove STRUCT_OFFSET from config.c
r6813@totoro (orig r6812): nickm | 2006-07-23 01:39:37 -0400
Add (void) lines for unused parameters in eventdns.c
r6814@totoro (orig r6813): nickm | 2006-07-23 01:40:24 -0400
Delete trailing whitespace in eventdns.c
r6815@totoro (orig r6814): arma | 2006-07-23 01:52:27 -0400
use tor_mem_is_zero() in more places.
r6816@totoro (orig r6815): arma | 2006-07-23 02:41:02 -0400
rewrite conn->address for GET commands as well as POST commands.
r6817@totoro (orig r6816): nickm | 2006-07-23 03:19:49 -0400
Whitespace fix
r6818@totoro (orig r6817): nickm | 2006-07-23 03:37:35 -0400
Don't tell anybody, but we're going OO here. This patch splits
circuit_t into origin_circuit_t and or_circuit_t. I fixed some
segaults; there may be more. We still need to move more rendezvous
stuff into subtypes.
This is a trial run for splitting up connection_t; if the approach is
insane, please say so soon so we can do something smarter.
Also, this discards the old HALF_OPEN code, which nobody seems to
want.
r6819@totoro (orig r6818): nickm | 2006-07-23 04:13:45 -0400
Fix another segfault in assert_circuit_ok.
r6820@totoro (orig r6819): arma | 2006-07-23 07:50:03 -0400
publish a new descriptor after we hup. this is important if our
config has changed such that we'll want to start advertising our
dirport now, etc.
r6821@totoro (orig r6820): phobos | 2006-07-23 08:52:06 -0400
These settings are required for those using launchd in OSX 10.4.
This is a crude but functional way to insert them for now.
r6822@totoro (orig r6821): weasel | 2006-07-23 23:24:25 -0400
Previously our defaults for DataDirectory, PidFile, RunAsDaemon, and
Log differed from upstreams. Now Tor behaves just like before (with
our own DataDirectory and all) only when run as the debian-tor user.
If invoked as any other user, Tor will behave just like the pristine
upstream version.
r6823@totoro (orig r6822): weasel | 2006-07-23 23:38:26 -0400
Tell users about the init script when they try to run Tor as root.
Should we also do this when they try to run their Tor as any other
(non root, non debian-tor) user? - add 11_tor_as_root_more_helpful
r6892@totoro (orig r6891): weasel | 2006-07-24 21:00:48 -0400
Remove .cvsignore files from trunk
r6894@totoro (orig r6893): nickm | 2006-07-24 21:13:04 -0400
Add an item to the TODO.
r6899@totoro (orig r6898): nickm | 2006-07-25 00:34:14 -0400
Remove code to special-case "-cvs" ending, since it has not actually mattered since 0.0.9. Perhaps we can special-case even more...
r6900@totoro (orig r6899): nickm | 2006-07-25 18:26:42 -0400
Add libor.a and libor-crypto.a to svn:ignore
r6901@totoro (orig r6900): nickm | 2006-07-25 18:30:50 -0400
Allow wide lines if they have svn id tags in them. (This matters for svk: those tags can be *big*.)
r6902@totoro (orig r6901): nickm | 2006-07-25 18:33:57 -0400
Apply checkSpace.pl to checkSpace.pl.
r6903@totoro (orig r6902): nickm | 2006-07-25 18:51:51 -0400
Clarify a TODO, and test tweaked commit-email.pl script.
r6904@totoro (orig r6903): nickm | 2006-07-26 15:05:34 -0400
r6902@Kushana: nickm | 2006-07-25 17:30:27 -0400
Move rend_query to origin_circuit_t where it belongs; save another 17 bytes per OR circuit.
r6905@totoro (orig r6904): nickm | 2006-07-26 15:05:41 -0400
r6903@Kushana: nickm | 2006-07-25 18:22:48 -0400
No circuit can be both an intro point and a rend point, so we can merge both the cookie and the pk digest into one "rend_token" field for or circuits. This saves another 20 bytes per or circuit.
r6906@totoro (orig r6905): nickm | 2006-07-26 15:07:23 -0400
r6907@Kushana: nickm | 2006-07-25 19:03:43 -0400
Realign circuit structs to avoid wasted space.
r6907@totoro (orig r6906): nickm | 2006-07-26 15:07:26 -0400
r6908@Kushana: nickm | 2006-07-26 12:38:52 -0400
Refactor connection_t into edge, or, dir, control, and base subtypes. This might save some RAM on busy exit servers, but really matters most in terms of correctness.
r6908@totoro (orig r6907): nickm | 2006-07-26 15:07:37 -0400
r6909@Kushana: nickm | 2006-07-26 13:05:58 -0400
Clean up wide lines from last patch.
r6909@totoro (orig r6908): nickm | 2006-07-26 15:29:30 -0400
r6918@Kushana: nickm | 2006-07-26 15:22:28 -0400
Fix compilation for eventdns dns.c with split structs.
r6910@totoro (orig r6909): nickm | 2006-07-26 15:39:47 -0400
r6920@Kushana: nickm | 2006-07-26 15:39:40 -0400
Mark some TODO items done
r6911@totoro (orig r6910): phobos | 2006-07-26 17:50:27 -0400
Set Soft & Hard resource limits to appease launchd.
r6912@totoro (orig r6911): phobos | 2006-07-26 19:52:59 -0400
OSX pre-install script to clean up Tor and force a fresh install, but
save the server keys if they exist.
r6913@totoro (orig r6912): phobos | 2006-07-26 20:19:36 -0400
Be better at finding the Tor install path, backup all of Tor just in
case, then blow Tor away
r6914@totoro (orig r6913): phobos | 2006-07-26 21:20:02 -0400
Backup only what is needed, write the file we created to a temp file for
TorPostFlight to restore the data and remove the temp files
r6915@totoro (orig r6914): phobos | 2006-07-26 23:02:47 -0400
The whole process works from preflight to postflight creating a clean
Tor install with proper config file edits
r6916@totoro (orig r6915): phobos | 2006-07-26 23:10:23 -0400
OSX gets confused when you have two ways to start the same program.
r6917@totoro (orig r6916): phobos | 2006-07-26 23:17:50 -0400
Oops, forgot the all important 'r'
r6919@totoro (orig r6918): nickm | 2006-07-27 00:10:51 -0400
Fix comments that implied that only dir connections had a purpose field, and the code that believed in those comments.
r6920@totoro (orig r6919): nickm | 2006-07-27 01:03:57 -0400
r6922@Kushana: nickm | 2006-07-26 16:32:24 -0400
Rename some fields, compress a bitfield, and document some structs and fields
r6923@totoro (orig r6922): nickm | 2006-07-27 13:16:10 -0400
Add more asserts in dns_found_answer. This may confirm my theory that dns_purge_resolve is the culprit.
r6924@totoro (orig r6923): phobos | 2006-07-27 13:19:32 -0400
Far better test and handling of existing torrc
r6925@totoro (orig r6924): nickm | 2006-07-27 13:37:37 -0400
Get better numbers out of HT_REP_OK
r6926@totoro (orig r6925): nickm | 2006-07-27 14:35:25 -0400
Fix a bug in HT_REMOVE.
r6927@totoro (orig r6926): nickm | 2006-07-27 14:35:56 -0400
More asserts in dns.c
r6928@totoro (orig r6927): arma | 2006-07-27 15:35:11 -0400
resolve typo
r6929@totoro (orig r6928): phobos | 2006-07-27 16:03:09 -0400
fi, fy fo fum, if-then- doesn't work without one
r6930@totoro (orig r6929): phobos | 2006-07-27 16:24:53 -0400
Forget leopard.
r6933@totoro (orig r6932): phobos | 2006-07-28 09:52:36 -0400
So long, farewell, auf Wiedersehen, adieu, launchd
r6934@totoro (orig r6933): nickm | 2006-07-28 11:11:11 -0400
r6948@Kushana: nickm | 2006-07-28 10:10:35 -0400
Identify some likely target fields for lowering; lower global_identifier (since we only use it for AP streams and origin circs).
r6935@totoro (orig r6934): nickm | 2006-07-28 11:11:20 -0400
r6949@Kushana: nickm | 2006-07-28 10:17:38 -0400
Shave another 8 bytes from connection_t: turn inbuf_reached_eof into a bit, and lower timestamp_lastempty to or_connection_t
r6936@totoro (orig r6935): nickm | 2006-07-28 11:11:28 -0400
r6950@Kushana: nickm | 2006-07-28 10:32:08 -0400
Document split fields better. Now, I think we can take a break from type splitting for a bit.
r6937@totoro (orig r6936): phobos | 2006-07-29 23:32:54 -0400
Remove the launchd plist file.
r6938@totoro (orig r6937): arma | 2006-07-29 23:34:44 -0400
a bit more debugging for phobos
r6939@totoro (orig r6938): phobos | 2006-07-29 23:53:18 -0400
Remove the last vestiges of launchd plist.
r6940@totoro (orig r6939): arma | 2006-07-30 00:32:58 -0400
defense in depth
r6945@totoro (orig r6944): arma | 2006-07-30 00:45:59 -0400
forward-port the website hack. note that with svn, our build
system seems to build in-place, so the website/ and img/ directories
actually get created in my sandbox. poo.
r6946@totoro (orig r6945): arma | 2006-07-30 00:54:13 -0400
fix assert found by DreadWingKnight: now that rendezvous streams
are attached to p_streams, the p_streams list can consist of both AP
and EXIT conns.
r6947@totoro (orig r6946): arma | 2006-07-30 01:36:17 -0400
forward-port the 0.1.1.23 changelog.
r6953@totoro (orig r6952): nickm | 2006-07-31 13:59:11 -0400
r6952@Kushana: nickm | 2006-07-28 11:09:37 -0400
Add completely untested find-my-nameservers code for win32.
r6954@totoro (orig r6953): nickm | 2006-07-31 13:59:37 -0400
r6958@Kushana: nickm | 2006-07-29 18:54:15 -0400
Looks like we might need a priority queue.
r6955@totoro (orig r6954): nickm | 2006-07-31 14:00:18 -0400
r6959@Kushana: nickm | 2006-07-29 22:33:18 -0400
start restructuring dns to use priority queues for expiring entries.
r6956@totoro (orig r6955): nickm | 2006-07-31 14:00:47 -0400
r6957@totoro (orig r6956): nickm | 2006-07-31 14:01:18 -0400
r6977@Kushana: nickm | 2006-07-31 13:01:28 -0400
Solve timing-out pending connections. Add pending resolves to expiry queue; when we find an answer, change the pending resolve to "done" and stick the actual answer in the expiry queue as a new entry. This uses a little more memory, but makes the code simpler than other solutions.
r6958@totoro (orig r6957): nickm | 2006-07-31 14:01:22 -0400
r6978@Kushana: nickm | 2006-07-31 13:16:14 -0400
Add isupper and islower wrappers to compat.h
r6959@totoro (orig r6958): nickm | 2006-07-31 14:01:27 -0400
r6979@Kushana: nickm | 2006-07-31 13:16:58 -0400
Add assert_ok functions for strmap and digestmap; use them in unit test code.
r6960@totoro (orig r6959): nickm | 2006-07-31 14:01:37 -0400
r6980@Kushana: nickm | 2006-07-31 13:18:22 -0400
Add a utility function to verify that a string has been through strlower.
r6961@totoro (orig r6960): nickm | 2006-07-31 14:01:45 -0400
r6981@Kushana: nickm | 2006-07-31 13:23:26 -0400
More asserts for cache correctness.
r6962@totoro (orig r6961): nickm | 2006-07-31 14:01:49 -0400
r6982@Kushana: nickm | 2006-07-31 13:47:19 -0400
documentation and naming tweaks in dns.c
r6963@totoro (orig r6962): nickm | 2006-07-31 16:19:58 -0400
r6993@Kushana: nickm | 2006-07-31 16:19:21 -0400
Interesting how much a ! can change the behavior of an assert.
r6964@totoro (orig r6963): arma | 2006-07-31 16:25:57 -0400
more compile options mean more codepaths
r6965@totoro (orig r6964): arma | 2006-08-01 00:08:15 -0400
man, our sample torrc sucked.
r6966@totoro (orig r6965): arma | 2006-08-02 01:17:22 -0400
explain that the exitlist isn't perfect, because some tor exit nodes
don't exit on their advertised address.
r6969@totoro (orig r6968): weasel | 2006-08-02 21:50:10 -0400
r8207@galaxy: weasel | 2006-08-03 03:22:17 +0200
Merge in local revisions 7944, 8205, and 8206: Forward port 0.1.1.x changelog to trunk
r6970@totoro (orig r6969): arma | 2006-08-03 00:22:25 -0400
fix a seg fault on exit for clients; and fix a comment.
r6971@totoro (orig r6970): arma | 2006-08-03 00:23:45 -0400
turn future seg faults into asserts
r6972@totoro (orig r6971): nickm | 2006-08-03 03:46:25 -0400
Patch from Frediano Ziglio: Windows compilation fixes on eventdns.c.
r6973@totoro (orig r6972): nickm | 2006-08-04 14:23:56 -0400
r6995@Kushana: nickm | 2006-07-31 13:30:42 -0700
Avoid segfault if we exit before we get our first dns answer.
r6974@totoro (orig r6973): nickm | 2006-08-04 14:24:13 -0400
r6975@totoro (orig r6974): nickm | 2006-08-04 14:24:25 -0400
r7007@Kushana: nickm | 2006-08-03 09:58:30 -0700
Export and use eventdns_config_windows_nameservers(); clean up some comments and log messages.
r6976@totoro (orig r6975): nickm | 2006-08-04 14:24:41 -0400
r7008@Kushana: nickm | 2006-08-03 10:03:39 -0700
Oops. We shouldnt initialize eventdns when we are not being a server and not resolving anything.
r6977@totoro (orig r6976): nickm | 2006-08-04 14:26:13 -0400
r7009@Kushana: nickm | 2006-08-03 10:44:58 -0700
Add functions to eventdns to allow detecting whether we have any nameservers configured, and to change the list of nameservers after initial configuration.
r6978@totoro (orig r6977): nickm | 2006-08-04 14:26:40 -0400
r7010@Kushana: nickm | 2006-08-03 10:47:36 -0700
Enable log message format checking in eventdns.c when __GNUC__ is defined.
r6979@totoro (orig r6978): nickm | 2006-08-04 14:27:10 -0400
r7011@Kushana: nickm | 2006-08-03 13:26:34 -0700
eventdns: Document functions added to API; make suspended requests go to the front of the queue; check (or explicitly ignore) return values on libevent functions.
r6980@totoro (orig r6979): nickm | 2006-08-04 14:31:13 -0400
r6981@totoro (orig r6980): nickm | 2006-08-04 14:32:43 -0400
r7012@Kushana: nickm | 2006-08-03 19:21:25 -0700
Add an "mmap handle" type to encapsulate bookkeeping elements of mmap issues; add prelim win32 impl
r6982@totoro (orig r6981): nickm | 2006-08-04 15:03:40 -0400
r7025@Kushana: nickm | 2006-08-04 12:03:22 -0700
Finish (I hope) windows mmap impl.
r6983@totoro (orig r6982): arma | 2006-08-04 16:30:45 -0400
make svn trunk link again. nick can fix this if it's wrong.
r6985@totoro (orig r6984): arma | 2006-08-04 23:08:56 -0400
a potential fix on the HELLO protocol design
r6986@totoro (orig r6985): nickm | 2006-08-05 13:52:51 -0400
r7027@Kushana: nickm | 2006-08-04 13:06:48 -0700
Oops. Fix downcast macro.
r6987@totoro (orig r6986): nickm | 2006-08-05 13:53:08 -0400
r7028@Kushana: nickm | 2006-08-04 13:10:16 -0700
Make data and size fields visible in tor_mmap_t; hide win magic differently.
r6988@totoro (orig r6987): nickm | 2006-08-05 13:53:21 -0400
r7029@Kushana: nickm | 2006-08-04 14:08:41 -0700
Remove now-spurious size and data arguments from tor_mmap_file
r6989@totoro (orig r6988): nickm | 2006-08-05 13:53:32 -0400
r7030@Kushana: nickm | 2006-08-04 14:46:52 -0700
Close an fd leak on failed mmap()
r6990@totoro (orig r6989): phobos | 2006-08-07 21:42:52 -0400
Update preflight to save Privoxy configs as well
r6991@totoro (orig r6990): arma | 2006-08-08 02:21:52 -0400
three more todo items
r6992@totoro (orig r6991): arma | 2006-08-08 18:56:26 -0400
change the dir-spec to say that it's version 2 of the dir spec,
and move the v0 file to v1.
r6996@totoro (orig r6995): nickm | 2006-08-08 20:58:27 -0400
Say more about reverse DNS
r6999@totoro (orig r6996): nickm | 2006-08-09 02:41:29 -0400
r7056@Kushana: nickm | 2006-08-08 23:40:53 -0700
Add a comment about v0 fallback approach. Why did we dislike discriminating on X.509 certs again?
r7000@totoro (orig r6997): nickm | 2006-08-09 02:54:02 -0400
r7058@Kushana: nickm | 2006-08-08 23:53:46 -0700
Clarify point about certs. We have been over this before, but it seems simpler than what we are considering now.
r7001@totoro (orig r6998): nickm | 2006-08-09 04:23:27 -0400
r7285@Kushana: nickm | 2006-08-09 01:23:11 -0700
Claim a TODO item (and test commit signing.)
r7002@totoro (orig r6999): nickm | 2006-08-09 04:30:11 -0400
r7286@Kushana: nickm | 2006-08-09 01:28:27 -0700
Claim another TODO item (and test commit signing.)
r7003@totoro (orig r7000): weasel | 2006-08-09 06:25:01 -0400
r8245@danube: weasel | 2006-08-09 12:24:00 +0200
Add a few comments so I find the rationale for the autoconf --build/--host split in the future when I need it again
r7006@totoro (orig r7001): nickm | 2006-08-09 17:42:38 -0400
r7005@totoro: nickm | 2006-08-09 17:42:18 -0400
Begin committing violence against the spec; add some TODO items at the top. Arma, if you disagree, better say so.
r7228@totoro (orig r7002): phobos | 2006-08-09 22:06:35 -0400
Update for universal binaries for OSX.
r7229@totoro (orig r7003): nickm | 2006-08-10 03:39:47 -0400
Recommend libevent 1.1b for kqueue and win32 methods; deprecate libevent 1.0b harder; make libevent recommendation system saner.
r7230@totoro (orig r7004): nickm | 2006-08-10 04:00:13 -0400
Experimentally re-enable kqueue on OSX when using libevent 1.1b or later. Log when we are doing this, so we can diagnose it when it fails.
r7231@totoro (orig r7005): arma | 2006-08-10 04:00:54 -0400
a way to make tor more stable in crummy situations
r7232@totoro (orig r7006): arma | 2006-08-10 04:13:41 -0400
initial skeleton for issues to resolve re: blocking resistance.
r7233@totoro (orig r7007): nickm | 2006-08-10 05:01:37 -0400
r7299@Kushana: nickm | 2006-08-10 01:08:58 -0700
Patch from Tup to add support for transparent AP connections: this basically bundles the functionality of trans-proxy-tor into the tor mainline. Now hosts with compliant pf/netfilter implementations can redirect TCP connections straight to Tor without diverting through SOCKS.
r7234@totoro (orig r7008): nickm | 2006-08-10 05:01:46 -0400
r7300@Kushana: nickm | 2006-08-10 01:36:40 -0700
Distinguish netfilter vs pf at configure time based on headers, not on OS.
r7235@totoro (orig r7009): nickm | 2006-08-10 05:01:54 -0400
r7301@Kushana: nickm | 2006-08-10 01:41:27 -0700
Only open /dev/pf once.
r7236@totoro (orig r7010): nickm | 2006-08-10 05:02:02 -0400
r7302@Kushana: nickm | 2006-08-10 01:48:44 -0700
Warn about open TransListenAddress values.
r7237@totoro (orig r7011): nickm | 2006-08-10 05:02:12 -0400
r7303@Kushana: nickm | 2006-08-10 01:52:19 -0700
whitespace fixes
r7238@totoro (orig r7012): nickm | 2006-08-10 05:02:26 -0400
r7304@Kushana: nickm | 2006-08-10 01:58:05 -0700
Fix verbose compilation errors; make sure transparent proxy fails when no method is configured.
r7240@totoro (orig r7013): nickm | 2006-08-10 05:14:57 -0400
r7313@Kushana: nickm | 2006-08-10 02:13:35 -0700
Add missing .h to header file name. My bad.
r7243@totoro (orig r7015): nickm | 2006-08-10 05:30:25 -0400
r7315@Kushana: nickm | 2006-08-10 02:30:13 -0700
add missing "test" and missing include.
r7248@totoro (orig r7018): nickm | 2006-08-10 15:56:10 -0400
Fix crash in first-time option validation. Oops.
r7249@totoro (orig r7019): nickm | 2006-08-11 03:09:09 -0400
r7323@Kushana: nickm | 2006-08-10 22:04:57 -0700
Add sys/socket.h prereq for net/if.h and net/pfvar.h
r7250@totoro (orig r7020): nickm | 2006-08-11 03:09:17 -0400
r7324@Kushana: nickm | 2006-08-10 23:23:15 -0700
Add more warnings to the list of those we tolerate. Start using GCC attributes more, for better error checking and better code generation.
r7251@totoro (orig r7021): nickm | 2006-08-11 03:09:28 -0400
r7325@Kushana: nickm | 2006-08-10 23:37:31 -0700
Use gcc offsetof where available.
r7252@totoro (orig r7022): nickm | 2006-08-11 03:09:35 -0400
r7326@Kushana: nickm | 2006-08-10 23:50:49 -0700
And another GCC change: predict that tor_frees() are usually real frees, and tor_asserts() usually wont happen. Other test should wait till -fprofile-arcs
r7253@totoro (orig r7023): nickm | 2006-08-11 03:09:45 -0400
r7327@Kushana: nickm | 2006-08-11 00:00:36 -0700
Fix a pedantic warning
r7254@totoro (orig r7024): nickm | 2006-08-11 03:09:52 -0400
r7328@Kushana: nickm | 2006-08-11 00:04:26 -0700
remove an assert that can never be false (an array in a non-null struct cannot be null).
r7255@totoro (orig r7025): nickm | 2006-08-11 03:17:16 -0400
r7335@Kushana: nickm | 2006-08-11 00:13:03 -0700
fix wide lines
r7256@totoro (orig r7026): arma | 2006-08-11 03:31:16 -0400
fix typo, add explanatory comment
r7257@totoro (orig r7027): arma | 2006-08-11 03:41:21 -0400
fix funny-looking assignment that crashes unit tests
r7258@totoro (orig r7028): nickm | 2006-08-11 03:42:11 -0400
r7337@Kushana: nickm | 2006-08-11 00:42:04 -0700
Only use __builtin_offsetof with gcc 4 or later
r7259@totoro (orig r7029): nickm | 2006-08-11 03:51:34 -0400
r7341@Kushana: nickm | 2006-08-11 00:51:05 -0700
Amazing how much difference adding a ! to all your asserts can make.
r7260@totoro (orig r7030): nickm | 2006-08-11 03:51:42 -0400
r7342@Kushana: nickm | 2006-08-11 00:51:25 -0700
Remove braindeadism.
r7262@totoro (orig r7032): arma | 2006-08-11 20:30:07 -0400
more todo items
r7263@totoro (orig r7033): arma | 2006-08-11 20:38:38 -0400
mark out the two biggest bugs
r7264@totoro (orig r7034): arma | 2006-08-12 03:44:13 -0400
two more thoughts to consider for blocking resistance
r7266@totoro (orig r7036): weasel | 2006-08-12 19:26:54 -0400
r8290@danube: weasel | 2006-08-13 01:26:01 +0200
Merge local r8289 into trunk: Update debian/copyright
r7267@totoro (orig r7037): phobos | 2006-08-12 20:13:55 -0400
Updated details on how to check for universal binary.
r7269@totoro (orig r7039): phobos | 2006-08-12 20:29:56 -0400
Apparently not everyone wants to build Universal binaries. Return valid
Archictecture detection for inclusion into final dmg naming.
r7272@totoro (orig r7042): arma | 2006-08-13 19:38:30 -0400
simplify the connection_write_to_buf_zlib() wrapper.
r7273@totoro (orig r7043): arma | 2006-08-13 20:28:44 -0400
note three more bugs we should resolve
r7274@totoro (orig r7044): arma | 2006-08-14 01:53:57 -0400
save 3-7 bytes per edge connection
r7275@totoro (orig r7045): nickm | 2006-08-14 02:03:26 -0400
r7369@Kushana: nickm | 2006-08-14 02:03:10 -0400
Shortening fields is only one part of making structs shorter. You must also consider alignment padding. Whee.
r7276@totoro (orig r7046): arma | 2006-08-14 02:27:39 -0400
two more todo's, plus a cleanup
r7277@totoro (orig r7047): arma | 2006-08-14 02:29:40 -0400
this is what i really meant
r7278@totoro (orig r7048): arma | 2006-08-14 03:08:29 -0400
simplify compare_cached_resolves_by_expiry() to make it match
the idioms of other compare functions. hopefully i didn't
break it?
r7279@totoro (orig r7049): arma | 2006-08-14 04:55:41 -0400
fix typos and stuff
r7280@totoro (orig r7050): arma | 2006-08-14 05:03:16 -0400
clarify that some old versions used an organizationName of "Tor",
and others used "TOR". We should avoid both now.
r7281@totoro (orig r7051): arma | 2006-08-14 05:04:27 -0400
rename HELLO cells to VERSIONS cells.
r7282@totoro (orig r7052): arma | 2006-08-14 05:44:54 -0400
note a compile warning that we should investigate one day.
r7283@totoro (orig r7053): arma | 2006-08-14 06:00:15 -0400
avoid complaining about our SOCKS proxy proxy.
r7284@totoro (orig r7054): arma | 2006-08-14 06:16:42 -0400
checkpoint in-progress changelog notes
r7287@totoro (orig r7057): nickm | 2006-08-14 16:16:21 -0400
r7383@Kushana: nickm | 2006-08-14 16:16:03 -0400
Stop walking entire dns cache for every request, now that we found our hash table bug.
r7288@totoro (orig r7058): nickm | 2006-08-14 17:44:29 -0400
r7045@Kushana: nickm | 2006-08-05 13:56:44 -0400
mess with the TODO a little.
r7289@totoro (orig r7059): nickm | 2006-08-14 17:44:34 -0400
r7046@Kushana: nickm | 2006-08-05 13:57:04 -0400
Make it possible for dns_init() to fail; note failure of eventdns configuratoin.
r7290@totoro (orig r7060): nickm | 2006-08-14 17:44:39 -0400
r7291@totoro (orig r7061): nickm | 2006-08-14 17:44:45 -0400
r7386@Kushana: nickm | 2006-08-14 17:43:44 -0400
Patch from Adam Langley.
* I meant getaddrinfo_a, not getaddrinfo_r - fixed
* Added more checks to the parsing code.
* It seems you switched an alloca to a malloc, but didn't add any frees
r7292@totoro (orig r7062): nickm | 2006-08-14 23:54:09 -0400
r7392@Kushana: nickm | 2006-08-14 23:50:32 -0400
Only do the expensive version of router_have_minimum_dir_info() when the dir info has changed. Backport candidate, since oprofile suggests that this function and ones it calls account for 25-35% of oprofile samples.
r7293@totoro (orig r7063): nickm | 2006-08-14 23:54:13 -0400
r7393@Kushana: nickm | 2006-08-14 23:51:07 -0400
remove extraneous #endif
r7305@totoro (orig r7064): nickm | 2006-08-15 00:50:17 -0400
r7397@Kushana: nickm | 2006-08-15 00:46:18 -0400
Fix eventdns version of dns.c. Man, we need to get rid of this eventdns/dnsworker split.
r7306@totoro (orig r7065): nickm | 2006-08-15 00:50:33 -0400
r7398@Kushana: nickm | 2006-08-15 00:49:50 -0400
Add missing backslash
r7308@totoro (orig r7067): arma | 2006-08-15 22:18:55 -0400
dns.c:173: warning: control reaches end of non-void function
r7309@totoro (orig r7068): arma | 2006-08-15 23:44:13 -0400
automatically avoid picking more than one node from the same
/16 network when constructing a circuit.
r7310@totoro (orig r7069): nickm | 2006-08-16 14:47:19 -0400
r7404@Kushana: nickm | 2006-08-16 09:32:19 -0400
Pass hints to getaddrinfo; fix bug 280 (?)
r7311@totoro (orig r7070): nickm | 2006-08-16 14:47:24 -0400
r7405@Kushana: nickm | 2006-08-16 14:38:46 -0400
Implement reverse DNS lookup in eventdns: add new entry point; rename old entry point; revise TODO a little; add facility for parsing dns names.
r7313@totoro (orig r7072): nickm | 2006-08-17 19:00:32 -0400
r7411@Kushana: nickm | 2006-08-17 19:00:25 -0400
patch suggested by Karsten Loesing: respond to SIGNAL command before we execute the signal, in case the signal shuts us down.
r7314@totoro (orig r7073): arma | 2006-08-18 13:46:14 -0400
clean up AllowInvalidNodes man page entry.
r7315@totoro (orig r7074): arma | 2006-08-18 14:19:35 -0400
fix typo pointed out by paul
r8455@totoro (orig r8214): nickm | 2006-08-22 02:10:53 -0400
r8556@Kushana: nickm | 2006-08-22 01:22:46 -0400
Enable eventdns by default on platforms where we autoconf. This should be everything but windows.
r8456@totoro (orig r8215): nickm | 2006-08-22 02:10:58 -0400
r8557@Kushana: nickm | 2006-08-22 02:10:12 -0400
dns-related TODO changes.
r8462@totoro (orig r8221): arma | 2006-08-23 19:45:03 -0400
tor --verify-config now exits with -1(255) or 0 depending on whether
the config options are bad or good.
r8463@totoro (orig r8222): arma | 2006-08-23 20:54:18 -0400
clean up logging conventions in do_list_fingerprint()
r8464@totoro (orig r8223): arma | 2006-08-24 00:51:55 -0400
make our socks5 handling more robust to broken socks clients:
throw out everything waiting on the buffer in between socks
handshake phases, since they can't possibly (so the theory
goes) have predicted what we plan to respond to them.
r8465@totoro (orig r8224): nickm | 2006-08-25 17:01:56 -0400
r8572@Kushana: nickm | 2006-08-25 16:35:49 -0400
Fix for bug 308: When we have a state file we cannot parse, tell the user, and move it aside.
r8466@totoro (orig r8225): nickm | 2006-08-25 17:02:01 -0400
r8573@Kushana: nickm | 2006-08-25 16:55:19 -0400
Resolve bug 321 when using dnsworkers: append a period to every address we resolve at the exit node, so that we do not accidentally pick up local addresses, and so that failing searches are retried in the resolver search domains. (This is already solved for eventdns.)
r8467@totoro (orig r8226): nickm | 2006-08-25 17:16:22 -0400
r8576@Kushana: nickm | 2006-08-25 17:16:01 -0400
Fix bug 314: Instead of checking address_in_virtualrange, check addressmap_have_mapping(). This should be more accurate. [Rename to addressmap_have_mapping() from addressmap_already_mapped().]
r8468@totoro (orig r8227): arma | 2006-08-26 00:48:50 -0400
fix a log level -- err is for things that kill tor, warn is for
things that tor can recover from.
also, avoid situations where people who don't read their logs
accumulate ten thousand useless files in their datadir.
r8469@totoro (orig r8228): arma | 2006-08-26 00:52:22 -0400
more todo items when we avoid putting /16 servers into the same circuit
r8470@totoro (orig r8229): arma | 2006-08-26 02:51:02 -0400
remove some more vestiges of cvs
r8471@totoro (orig r8230): arma | 2006-08-26 02:56:16 -0400
complete an item, abandon an item, defer an item.
r8472@totoro (orig r8231): arma | 2006-08-26 02:57:48 -0400
remove more completed items
r8473@totoro (orig r8232): arma | 2006-08-26 03:13:54 -0400
make a 'real soon now' section of the todo, to point out
what's more urgent
r8474@totoro (orig r8233): arma | 2006-08-26 21:33:35 -0400
fix typo
r8475@totoro (orig r8234): arma | 2006-08-26 21:41:08 -0400
i'm guessing nick meant to check the return value of dns_init.
and revise some log severities to match the convention.
r8476@totoro (orig r8235): arma | 2006-08-26 22:07:54 -0400
stop three memory leaks. nick, fix these if i'm wrong.
r8477@totoro (orig r8236): arma | 2006-08-26 22:12:12 -0400
stop a big memory leak: we were leaking the whole contents
of cached-routers.new every time we read it.
r8478@totoro (orig r8237): arma | 2006-08-27 02:49:33 -0400
typo and whitespace
r8479@totoro (orig r8238): arma | 2006-08-27 02:49:51 -0400
a usability improvement i just thought of
r8480@totoro (orig r8239): arma | 2006-08-27 02:51:19 -0400
0.1.2.1-alpha will break blossom exit node functionality; put that
on the todo list.
r8481@totoro (orig r8240): arma | 2006-08-27 02:55:48 -0400
clean up the 0.1.2.1-alpha changelog
r8482@totoro (orig r8241): arma | 2006-08-27 02:58:30 -0400
comment out an unused variable; nick, feel free to excise it further.
r8483@totoro (orig r8242): arma | 2006-08-27 03:02:41 -0400
bump the doxyfile version number, since i made a doxygen
ref man last week.
r8484@totoro (orig r8243): arma | 2006-08-27 03:03:05 -0400
fix typo
r8485@totoro (orig r8244): arma | 2006-08-27 03:03:17 -0400
bump to 0.1.2.1-alpha
(does it work?)
r8486@totoro (orig r8245): arma | 2006-08-27 03:26:00 -0400
turn eventdns off by default until we can get some more useful
log messages into it. no use learning that it's broken without
being able to learn what is broken too.
r8488@totoro (orig r8247): phobos | 2006-08-27 20:40:11 -0400
Update rpms to require libevent 1.1b.
r8490@totoro (orig r8249): phobos | 2006-08-27 21:06:44 -0400
Remove architecture from builds. The official builds are universal
binaries.
r8491@totoro (orig r8250): phobos | 2006-08-27 22:20:35 -0400
Replace nickm as packager of rpms.
r8492@totoro (orig r8251): arma | 2006-08-27 23:01:07 -0400
bump trunk to 0.1.2.1-alpha-dev
r8493@totoro (orig r8252): nickm | 2006-08-27 23:15:38 -0400
r8605@Kushana: nickm | 2006-08-27 14:01:11 -0400
divide eventdns log messages into warn and info messages.
r8494@totoro (orig r8253): nickm | 2006-08-27 23:15:47 -0400
r8606@Kushana: nickm | 2006-08-27 14:04:19 -0400
Now that 0.1.2.1-alpha is out, make eventdns on-by-default again. (Hoop-lah.)
r8495@totoro (orig r8254): nickm | 2006-08-27 23:15:50 -0400
r8607@Kushana: nickm | 2006-08-27 15:45:42 -0400
Change configuration strategy for eventdns. Instead of elaborate option set, just allow the user to specify another resolv.conf to use.
r8496@totoro (orig r8255): nickm | 2006-08-27 23:15:55 -0400
r8608@Kushana: nickm | 2006-08-27 16:57:47 -0400
Make it possible to change nameserver options while Tor is running.
r8497@totoro (orig r8256): nickm | 2006-08-27 23:16:02 -0400
r8609@Kushana: nickm | 2006-08-27 17:24:27 -0400
Add some doxygen, concentrated in dns.c
r8499@totoro (orig r8258): nickm | 2006-08-27 23:29:51 -0400
Kill some SVK branches that I am done with.
r8500@totoro (orig r8259): nickm | 2006-08-27 23:46:21 -0400
r8626@Kushana: nickm | 2006-08-27 23:45:46 -0400
Aw, crap. Non-gcc bug. We need regular windows builds.
r8511@totoro (orig r8270): arma | 2006-08-28 04:24:36 -0400
put some symbols on the eventdns logging. leave the haphazard
formatting alone.
r8512@totoro (orig r8271): arma | 2006-08-28 14:51:36 -0400
fix bug found by Adam J. Richter: directory clients shouldn't
hang up five minutes after they begin. but directory servers
should still hang up after 5 minutes of failing to deliver
any bytes.
r8513@totoro (orig r8272): arma | 2006-08-28 15:00:17 -0400
an even better check -- now servers hang up if they can't write
for a while, and clients hang up if they can't read for a while.
r8514@totoro (orig r8273): arma | 2006-08-28 15:02:57 -0400
and make the comment better, while i'm at it
r8518@totoro (orig r8277): nickm | 2006-08-28 16:42:14 -0400
r8652@Kushana: nickm | 2006-08-28 16:41:44 -0400
Resolve bug 324: strdup the right variable.
r8519@totoro (orig r8278): nickm | 2006-08-28 16:50:47 -0400
r8654@Kushana: nickm | 2006-08-28 16:50:36 -0400
I bet real programmers even have programs that compile.
r8521@totoro (orig r8280): nickm | 2006-08-29 00:22:51 -0400
r8657@Kushana: nickm | 2006-08-29 00:22:31 -0400
Mark an item complete; test new commit-email.pl script.
r8539@totoro (orig r8298): weasel | 2006-08-29 16:43:52 -0400
r9565@galaxy: weasel | 2006-08-29 22:20:38 +0200
Do not reload or start when our config is known to be broken
r8540@totoro (orig r8299): weasel | 2006-08-29 16:43:59 -0400
r9566@galaxy: weasel | 2006-08-29 22:32:26 +0200
Unnest all those nested if blocks
r8541@totoro (orig r8300): weasel | 2006-08-29 16:44:05 -0400
r9567@galaxy: weasel | 2006-08-29 22:32:46 +0200
Prepare 0.1.2.1-alpha-1
r8542@totoro (orig r8301): weasel | 2006-08-29 16:44:11 -0400
r9568@galaxy: weasel | 2006-08-29 22:34:36 +0200
Upload target is experimental, not unstable
r8543@totoro (orig r8302): weasel | 2006-08-29 16:44:17 -0400
r9569@galaxy: weasel | 2006-08-29 22:38:54 +0200
Change Standards-Version to 3.7.2. No changes required.
r8544@totoro (orig r8303): nickm | 2006-08-29 17:59:20 -0400
r8682@Kushana: nickm | 2006-08-29 17:58:59 -0400
Fix compilation on GCC2 by disabling fun attributes unless __GNUC_MAJOR__ >= 3.
r8548@totoro (orig r8307): nickm | 2006-08-30 19:34:49 -0400
r8686@Kushana: nickm | 2006-08-30 18:01:35 -0400
Remove bogus whitespace
r8549@totoro (orig r8308): nickm | 2006-08-30 19:34:56 -0400
r8687@Kushana: nickm | 2006-08-30 19:33:28 -0400
Apply patch from Adam Langley: fix assert() in eventdns.c. [Fuzzing, apparently, is cool.]
r8550@totoro (orig r8309): nickm | 2006-08-31 13:39:47 -0400
r8691@Kushana: nickm | 2006-08-31 13:30:46 -0400
Fix bug 327 (part 1): Use correct macro to test for GCC 3 or later.
r8551@totoro (orig r8310): nickm | 2006-08-31 13:39:51 -0400
r8692@Kushana: nickm | 2006-08-31 13:38:07 -0400
Fix bug 327 (part 2): Cast char to unsigned char before passing to toupper/tolower. (Follow the same idiom as with isupper and friends, in case we run into the same problem on SGI or whereever it was.)
r8552@totoro (orig r8311): nickm | 2006-08-31 14:46:46 -0400
r8695@Kushana: nickm | 2006-08-31 14:35:36 -0400
Fix two corner cases in router_dir_info_changed(). This should have no observable effect.
r8553@totoro (orig r8312): nickm | 2006-08-31 14:47:54 -0400
r8696@Kushana: nickm | 2006-08-31 14:43:44 -0400
Try to appease some warnings with newer gccs that believe that ignoring a return value is okay, but casting a return value and then ignoring it is a sign of madness.
r8556@totoro (orig r8315): phobos | 2006-08-31 19:52:41 -0400
Remove old TODO that I can't remember the details of it.
r8557@totoro (orig r8316): nickm | 2006-09-02 01:33:11 -0400
r8704@Kushana: nickm | 2006-09-02 01:32:34 -0400
correct includes for net/if.h and net/pfvar.h on openbsd (from Tup)
r8558@totoro (orig r8317): arma | 2006-09-02 19:26:42 -0400
Tor is more than code these days.
r8559@totoro (orig r8318): nickm | 2006-09-02 22:13:52 -0400
r8708@Kushana: nickm | 2006-09-02 20:34:15 -0400
Fix warnings reported by weasel when compiling Tor on Debian woody.
r8560@totoro (orig r8319): nickm | 2006-09-02 22:13:56 -0400
r8709@Kushana: nickm | 2006-09-02 20:59:54 -0400
Remove unused autoconf checks.
r8561@totoro (orig r8320): weasel | 2006-09-03 16:19:35 -0400
r9648@danube: weasel | 2006-09-03 19:54:42 +0200
Document minor update of debian/copyright by arma in debian changelog
r8562@totoro (orig r8321): arma | 2006-09-05 10:30:06 -0400
patch from tup
r8563@totoro (orig r8322): chiussi | 2006-09-05 21:49:55 -0400
- made configure check if we are building for win32
- made configure link to required system dll's if building for win32
- added diffs for libevent 1.1b
- forced user to turn off eventdns if win32 is set
- cleaned up tor_mmap_file()_win32 (not sure if it's stable)
- cleaned up some warnings and typos
r8564@totoro (orig r8323): chiussi | 2006-09-05 21:58:55 -0400
- removed windows line endings
r8565@totoro (orig r8324): weasel | 2006-09-05 22:08:52 -0400
r9665@danube: weasel | 2006-09-06 04:08:12 +0200
Remove svn:executable properties from all files in Win32Build
r8566@totoro (orig r8325): nickm | 2006-09-06 04:42:12 -0400
r8723@Kushana: nickm | 2006-09-06 04:24:54 -0400
Clean up configure.in spaces, and make it work on Mac OS X again (for me).
r8567@totoro (orig r8326): nickm | 2006-09-06 04:42:16 -0400
r8724@Kushana: nickm | 2006-09-06 04:32:28 -0400
Fix spaces; restore support for mapping files over 4GB on win32 (?)
r8568@totoro (orig r8327): nickm | 2006-09-06 04:42:20 -0400
r8725@Kushana: nickm | 2006-09-06 04:39:29 -0400
spawn_func fixes: have cpuworker_main and dnsworker_main confirm to the right interfaces [casting func to void* is icky]. Also, make pthread_create() build without warnings.
r8569@totoro (orig r8328): nickm | 2006-09-06 16:22:05 -0400
r8729@Kushana: nickm | 2006-09-06 16:20:40 -0400
Fix dns_cancel_pending_resolve() to realize pending resolves have expiry times, and should not be freed except when they fall off the pqueue.
r8570@totoro (orig r8329): nickm | 2006-09-06 16:33:28 -0400
r8731@Kushana: nickm | 2006-09-06 16:33:19 -0400
Try to fix eventdns.c build on windows.
r8571@totoro (orig r8330): nickm | 2006-09-06 17:35:30 -0400
r8733@Kushana: nickm | 2006-09-06 17:35:22 -0400
Add some missing defines to eventdns.c for windows.
r8572@totoro (orig r8331): arma | 2006-09-06 20:30:29 -0400
new config option AvoidDiskWrites for people running tors on
usb keys and other media that degrades when you write.
not implemented yet, so just a reminder.
r8573@totoro (orig r8332): chiussi | 2006-09-06 20:51:20 -0400
- fixed up typos in eventdns.c
- configure lets user build with eventdns on win32
r8574@totoro (orig r8333): chiussi | 2006-09-06 20:54:28 -0400
rm'ed a tab (sorry)
r8575@totoro (orig r8334): arma | 2006-09-06 21:00:37 -0400
prefer calling it a client rather than an OP
r8576@totoro (orig r8335): arma | 2006-09-06 21:01:11 -0400
flesh out TODO re: AvoidDiskWrites
r8577@totoro (orig r8336): arma | 2006-09-06 21:02:23 -0400
clean up and correct the spec
r8578@totoro (orig r8337): arma | 2006-09-06 21:22:55 -0400
another controller event we left out
r8579@totoro (orig r8338): arma | 2006-09-06 21:23:15 -0400
clean up some comments
r8580@totoro (orig r8339): arma | 2006-09-06 23:26:17 -0400
help the rpm spec file recognize development versions now that
they're not called foo-cvs
r8581@totoro (orig r8340): arma | 2006-09-06 23:40:23 -0400
packages now start including svn website docs (oops)
r8583@totoro (orig r8342): chiussi | 2006-09-06 23:53:14 -0400
- added some debugging lines to load_nameservers_from_registry()
r8584@totoro (orig r8343): chiussi | 2006-09-06 23:58:46 -0400
- got rid of tabs (i think)
r8585@totoro (orig r8344): chiussi | 2006-09-07 00:02:52 -0400
- got rid of tabs
r8586@totoro (orig r8345): chiussi | 2006-09-07 02:34:20 -0400
fixed win32 eventdns snafu
r8587@totoro (orig r8346): chiussi | 2006-09-07 02:36:22 -0400
undid whitespace changes to dns.c
r8588@totoro (orig r8347): nickm | 2006-09-07 12:24:19 -0400
r8753@Kushana: nickm | 2006-09-07 12:23:25 -0400
Make eventdns spacing consistant
r8589@totoro (orig r8348): nickm | 2006-09-07 12:32:06 -0400
r8755@Kushana: nickm | 2006-09-07 12:31:57 -0400
Tabify eventdns
r8590@totoro (orig r8349): nickm | 2006-09-07 15:00:51 -0400
r8757@Kushana: nickm | 2006-09-07 13:07:46 -0400
Fix more compile warnings on Woody.
r8591@totoro (orig r8350): chiussi | 2006-09-08 05:02:14 -0400
fixed typo in eventdns.c
r8592@totoro (orig r8351): chiussi | 2006-09-08 05:05:07 -0400
another typo
r8593@totoro (orig r8352): weasel | 2006-09-08 12:12:15 -0400
r9695@danube: weasel | 2006-09-08 18:11:53 +0200
Make tor build with -O0
r8594@totoro (orig r8353): arma | 2006-09-08 16:48:43 -0400
possibly make tor build and run on cygwin again.
r8595@totoro (orig r8354): arma | 2006-09-08 23:18:39 -0400
add a "getinfo address" controller command.
r8596@totoro (orig r8355): arma | 2006-09-08 23:38:03 -0400
fix encoding in "getinfo addr-mappings" response.
fix error code when "getinfo dir/status/" fails.
r8597@totoro (orig r8356): nickm | 2006-09-08 23:46:52 -0400
r8766@Kushana: nickm | 2006-09-08 23:46:12 -0400
Remove some completed items from TODO; defer a bunch of stuff from 0.1.2.x (which we want to have an RC for by the end of October)
r8598@totoro (orig r8357): arma | 2006-09-09 15:16:07 -0400
fix typo, add log message
r8599@totoro (orig r8358): arma | 2006-09-09 15:20:27 -0400
parameterize the loudness of get_interface_address()
r8600@totoro (orig r8359): arma | 2006-09-09 15:36:51 -0400
start remembering X-Your-Address-Is hints even if you're
a client, so you can become a server more smoothly.
r8602@totoro (orig r8361): phobos | 2006-09-10 23:46:00 -0400
First attempt to document the process to build Win32 executables with
mingw. Relies upon magic to actually work right now.
r8611@totoro (orig r8370): nickm | 2006-09-11 22:50:14 -0400
r8776@Kushana: nickm | 2006-09-11 22:49:53 -0400
Avoid crash when telling controller stream-status and a stream is detached. Fixes bug 334. Backport candidate.
r8612@totoro (orig r8371): phobos | 2006-09-11 22:58:33 -0400
Updates to win32-mingw. libevent1.1b builds. openssl doesn't.
r8614@totoro (orig r8373): nickm | 2006-09-12 14:05:54 -0400
r8785@Kushana: nickm | 2006-09-12 14:05:46 -0400
Add non-dist utility for weasel
r8615@totoro (orig r8374): weasel | 2006-09-12 14:11:19 -0400
r9734@danube: weasel | 2006-09-12 20:11:02 +0200
Set svn:keywords
r8616@totoro (orig r8375): arma | 2006-09-12 14:31:03 -0400
avoid getting mail from ilja in 2008 about this.
r8617@totoro (orig r8376): nickm | 2006-09-12 15:00:55 -0400
r8791@Kushana: nickm | 2006-09-12 15:00:48 -0400
As long as we are being pedantic, we may as well be extra-pedantic.
r8624@totoro (orig r8383): nickm | 2006-09-13 13:34:59 -0400
r8795@Kushana: nickm | 2006-09-13 13:34:54 -0400
Try to split off the mingw portions of the libevent diff so we can send them to niels separately. Adding two patches: one that applies against 1.1b; one that applies against the libevent svn trunk.
r8625@totoro (orig r8384): arma | 2006-09-13 18:24:05 -0400
two todo items we'll need for the blocking-resistance scheme
r8626@totoro (orig r8385): arma | 2006-09-13 18:24:43 -0400
continue fleshing out the blocking-resistance design doc
r8627@totoro (orig r8386): weasel | 2006-09-14 00:53:23 -0400
r9736@danube: weasel | 2006-09-14 05:53:06 +0200
Refactor dirserv_parse_fingerprint_file(fname) into dirserv_load_fingerprint_file():
There is not need to put together the path to the approved-routers file in more than one place.
r8628@totoro (orig r8387): weasel | 2006-09-14 00:53:42 -0400
r9749@danube: weasel | 2006-09-14 06:53:12 +0200
Do not graciously increase the size to be mmaped if the current size already is
at a page_size boundary. This is important since if a file has a size of zero
and we mmap() it with length > 0, then accessing the mmaped memory area causes
a bus error. However, if we pass a length of 0 to mmap() it will return with -1
and things work from there.
r8629@totoro (orig r8388): nickm | 2006-09-14 01:00:02 -0400
r8808@senior-two-eighty: nickm | 2006-09-14 00:59:54 -0400
House style for no-args functions is old-style C, not C++
r8630@totoro (orig r8389): weasel | 2006-09-14 01:07:26 -0400
r9752@danube: weasel | 2006-09-14 07:06:49 +0200
Fix some dead code that is on occasion useful
r8631@totoro (orig r8390): weasel | 2006-09-14 01:07:34 -0400
r9753@danube: weasel | 2006-09-14 07:07:02 +0200
Remove duplicates from MyFamily
r8632@totoro (orig r8391): weasel | 2006-09-14 01:17:02 -0400
r9758@danube: weasel | 2006-09-14 07:10:12 +0200
Add a comment
r8633@totoro (orig r8392): weasel | 2006-09-14 01:17:12 -0400
r9759@danube: weasel | 2006-09-14 07:14:37 +0200
Add smartlist_uniq() to TODO list
r8634@totoro (orig r8393): arma | 2006-09-14 01:49:03 -0400
we should add a preamble to tor-design saying it's out of date.
r8636@totoro (orig r8395): nickm | 2006-09-14 18:34:57 -0400
r8817@Kushana: nickm | 2006-09-14 18:31:29 -0400
Patch from steve hildrey: Generate network status correctly on non-versioning dirservers
r8637@totoro (orig r8396): nickm | 2006-09-15 00:27:58 -0400
r8819@Kushana: nickm | 2006-09-15 00:27:45 -0400
Implement a smartlist_uniq() that will with luck not end the world.
r8638@totoro (orig r8397): weasel | 2006-09-15 00:29:36 -0400
r9767@danube: weasel | 2006-09-15 06:27:48 +0200
Minor documentation fix
r8639@totoro (orig r8398): weasel | 2006-09-15 01:20:16 -0400
r9770@danube: weasel | 2006-09-15 07:20:05 +0200
router_set_networkstatus() gets a list of status documents we asked for from
connection_dir_client_reached_eof(). However, as a cache we (sometimes?) just
ask for "all". router_set_networkstatus() would freak out over that, meaning
it would log a warning and drop the status document instead of caching it
as it is supposed to. Now we let router_set_networkstatus() know if the
data comes from an all-request so it can do the right thing.
r8640@totoro (orig r8399): arma | 2006-09-15 01:30:25 -0400
Send out a burst of long-range drop cells after we've established that
we're reachable. Spread them over 4 circuits, so hopefully a few will
be fast. This exercises our bandwidth and bootstraps us quicker.
r8641@totoro (orig r8400): arma | 2006-09-15 01:53:00 -0400
make my bandwidth exercises actually happen
r8642@totoro (orig r8401): arma | 2006-09-15 02:06:09 -0400
two todo items
r8643@totoro (orig r8402): arma | 2006-09-15 14:03:53 -0400
avoid thrashing the bandwidth exercise when we change IPs a lot.
(we avoid simply by not doing any new tests when we change IPs --
it looks like we retain our previous bandwidth estimates, so there's
no need to do new exercise. though in some cases new exercises may
still be useful. one day we'll do something smarter.)
r8644@totoro (orig r8403): arma | 2006-09-15 14:07:11 -0400
actually, do the bandwidth test anyway, if you've been up at
least 6 hours at your previous address.
r8645@totoro (orig r8404): nickm | 2006-09-15 16:19:55 -0400
r8821@Kushana: nickm | 2006-09-15 16:19:16 -0400
Clean up libevent difmingw f problems noticed by mikec.
r8647@totoro (orig r8406): phobos | 2006-09-16 15:54:03 -0400
Re-create symlinks through cp.
r8648@totoro (orig r8407): arma | 2006-09-17 02:18:06 -0400
build testing circuits more rapidly. this has a failure mode
where if circuits fail quickly, we'll fail a lot of them very
quickly and not retry for a while. so be it.
r8649@totoro (orig r8408): arma | 2006-09-17 13:57:56 -0400
a few more todo changes
r8650@totoro (orig r8409): nickm | 2006-09-17 15:58:24 -0400
r8837@Kushana: nickm | 2006-09-17 15:58:04 -0400
More TODO work; tenatively mark assignments.
r8651@totoro (orig r8410): nickm | 2006-09-17 16:12:10 -0400
r8839@Kushana: nickm | 2006-09-17 16:11:59 -0400
Add some client performance XXXXs; try to move some common case tests higher on their decision trees.
r8652@totoro (orig r8411): nickm | 2006-09-17 16:20:23 -0400
r8841@Kushana: nickm | 2006-09-17 16:20:16 -0400
Move more todo items around
r8653@totoro (orig r8412): arma | 2006-09-18 00:24:41 -0400
Fix two bugs: first, "extendcircuit" would crash if you gave it a purpose.
Second, if you give an unknown purpose, it would say:
552 Unknown purpose "purpose=foo"
Now it just says
552 Unknown purpose "foo"
r8656@totoro (orig r8415): arma | 2006-09-18 00:59:15 -0400
good god, that was a bad idea. i've built 500 circuits in the past 5
minutes, trying to establish reachability of my unreachable server.
r8657@totoro (orig r8416): phobos | 2006-09-18 01:35:08 -0400
Fix a mis-spelled Privoxy.
r8668@totoro (orig r8427): nickm | 2006-09-19 16:41:31 -0400
Merge in some bsockets calls, all wrapped inside #if defined(USE_BSOCKETS)
r8669@totoro (orig r8428): nickm | 2006-09-19 16:45:20 -0400
Fix a bogus free() in eventdns.c. Bug reported by xiando.
r8670@totoro (orig r8429): arma | 2006-09-19 17:37:03 -0400
eventdns patch from Adam Langley
r8671@totoro (orig r8430): nickm | 2006-09-19 18:20:09 -0400
Switch routerlist.c to using memcmp on digests rather than crypto_pk_cmp_keys(); speed up find_whitespace a lot (8x for me) by using a switch statement. This should speed parsing a lot of routers at once by a lot.
r8672@totoro (orig r8431): nickm | 2006-09-19 18:36:48 -0400
Malloc and friends are critical-path: Thus, add an it-wont-happen branch prediction for NULL returns, and skip the malloc(0) check on platforms where malloc(0) returns a pointer.
r8673@totoro (orig r8432): nickm | 2006-09-19 19:18:30 -0400
Stop searching routerlist for routers with the same identity as other routers (on router insert): we already have a map for that. (We need to add an index field to routerinfo_t so we can figure out which point in the routerlist to replace.) Also, add a comment to routerlist.c; arma, please advise?
r8674@totoro (orig r8433): nickm | 2006-09-19 19:48:14 -0400
NEEDS REVIEW. Act on previous comment, and handle named servers differently: now, we allow multiple servers with the same name in the routerlist even if that name is reserved, but we check whether names are reserved when we try to look up routers by nickname. This is a minor security fix. This makes router_add_to_routerlist O(1). This is a backport candidate.
r8675@totoro (orig r8434): nickm | 2006-09-19 19:55:35 -0400
Speed up eat_whitespace by a lot.
r8678@totoro (orig r8437): nickm | 2006-09-21 17:48:06 -0400
r8872@Kushana: nickm | 2006-09-21 14:00:20 -0400
Implement server-side reverse DNS using eventdns. Add an option to routerdescs so we can tell which servers have eventdns enabled.
r8679@totoro (orig r8438): nickm | 2006-09-21 17:48:11 -0400
r8873@Kushana: nickm | 2006-09-21 14:38:22 -0400
Fix a bug: Remember, each call to escaped() replaces the value returned from the last call to escaped().
r8680@totoro (orig r8439): nickm | 2006-09-21 17:48:16 -0400
r8874@Kushana: nickm | 2006-09-21 15:22:27 -0400
Rename and document SearchDomains and ResolvConf options; warn if ServerDNSResolvConfFile is given but eventdns isnt enabled.
r8681@totoro (orig r8440): nickm | 2006-09-21 17:48:22 -0400
r8875@Kushana: nickm | 2006-09-21 16:46:28 -0400
Resolve bug 330: detect ISPs that want to hijack failing DNS requests and basically domain-squat the entire internet.
r8682@totoro (orig r8441): nickm | 2006-09-21 17:48:37 -0400
r8876@Kushana: nickm | 2006-09-21 16:58:46 -0400
Trivial cleanup: !!x is a weird way to spell x != NULL.
r8683@totoro (orig r8442): nickm | 2006-09-21 17:48:42 -0400
r8877@Kushana: nickm | 2006-09-21 17:12:33 -0400
Consider non-exit servers unsuitable for RESOLVE commands.
r8684@totoro (orig r8443): nickm | 2006-09-21 17:48:55 -0400
r8878@Kushana: nickm | 2006-09-21 17:15:47 -0400
Trivial whitespace cleanups.
r8685@totoro (orig r8444): nickm | 2006-09-21 17:49:03 -0400
r8879@Kushana: nickm | 2006-09-21 17:20:31 -0400
Oops from earlier patch; add "opt eventdns" to unittests, fix bug in routerdesc generation.
r8686@totoro (orig r8445): nickm | 2006-09-21 17:49:15 -0400
r8880@Kushana: nickm | 2006-09-21 17:26:02 -0400
Trivial patch to appease warn-happy gcc: mark unused parameter
r8687@totoro (orig r8446): nickm | 2006-09-21 17:49:36 -0400
r8881@Kushana: nickm | 2006-09-21 17:27:59 -0400
Allow resolve requests to non-exits when they are specifically requested (via resolve foo.bar.exit).
r8688@totoro (orig r8447): nickm | 2006-09-21 18:24:45 -0400
Fix bug in r8440: base32 uses 5 bytes per char, not 4.
r8689@totoro (orig r8448): nickm | 2006-09-21 18:57:07 -0400
Fix stupid C mistake. Glad I caught that one fast.
r8690@totoro (orig r8449): nickm | 2006-09-21 20:10:26 -0400
Fix a couple of server-side reverse dns bugs
r8691@totoro (orig r8450): arma | 2006-09-21 20:24:27 -0400
remove some loud log messages
r8692@totoro (orig r8451): nickm | 2006-09-21 20:43:55 -0400
r8894@Kushana: nickm | 2006-09-21 18:30:42 -0400
Specify and implement SOCKS5 interface for reverse hostname lookup.
r8693@totoro (orig r8452): nickm | 2006-09-21 20:44:07 -0400
r8895@Kushana: nickm | 2006-09-21 20:05:11 -0400
Debug client-side reverse dns code.
r8694@totoro (orig r8453): nickm | 2006-09-21 20:44:21 -0400
r8695@totoro (orig r8454): nickm | 2006-09-21 20:45:08 -0400
r8901@Kushana: nickm | 2006-09-21 20:43:48 -0400
I tracked a bug in server-side reverse DNS to something concerning the caching code. Ive disabled server-side cacheing for reverse DNS answers for now, and I am noting the bug in the TODO.
r8696@totoro (orig r8455): phobos | 2006-09-21 21:15:07 -0400
Removed verbose patch output.
r8697@totoro (orig r8456): nickm | 2006-09-21 21:23:28 -0400
r8906@Kushana: nickm | 2006-09-21 21:23:22 -0400
Revise patch for libevent 1.1b to handle the "wait, I *do* have a gettimeofday()" case.
r8698@totoro (orig r8457): phobos | 2006-09-22 11:08:05 -0400
Finally, valid tor.exe directions from scratch with mingw.
r8699@totoro (orig r8458): phobos | 2006-09-22 11:37:52 -0400
One last change to openssl to get it to compile in mingw.
r8702@totoro (orig r8461): arma | 2006-09-22 15:29:26 -0400
remove 8888 as a long lived port. i can't remember why it's
on the list.
r8703@totoro (orig r8462): nickm | 2006-09-22 16:18:58 -0400
r8843@Kushana: nickm | 2006-09-17 16:57:20 -0400
A couple of last minor TODO cleanup items
r8704@totoro (orig r8463): nickm | 2006-09-22 16:19:34 -0400
r8912@Kushana: nickm | 2006-09-22 16:18:51 -0400
Write more of path-spec.txt
r8705@totoro (orig r8464): nickm | 2006-09-22 16:20:21 -0400
r8909@Kushana: nickm | 2006-09-22 00:38:13 -0400
Consider changes to has_old_dnsworkers as noncosmetic
r8706@totoro (orig r8465): nickm | 2006-09-22 16:20:26 -0400
r8910@Kushana: nickm | 2006-09-22 12:14:05 -0400
Instead of just checking known-invalid addresses for DNS hijacking, we
now check randomly generated addresses, and if too many of them map to
the same IP, we assume that IP is the destination of a DNS hijack
attempt.
A little bird tells me that some DNS hijackers think that declining to
give an A record for RFC2606 addresses (like .invalid and .example)
makes them more standards compliant. Standardswise, this is like an
illicit brothel making sure that nobody has pulled the tags off the
mattresss, but that doesn't get us out of working around it.
r8707@totoro (orig r8466): nickm | 2006-09-22 16:20:35 -0400
r8911@Kushana: nickm | 2006-09-22 12:24:52 -0400
Make exitlist only output each result line once.
r8708@totoro (orig r8467): phobos | 2006-09-22 22:18:18 -0400
Special mingw package_nsis and nsi scripts.
r8713@totoro (orig r8472): phobos | 2006-09-23 15:17:08 -0400
Changes to doc as a result of creating a MinGW tor.exe on a bare system.
Add two TODO items relating to MinGW instructions.
svn:r8473
|
