 Nick Mathewson
|
f28fc83ea5
More logging for mtbf/wfu calculations.
|
16 years ago |
 Roger Dingledine
|
29abfab8bf
Update to the "September 1 2008" ip-to-country file.
|
16 years ago |
 Karsten Loesing
|
e4942e809e
Exclude v0 rendezvous descriptors when considering republication.
|
16 years ago |
|
Roger Dingledine
|
6a7abbd25d
ok, but inaccurate error strings worse
|
16 years ago |
|
Roger Dingledine
|
0fc0d5266c
identical error strings in different contexts bad
|
16 years ago |
|
Nick Mathewson
|
b0c48d8e45
Oops; we need to make sure that DNS request names are matched in the _questions_ section of the replies. Rejecting answers whether the _answers_ section did not match made us reject A records waiting at the end of a CNAME record. Bug 823.
|
16 years ago |
|
Roger Dingledine
|
e761a3a50e
add some more papers we tend to cite
|
16 years ago |
|
Roger Dingledine
|
4fd9880540
give rend_client_desc_here a slightly more accurate name
|
16 years ago |
|
Roger Dingledine
|
38f2272d5d
patch from karsten for bug 814. whew.
|
16 years ago |
|
Nick Mathewson
|
3db8c15287
Add a circuit_conforms_to_options() function for use in debugging paths [and inother stuff too]. Untested, and so far unused.
|
16 years ago |
|
Roger Dingledine
|
2bde30efa6
read_to_buf should take an int *error_socket and return it.
|
16 years ago |
|
Roger Dingledine
|
7f2fd34645
stop telling us every time we fetch a consensus. that's not really news.
|
16 years ago |
|
Roger Dingledine
|
dbb481c169
Serve the latest v3 networkstatus consensus via the control
|
16 years ago |
|
Roger Dingledine
|
4045ef72f5
make it explicit that the uncompressed cached dir items are nul-terminated
|
16 years ago |
|
Roger Dingledine
|
2d95accc75
Allow ports 465 and 587 in the default exit policy again. We had
|
16 years ago |
 Andrew Lewman
|
abb5febd8a
oops, forgot the changelog reference for r16867.
|
16 years ago |
|
Andrew Lewman
|
abce15b993
Patch from Camilo Viecco to make the rpm spec arch independent, as far
|
16 years ago |
|
Karsten Loesing
|
f0a5ef804f
Directory mirrors store and serve v2 hidden service descriptors by default.
|
16 years ago |
|
Roger Dingledine
|
535e2074bc
give lasse's last name a different charset
|
16 years ago |
|
Roger Dingledine
|
2f18370821
and here too
|
16 years ago |
|
Nick Mathewson
|
b322348e8a
Fix more actual test leaks
|
16 years ago |
|
Nick Mathewson
|
45205126fe
Fix some apparent leaks in voting. If authorities suddenly start segfaulting, this could be a patch to look harder at.
|
16 years ago |
|
Nick Mathewson
|
aa69d586ea
Make buffer unit tests handle resource leaks properly.
|
16 years ago |
|
Nick Mathewson
|
38f56608d9
Make more unit tests handle resource leaks properly.
|
16 years ago |
|
Nick Mathewson
|
1203850350
Coverity says it is okay to acknowledge them. Do so. Their tool is awesome.
|
16 years ago |
|
Karsten Loesing
|
52fbfc5d45
Correct indentation.
|
16 years ago |
|
Roger Dingledine
|
659f1651e0
tweak karsten's patch
|
16 years ago |
|
Roger Dingledine
|
2c14705a4d
patch from karsten to fix more of bug 767
|
16 years ago |
|
Nick Mathewson
|
f95d7c189b
Refactor unit test macros and tor_free_all() logic a bit so as to make it easier to free memory on failing tests, in order to suppress scanner warnings and to make dmalloc() usable with tests.
|
16 years ago |
|
Nick Mathewson
|
6c432a5565
Replace the dummy-use var in SMARTLIST_FOREACH_END() with one that is less likely to confuse analysis tools into thinking we do use after free. Arguably, (void)x should count as use in suppressing unused variable warnings, but not in generating hey-you-used-a-variable warnings. Arguably, though, it shouldn't.
|
16 years ago |
