Commit Graph

23659 Commits

Author SHA1 Message Date
Nick Mathewson
f5fdf188b9 Merge branch 'maint-0.2.6' into maint-0.2.7 2016-11-07 09:30:42 -05:00
Nick Mathewson
6c2174d44d Merge branch 'maint-0.2.5' into maint-0.2.6 2016-11-07 09:30:16 -05:00
Nick Mathewson
db2571be61 Merge branch 'maint-0.2.4' into maint-0.2.5 2016-11-07 09:29:54 -05:00
Nick Mathewson
d82ffb77f3 Merge branch '20499_part1_029_squashed', remote-tracking branches 'teor/bug20591_029' and 'teor/bug20533_029' into maint-0.2.9 2016-11-07 09:20:13 -05:00
Nick Mathewson
858867a31a Allow infinitely long delays in exponential-backoff downloads
It's only safe to remove the failure limit (per 20536) if we are in
fact waiting a bit longer each time we try to download.

Fixes bug 20534; bugfix on 0.2.9.1-alpha.
2016-11-07 09:19:35 -05:00
Nick Mathewson
ce8a1e25a2 Merge branch 'bug20588' into maint-0.2.9 2016-11-07 09:08:58 -05:00
Karsten Loesing
ea597832e2 Update geoip and geoip6 to the November 3 2016 database. 2016-11-07 15:05:19 +01:00
teor
e819d420c5
When downloading certificates, check for related failures
If a consensus expires while we are waiting for certificates to download,
stop waiting for certificates.

If we stop waiting for certificates less than a minute after we started
downloading them, do not consider the certificate download failure a
separate failure.

Fixes bug 20533; bugfix on commit e0204f21 in 0.2.0.9-alpha.
2016-11-08 00:01:20 +11:00
teor
1bb28cecd9
Ensure relays don't make multiple connections during bootstrap
Relays do not deliberately launch multiple attempts, so the impact of this
bug should be minimal. This fix also defends against bugs like #20499.

Bugfix on 0.2.8.1-alpha.
2016-11-07 23:05:55 +11:00
Nick Mathewson
0bd55ed96a Always Use EVP_aes_*_ctr() with openssl 1.1
(OpenSSL 1.1 makes EVP_CIPHER_CTX opaque, _and_ adds acceleration
for counter mode on more architectures.  So it won't work if we try
the older approach, and it might help if we try the newer one.)

Fixes bug 20588.
2016-11-06 21:01:25 -05:00
Nick Mathewson
f6a3d213e4 Finish a sentence in a comment. Close 20576. 2016-11-06 20:31:50 -05:00
Nick Mathewson
def41e93bd In test_tortls_classify_client_ciphers(), s/ECDH/ECDHE/
(We weren't actually using these ciphers; we were just requing that
ciphers of that name existed.)

Patch from rubiate.  Fixes 20460
2016-11-06 20:23:40 -05:00
Nick Mathewson
5385a023e1 Do not apply 'max_failures' to random-exponential schedules.
Fixes bug 20536; bugfix on 0.2.9.1-alpha.
2016-11-06 20:08:11 -05:00
Nick Mathewson
e9ce181955 Change a BUG warning to be a warning, not an info. 2016-11-06 20:01:24 -05:00
Nick Mathewson
1b22eae120 Fix get_delay() code to avoid TIME_MAX overflow, not INT_MAX.
Fixes bug 20587; bugfix on 35bbf2e4a4 in 0.2.8.1-alpha.
2016-11-06 19:50:08 -05:00
Nick Mathewson
add164aa41 Fix warnings from lintChanges.py 2016-11-06 16:39:46 -05:00
Nick Mathewson
59f4cae68c Merge branch 'maint-0.2.8' into maint-0.2.9 2016-11-03 18:36:43 -04:00
Nick Mathewson
61bdc452b0 Merge branch 'bug20551_028' into maint-0.2.8 2016-11-03 18:36:25 -04:00
Nick Mathewson
3cd520a52d Merge branch 'maint-0.2.8' into maint-0.2.9 2016-11-03 15:44:46 -04:00
Nick Mathewson
7a45ef5a47 Merge remote-tracking branch 'arma/bug19969_028_squashed' into maint-0.2.8 2016-11-03 15:44:30 -04:00
Nick Mathewson
3bb49c0110 Merge branch 'maint-0.2.8' into maint-0.2.9 2016-11-03 15:41:04 -04:00
Nick Mathewson
9b18b215bb Work around a behavior change in openssl's BUF_MEM code
In our code to write public keys to a string, for some unfathomable
reason since 253f0f160e, we would allocate a memory BIO, then
set the NOCLOSE flag on it, extract its memory buffer, and free it.
Then a little while later we'd free the memory buffer with
BUF_MEM_free().

As of openssl 1.1 this doesn't work any more, since there is now a
BIO_BUF_MEM structure that wraps the BUF_MEM structure.  This
BIO_BUF_MEM doesn't get freed in our code.

So, we had a memory leak!

Is this an openssl bug?  Maybe.  But our code was already pretty
silly.  Why mess around with the NOCLOSE flag here when we can just
keep the BIO object around until we don't need the buffer any more?

Fixes bug 20553; bugfix on 0.0.2pre8
2016-11-03 10:51:10 -04:00
Nick Mathewson
464783a8dc Use explicit casts to avoid warnings when building with openssl 1.1
fixes bug 20551; bugfix on 0.2.1.1-alpha
2016-11-03 09:35:41 -04:00
teor
0ee9049e94
Use the latest options in rend_service_check_private_dir
Fixup on both:
* Refactor, adding a create argument... and
* Check every hidden service directory's permissions...
2016-11-02 14:55:14 +11:00
teor
1747f28861
Check every hidden service directory's permissions when configuring
Previously, we would only check the last hidden service directory.

Fixes #20529, bugfix on ticket 13942 commit 85bfad1 in 0.2.6.2-alpha.
2016-11-02 14:32:04 +11:00
teor
01fe039b78
Test single onion service configs where the directory does not exist
Runs a test for each combination of create/don't create directories.

Tests #20484.
2016-11-02 14:17:52 +11:00
teor
13fbbe9cce
Make sure passthrough_test_setup doesn't inadvertently fail or skip tests
passthrough_test_setup doesn't pass through arguments if the argument
is equal to 0 or TT_SKIP. Instead, it fails or skips the test.

Assert on this, so we don't accidentally fail or skip tests.
2016-11-02 14:16:14 +11:00
teor
77e1d660ee
Add get_fname_rnd for unit tests that want a unique path every time 2016-11-02 14:14:19 +11:00
teor
d7634dc519
Create get_fname_suffix, and refactor get_fname to use it 2016-11-02 14:13:34 +11:00
teor
2f48693663
Improve comments in check_private_dir and onion poisoning
Comment changes only
2016-11-02 14:11:26 +11:00
teor
a906ff88a3
fixup! Refactor, adding a create argument to rend_service_check_private_dir 2016-11-02 14:10:52 +11:00
teor
fedafe7c0e
Use check_private_dir in test_single_onion_poisoning
This avoids Win32 conditionals for mkdir.
2016-11-02 11:37:11 +11:00
teor
6c54181559
Remove redundant group permission code from rend_service_check_private_dir
check_private_dir already does this for existing directories.
2016-11-02 11:20:49 +11:00
teor
c9db775243
Refactor, adding a create argument to rend_service_check_private_dir
It used to be rend_service_check_and_create_private_dir, which always
created the directory.

No behaviour change.
2016-11-02 11:20:19 +11:00
Nick Mathewson
b2f82d45b7 Always call connection_ap_attach_pending() once a second.
Fixes bug 19969; bugfix on b1d56fc58.  We can fix this some more in
later Tors, but for now, this is probably the simplest fix possible.

This is a belt-and-suspenders fix, where the earlier fix ("Ask
event_base_loop to finish when we add a pending stream") aims to respond
to new streams as soon as they arrive, and this one aims to make sure
that we definitely respond to all of the streams.
2016-11-01 20:09:44 -04:00
Roger Dingledine
d89804a69d Ask event_base_loop to finish when we add a pending stream
Fixes bug 19969; bugfix on b1d56fc58. We can fix this some more in
later Tors, but for now, this is probably the right fix for us.
2016-11-01 19:52:55 -04:00
Roger Dingledine
28b755e660 refactor out the tor_event_base_loopexit() call
no actual changes
2016-11-01 19:52:54 -04:00
Nick Mathewson
ff3e08f2af Attempt to fix unit tests on netbsd 2016-11-01 13:32:21 -04:00
Nick Mathewson
733b245283 Merge remote-tracking branch 'teor/bug20472-029-v2' into maint-0.2.9 2016-11-01 13:05:33 -04:00
Nick Mathewson
9248466f0a Merge branch 'bug20487_029' into maint-0.2.9 2016-11-01 12:44:20 -04:00
Nick Mathewson
e712b5d784 Merge branch 'bug19968_029' into maint-0.2.9 2016-10-31 16:33:03 -04:00
Nick Mathewson
24b7b922ae Actually free the worker_state_t object when we do an update with it
Previously we freed the old "keys" object, but leaked the
worker_state_t that we had taken it from.

Fixes bug 20401; bugfix on 0.2.6.3-alpha.
2016-10-31 15:20:25 -04:00
Nick Mathewson
b858452f94 Add a sentence to the manpage about nonanonymous=>Socksport 0.
Closes 20487.
2016-10-31 15:13:27 -04:00
Nick Mathewson
d73c671d6d policy_is_reject_star():
ome policies are default-reject, some default-accept.  But
policy_is_reject_star() assumed they were all default_reject.  Fix
that!

Also, document that policy_is_reject_star() treats a NULL policy as
empty. This allows us to simplify the checks in
parse_reachable_addresses() by quite a bit.

Fxes bug 20306; bugfix on 0.2.8.2-alpha.
2016-10-31 15:05:56 -04:00
Nick Mathewson
8841a9e396 Create single-onion-service directory before poisoning it, if needed
(Also, refactor the code to create a hidden service directory into a
separate funcion, so we don't have to duplicate it.)

Fixes bug 20484; bugfix on 0.2.9.3-alpha.
2016-10-31 14:54:20 -04:00
Nick Mathewson
becc957839 Actually clamp the number of detected CPUs to 16.
Previously we said we did, but didn't.

Fixes #19968; bugfix on 0.2.3.1-alpha.
2016-10-31 14:19:39 -04:00
teor
3158777d35
Rename routerstatus_version_supports_ntor to *_supports_extend2_cells
This helps avoid the confusion that caused bug 20472.
Bugfix on commit 10aa913 from #19163 in tor-0.2.9.3-alpha.
2016-10-31 17:02:21 +11:00
teor
d81f2b5d81
Refactor circuit_pick_create_handshake
Make the structure of circuit_pick_create_handshake consistent with
circuit_pick_extend_handshake.

No behaviour change.
2016-10-31 17:01:33 +11:00
teor
26d7a07ad7
In circuit_pick_extend_handshake, assume all hops support EXTEND2 and ntor
This simplifies the function: if we have an ntor key, use ntor/EXTEND2,
otherwise, use TAP/EXTEND.

Bugfix on commit 10aa913 from 19163 in 0.2.9.3-alpha.
2016-10-31 17:01:04 +11:00
Nick Mathewson
d288704023 Avoid tor_fragile_assert() failure with DNSPort on RESOLVED_TYPE_ERROR
The tor_fragile_assert() bug has existed here since c8a5e2d588
in tor-0.2.1.7-alpha forever, but tor_fragile_assert() was mostly a
no-op until 0.2.9.1-alpha.

Fixes bug 19869.
2016-10-26 14:16:40 -04:00