Commit Graph

26777 Commits

Author SHA1 Message Date
Nick Mathewson
7dddef2a76 Merge branch 'maint-0.3.1' into maint-0.3.2 2017-11-13 11:15:50 -05:00
Nick Mathewson
666653758d Merge branch 'bug8185_025' into maint-0.2.9 2017-11-13 11:15:47 -05:00
Nick Mathewson
0fc2b7f82e Merge branch 'bug8185_031' into maint-0.3.1 2017-11-13 11:15:40 -05:00
Nick Mathewson
ece63105d9 Merge branch 'maint-0.2.9' into maint-0.3.0 2017-11-13 11:13:22 -05:00
Nick Mathewson
441bfcd885 Merge branch 'maint-0.3.1' into maint-0.3.2 2017-11-13 11:13:22 -05:00
Nick Mathewson
2777b469fa Merge branch 'maint-0.3.0' into maint-0.3.1 2017-11-13 11:13:22 -05:00
Nick Mathewson
c928fb988a Merge branch 'ticket21953_029' into maint-0.2.9 2017-11-13 11:13:18 -05:00
Nick Mathewson
a7ca71cf6b Fix mock_crypto_pk_public_checksig__nocheck() to handle short RSA keys
This function -- a mock replacement used only for fuzzing -- would
have a buffer overflow if it got an RSA key whose modulus was under
20 bytes long.

Fortunately, Tor itself does not appear to have a bug here.

Fixes bug 24247; bugfix on 0.3.0.3-alpha when fuzzing was
introduced.  Found by OSS-Fuzz; this is OSS-Fuzz issue 4177.
2017-11-11 14:44:45 -05:00
Nick Mathewson
512dfa15ed Merge branch 'maint-0.3.1' into maint-0.3.2 2017-11-11 12:23:34 -05:00
Nick Mathewson
ef92511577 Merge branch 'maint-0.3.0' into maint-0.3.1 2017-11-11 12:23:34 -05:00
Nick Mathewson
9666c620d5 Merge branch 'maint-0.2.8' into maint-0.2.9 2017-11-11 12:23:33 -05:00
Nick Mathewson
c9b76a12f5 Merge branch 'maint-0.2.5' into maint-0.2.8 2017-11-11 12:23:33 -05:00
Nick Mathewson
5348c8996f Merge branch 'maint-0.2.9' into maint-0.3.0 2017-11-11 12:23:33 -05:00
David Goulet
42ce6f78c8 control: Fix wrong Action on failure to upload an HS descriptor
On failure to upload, the HS_DESC event would report "UPLOAD_FAILED" as the
Action but it should have reported "FAILED" according to the spec.

Fixes #24230

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-11-10 15:31:31 -05:00
Karsten Loesing
6f8c32b7de Update geoip and geoip6 to the November 6 2017 database. 2017-11-10 10:14:09 +01:00
Nick Mathewson
aa1d90b40c Merge branch 'bug23985_029' into maint-0.3.2 2017-11-09 08:15:46 -05:00
Nick Mathewson
40c64f45f0 Re-remove AllowDotExit deprecation.
This was accidentally reintroduced in 5a46074e55.  It should be
harmless, though: the option is obsolete, so calling it deprecated
is redundant.
2017-11-09 07:58:27 -05:00
Nick Mathewson
c8859582a2 bump to 0.3.2.4-alpha-dev 2017-11-08 16:44:53 -05:00
Nick Mathewson
b84b008172 Merge remote-tracking branch 'dgoulet/bug24186_032_01' into maint-0.3.2 2017-11-08 14:49:16 -05:00
David Goulet
fa70aabb62 dirauth: Recalculate voting schedule at first vote
Commit e67f4441eb introduced a safeguard against
using an uninitialized voting schedule object. However, the dirvote_act() code
was looking roughly at the same thing to know if it had to compute the timings
before voting with this condition:

  if (!voting_schedule.voting_starts) {
    ...
    dirvote_recalculate_timing(options, now);
  }

The sr_init() function is called very early and goes through the safeguard
thus the voting schedule is always initilized before the first vote.

That first vote is a crucial one because we need to have our voting schedule
aligned to the "now" time we are about to use for voting. Then, the schedule
is updated when we publish our consensus or/and when we set a new consensus.
From that point on, we only want to update the voting schedule through that
code flow.

This "created_on_demand" is indicating that the timings have been recalculated
on demand by another subsystem so if it is flagged, we know that we need to
ignore its values before voting.

Fixes #24186

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-11-08 14:46:49 -05:00
Nick Mathewson
61330f40a5 Merge remote-tracking branch 'dgoulet/ticket24134_032_01' into maint-0.3.2 2017-11-08 13:26:46 -05:00
Nick Mathewson
0dc55fb247 Don't delay descriptor fetches when missing info needed for circuits
When we have fewer than 15 descriptors to fetch, we will delay the
fetch for a little while.  That's fine, if we can go ahead and build
circuits... but if not, it's a poor choice indeed.

Fixes bug 23985; bugfix on 0.1.1.11-alpha.

In 0.3.0.3-alpha, when we made primary guard descriptors necessary
for circuit building, this situation got worse.
2017-11-08 13:22:16 -05:00
Nick Mathewson
31a907881e downgrade severity for get_voting_schedule() 2017-11-08 11:50:57 -05:00
Nick Mathewson
02f05f2de2 Merge remote-tracking branch 'dgoulet/bug23751_032_01' into maint-0.3.2 2017-11-08 10:56:33 -05:00
Nick Mathewson
c74a817e83 Merge branch 'bug23318-redux_029' into maint-0.3.2 2017-11-08 10:47:22 -05:00
teor
6e4ebd41bb Stop calculating total twice in frac_nodes_with_descriptors()
Cleanup after 23318.
2017-11-08 10:45:18 -05:00
teor
4f944cc4cc Check arguments and initialise variables in compute_weighted_bandwidths()
Cleanup after 23318.
2017-11-08 10:45:18 -05:00
teor
fcaa4ab824 Actually log the total bandwidth in compute_weighted_bandwidths()
Fixes bug 24170; bugfix on 0.2.4.3-alpha.
2017-11-08 10:45:18 -05:00
teor
14b0bba06e Use node counts in networks with all zero-bandwidths
When calculating the fraction of nodes that have descriptors, and all
all nodes in the network have zero bandwidths, count the number of nodes
instead.

Fixes bug 23318; bugfix on 0.2.4.10-alpha.
2017-11-08 10:45:18 -05:00
Nick Mathewson
3c03e237ab Remove an erroneous 0.5 in compute_weighted_bandwidths()
Back in 0.2.4.3-alpha (e106812a77), when we switched from using
double to using uint64 for selecting by bandwidth, I got the math
wrong: I should have used llround(x), or (uint64_t)(x+0.5), but
instead I wrote llround(x+0.5).  That means we would always round
up, rather than rounding to the closest integer

Fixes bug 23318; bugfix on 0.2.4.3-alpha.
2017-11-08 10:45:18 -05:00
David Goulet
ac79f831e1 doc: Put the release timeline link in README
Closes #24134

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-11-08 10:44:00 -05:00
David Goulet
dcabf801e5 sched: Ignore closed channel after flushing cells
The flush cells process can close a channel if the connection write fails but
still return that it flushed at least one cell. This is due because the error
is not propagated up the call stack so there is no way of knowing if the flush
actually was successful or not.

Because this would require an important refactoring touching multiple
subsystems, this patch is a bandaid to avoid the KIST scheduler to handle
closed channel in its loop.

Bandaid on #23751.

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-11-08 09:44:39 -05:00
Nick Mathewson
e5a83062ed Merge branch 'bug23816_029_squashed' into maint-0.3.2 2017-11-08 08:11:22 -05:00
Nick Mathewson
cb29687e93 Replace our random-exponential-delay algorithm.
This patch has implementations of the "decorrelated" and "full"
algorithms from https://www.awsarchitectureblog.com/2015/03/backoff.html
2017-11-08 07:56:16 -05:00
Nick Mathewson
3dc61a5d71 Revert "Remove an erroneous 0.5 in compute_weighted_bandwidths()"
This reverts commit 01e984870a.
2017-11-07 20:35:19 -05:00
Nick Mathewson
c58471325a Merge remote-tracking branch 'dgoulet/bug24161_032_01' into maint-0.3.2 2017-11-07 14:00:18 -05:00
Nick Mathewson
c7d1d1ad2e Bump to 0.3.2.4-alpha 2017-11-07 13:22:27 -05:00
David Goulet
e67f4441eb Add a safe guard to avoid using a zeroed voting schedule
dirvote_get_next_valid_after_time() is the only public function that uses the
voting schedule outside of the dirvote subsystem so if it is zeroed,
recalculate its timing if we can that is if a consensus exists.

Part of #24161

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-11-07 11:14:45 -05:00
David Goulet
8171d9f50f Recalculate voting schedule first when getting a new consensus
Because the HS and SR subsystems can use the voting schedule early (with the
changes in #23623 making the SR subsystem using the static voting schedule
object), we need to recalculate the schedule very early when setting the new
consensus.

Fixes #24161

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-11-07 11:08:12 -05:00
Nick Mathewson
9d848ca58a Merge remote-tracking branch 'asn/ticket23623_032_01' into maint-0.3.2 2017-11-07 09:58:42 -05:00
George Kadianakis
023f62bda4 Add changes file for #23623. 2017-11-07 00:28:14 +02:00
Nick Mathewson
7d767706ef Merge branch 'bug24150_032_squashed' into maint-0.3.2 2017-11-06 12:59:23 -05:00
Nick Mathewson
5240afa713 Fix a memory leak on decryption non-failure of v3 hsdesc
If it decrypts something that turns out to start with a NUL byte,
then decrypt_desc_layer() will return 0 to indicate the length of
its result.  But 0 also indicates an error, which causes the result
not to be freed by decrypt_desc_layer()'s callers.

Since we're trying to stabilize 0.3.2.x, I've opted for the simpler
possible fix here and made it so that an empty decrypted string will
also count as an error.

Fixes bug 24150 and OSS-Fuzz issue 3994.

The original bug was present but unreachable in 0.3.1.1-alpha. I'm
calling this a bugfix on 0.3.2.1-alpha since that's the first version
where you could actually try to decrypt these descriptors.
2017-11-06 12:59:11 -05:00
Nick Mathewson
bebdd91058 Fix a 32-bit formatting warning 2017-11-06 10:16:10 -05:00
Nick Mathewson
c80293706f Merge remote-tracking branch 'public/bug23318_029' into maint-0.3.2 2017-11-06 09:48:51 -05:00
Nick Mathewson
f46cdb3ff8 lintchanges: Allow 'fixes bugs a, b, and c' 2017-11-05 14:41:39 -05:00
Nick Mathewson
9f3d9c68ab Merge branch 'bug21394_029' into maint-0.3.2 2017-11-05 14:39:37 -05:00
Sebastian Hahn
9827574308 Fix DNS resolution on busy exit relays
Thanks to Dhalgren who analyzed this issue two years ago and found a
solution!
2017-11-05 14:36:32 -05:00
Nick Mathewson
7c1f33dd0e ticket21031 changes file revised. 2017-11-05 14:09:07 -05:00
Sebastian Hahn
d3fee8b823 ClientDNSRejectInternalAddresses in non-default networks
Once again allow the flag to be set, unless the default network is used.
Thanks to nickm for a suggestion for the workaround to a test failure.
2017-11-04 18:31:02 +01:00