Commit Graph

25694 Commits

Author SHA1 Message Date
Nick Mathewson
be741d7e63 Merge branch 'maint-0.3.0' 2017-06-05 09:51:57 -04:00
Nick Mathewson
e5bdfd66cf Make code more clear about own_link_cert safety
It's okay to call add_ed25519_cert with a NULL argument: so,
document that.  Also, add a tor_assert_nonfatal() to catch any case
where we have failed to set own_link_cert when conn_in_server_mode.
2017-06-05 09:35:55 -04:00
rl1987
f8c98759e5 Use string_is_valid_hostname in SOCKS4 request parsing codepath 2017-06-04 13:22:45 +02:00
rl1987
7f05f89663 Don't reject SOCKS5 requests that contain IP strings 2017-06-04 13:14:55 +02:00
rl1987
9e2f780923 Refrain from needless SOCKS5 warning 2017-06-03 18:04:47 +02:00
Nick Mathewson
41ed9e978b Regenerate RSA->ed25519 identity crosscertificate as needed 2017-06-01 10:04:52 -04:00
Nick Mathewson
f2068ef862 Use tor_assert_nonfatal() to try to detect #22466 2017-06-01 09:42:32 -04:00
Nick Mathewson
34a6755b94 Fix ed25519 link certificate race on tls context rotation
Whenever we rotate our TLS context, we change our Ed25519
Signing->Link certificate.  But if we've already started a TLS
connection, then we've already sent the old X509 link certificate,
so the new Ed25519 Signing->Link certificate won't match it.

To fix this, we now store a copy of the Signing->Link certificate
when we initialize the handshake state, and send that certificate
as part of our CERTS cell.

Fixes one case of bug22460; bugfix on 0.3.0.1-alpha.
2017-06-01 09:26:24 -04:00
Nick Mathewson
a9be768959 Bugfix: Regenerate more certificates when appropriate
Previously we could sometimes change our signing key, but not
regenerate the certificates (signing->link and signing->auth) that
were signed with it.  Also, we would regularly replace our TLS x.509
link certificate (by rotating our TLS context) but not replace our
signing->link ed25519 certificate.  In both cases, the resulting
inconsistency would make other relays reject our link handshakes.

Fixes two cases of bug 22460; bugfix on 0.3.0.1-alpha.
2017-05-31 18:45:35 -04:00
Andreas Stieger
1763aa058b Fix GCC 7 -Wimplicit-fallthrough warnings (32 bit)
Add magic comments recognized by default -Wimplicit-fallthrough=3
Follow-up to e5f464, fixes Ticket 22446 for 32 bit.
2017-05-31 09:30:35 -04:00
Roger Dingledine
c361458998 take some of the suggestions from cypherpunks on #6892 2017-05-30 14:17:49 -04:00
Nick Mathewson
5860f0a7c8 Update the torify.1 manpage
I went into this to fix 6892 and say "we don't do anything for
circuit isolation."  But instead I did a fair amount of text-removal
to stop implying that torify does anything more than call torsocks.
2017-05-30 14:15:42 -04:00
Nick Mathewson
4a15b31696 Merge remote-tracking branch 'argonblue/bug22413' 2017-05-30 13:33:49 -04:00
Taylor Yu
90dd7dc92a Check for libzstd >= 1.1
The consensus compression code depends on a streaming compression API
that is new in libzstd-1.1.

Fixes #22413.
2017-05-30 13:25:34 -04:00
Nick Mathewson
c0559b97b5 Merge branch 'maint-0.3.0' 2017-05-30 13:21:06 -04:00
Nick Mathewson
9d59769db7 Improve error message when all permitted Exits are down
The old "No specified non-excluded exit routers seem to be running"
message was somewhat confusing.

Fix for 7890.
2017-05-30 10:59:04 -04:00
David Goulet
5b33d95a3d hs: Correctly validate v3 descriptor encrypted length
The encrypted_data_length_is_valid() function wasn't validating correctly the
length of the encrypted data of a v3 descriptor. The side effect of this is
that an HSDir was rejecting the descriptor and ultimately not storing it.

Fixes #22447

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-05-30 10:27:42 -04:00
Nick Mathewson
671c5dcde3 Merge remote-tracking branch 'public/bug6298' 2017-05-30 10:27:37 -04:00
Nick Mathewson
0fbe1a2c6f Merge remote-tracking branch 'teor/bug22424' 2017-05-30 08:50:45 -04:00
Nick Mathewson
6bc9ff87b8 changes file for 22417 2017-05-30 08:48:50 -04:00
Nick Mathewson
184d889f8a Merge remote-tracking branch 'jigsaw/fix-22417-without-3-star' 2017-05-30 08:47:14 -04:00
Nick Mathewson
f9615f9d77 Merge remote-tracking branch 'teor/bug22421' 2017-05-30 08:42:20 -04:00
Nick Mathewson
167e5be056 changes file for 22446 2017-05-30 08:33:27 -04:00
Andreas Stieger
e5f4642db3 Fix GCC 7 -Wimplicit-fallthrough warnings
Add magic comments recognized by default -Wimplicit-fallthrough=3
or break, as required.
2017-05-30 08:33:27 -04:00
Roger Dingledine
d1580ad49b remove obsolete comment
we should have taken out this comment with commit aadff6274
during ticket 16480.
2017-05-30 02:42:32 -04:00
Daniel Pinto
94d321120e Replace 3-star pointer with 2-star pointer 2017-05-28 20:24:48 +01:00
teor
af891e7f2c
Changes file for bug 22424 2017-05-28 22:34:56 +10:00
teor
79725289e1
If we do underflow the know usage of a storage, recalculate it
Fixes bug #22424 on 0.3.1.1-alpha.
2017-05-28 22:34:43 +10:00
teor
69b234a0a8
Refactor storage usage reductions into a static function
No behaviour change.

Part of #22424.
2017-05-28 22:28:43 +10:00
teor
334fe6bb6b
Don't underflow usage when it is unknown and a file is removed
Part of #22424.
2017-05-28 22:16:00 +10:00
teor
9e36b0beb9
Always check for usage underflow when removing a file in storage.c
Part of #22424.
2017-05-28 22:12:09 +10:00
teor
f6841ae263
Fix comment typos in storage.c 2017-05-28 22:11:22 +10:00
teor
25ea8be9de
Update the client bootstrap comment in config.c for exponential backoff
This brings the description up to date with the exponential backoff
code introduced in 0.2.9.1-alpha.

Fixes bug #22421.
2017-05-28 21:01:08 +10:00
Daniel Pinto
f8ccf8d9a9 Fix crash with %include
Fixes crash when including a folder that contains a non-empty file
without any values followed by any other non-empty file.
2017-05-28 09:53:14 +01:00
Roger Dingledine
084b64ba2e simplify because relay_crypt_one_payload can't fail 2017-05-28 01:51:22 -04:00
Nick Mathewson
6fcaf83c98 Cleanup MOCK_IMPL (etc) to be findable with etags
A fair number of our mock_impl declarations were messed up so that
even our special AM_ETAGSFLAGS couldn't find them.

This should be a whitespace-only patch.
2017-05-26 14:07:06 -04:00
Nick Mathewson
159a8061cf Mock tor_addr_lookup() during part of addr/basic test.
If this function isn't mocked, then our unit tests break on stupid
networks where localhost is broken or absent. Fixes bug 6298; bugfix
on 0.0.9pre2.
2017-05-26 13:53:32 -04:00
Nick Mathewson
5742e4fd8e bump version to 0.3.1.2-alpha-dev 2017-05-26 10:23:53 -04:00
Nick Mathewson
1405bdebb0 Update releasing-tor to reflect current versions and tooling
(Note that a lot of the removed guidance is stuff that the tools
will do automatically.)
2017-05-26 10:01:04 -04:00
Nick Mathewson
61625b8f26 Pedantry in the changelog. 2017-05-26 09:10:27 -04:00
Nick Mathewson
0870660716 edit changelog 2017-05-26 09:04:51 -04:00
Nick Mathewson
8cd01f5e5b Start an 0.3.1.2-alpha mini-changelog 2017-05-26 09:00:38 -04:00
Nick Mathewson
2550ac7bb1 update version to 0.3.1.2-alpha 2017-05-26 08:58:08 -04:00
Nick Mathewson
ab9976b724 Merge remote-tracking branch 'arma/bug22368' 2017-05-25 08:54:51 -04:00
Roger Dingledine
657297a9f8 Merge branch 'maint-0.3.0' 2017-05-25 00:28:11 -04:00
Roger Dingledine
83439e78cc Merge branch 'maint-0.2.9' into maint-0.3.0 2017-05-25 00:27:27 -04:00
teor
ec61ae59a5 Stop leaking keypin-rejected routerinfos on directory authorities
When directory authorities reject a router descriptor due to keypinning,
free the router descriptor rather than leaking the memory.

Fixes bug 22370; bugfix on 0.2.7.2-alpha.
2017-05-25 00:09:40 -04:00
Roger Dingledine
5f74749fba get rid of some dead code (leftover from commit fa04fe1) 2017-05-24 23:37:00 -04:00
Roger Dingledine
d22d565331 add copy of MyFamily element to the descriptor, not the element itself
If we add the element itself, we will later free it when we free the
descriptor, and the next time we go to look at MyFamily, things will
go badly.

Fixes the rest of bug 22368; bugfix on 0.3.1.1-alpha.
2017-05-24 23:37:00 -04:00
Roger Dingledine
a7e75ff796 don't free the values in options->MyFamily when we make a descriptor
If we free them here, we will still attempt to access the freed memory
later on, and also we will double-free when we are freeing the config.

Fixes part of bug 22368.
2017-05-24 23:32:32 -04:00