Commit graph

25708 commits

Author SHA1 Message Date
David Goulet
40a321659a hs-v3: Remove BUG() that can occur normally
Fixes #28992

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 18:40:55 +02:00
David Goulet
f0964628e6 Merge branch 'ticket33029_042_01' into ticket33029_043_03
Conflicts:
	doc/tor.1.txt
	src/app/config/config.c
	src/app/config/or_options_st.h
	src/core/mainloop/connection.h

Between 042 and 043, the dirauth options were modularized so this merge commit
address this by moving the AuthDirRejectUncompressedRequests to the module
along with a series of accessors.

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 10:30:29 -05:00
David Goulet
7b4d9fabef dirauth: Rename function for better clarity
Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 10:15:23 -05:00
David Goulet
33414e5494 test: Add unit test for connection_dir_is_global_write_low()
Part of #33029

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 10:15:04 -05:00
David Goulet
6d9113d2f6 dirauth: Resume sending 503 directory error code
Authorities were never sending back 503 error code because by design they
should be able to always answer directory requests regardless of bandwidth
capacity.

However, that recently backfired because of a large number of requests from
unknown source using the DirPort that are _not_ getting their 503 code which
overloaded the DirPort leading to the authority to be unable to answer to its
fellow authorities.

This is not a complete solution to the problem but it will help ease off the
load on the authority side by sending back 503 codes *unless* the connection
is from a known relay or an authority.

Fixes #33029

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:58:28 -05:00
David Goulet
735aa208b1 dirauth: Add option AuthDirRejectRequestsUnderLoad
This controls the previous feature added that makes dirauth send back a 503
error code on non relay connections if under bandwidth pressure.

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:57:05 -05:00
David Goulet
bd4f4cb5f0 dirlist: Add configured trusted dir to the nodelist address set
The configured, within the torrc or hardcoded, directory authorities addresses
are now added to the nodelist address set.

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:35:54 -05:00
David Goulet
4152c349b4 nodelist: Helper to add an address to the nodelist address set
We separate v4 and v6 because we often use an IPv4 address represented with
a uint32_t instead of a tor_addr_t.

This will be used to also add the trusted directory addresses taken from the
configuration.

The trusted directories from the consensus are already added to the address
set from their descriptor.

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:35:50 -05:00
David Goulet
872f95ca06 mainloop: Modernize a bit connection_dir_is_global_write_low()
Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:35:46 -05:00
David Goulet
c1e0ac63b8 mainloop: Remove unused parameter from connection_dir_is_global_write_low()
Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:35:40 -05:00
David Goulet
ee55823a11 mainloop: Rename global_write_bucket_low()
That function is only used to test the global bucket write limit for a
directory connection.

It should _not_ be used for anything else since that function looks to see if
we are a directory authority.

Rename it to something more meaningful. No change in behavior at this commit,
only renaming.

Part of #33029

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-11 09:34:55 -05:00
Nick Mathewson
384a771fcc Merge branch 'ticket32362_squashed' 2020-02-11 08:47:19 -05:00
Neel Chauhan
777d90fa23 Check for leading zeros in tor_inet_aton() 2020-02-11 08:47:13 -05:00
Nick Mathewson
71d96cefc0 Update to 0.4.4.0-alpha-dev 2020-02-10 19:43:54 -05:00
Nick Mathewson
5d56c1f4b2 Update to 0.4.3.2-alpha-dev 2020-02-10 19:42:31 -05:00
George Kadianakis
901ed35709 Make n_subcredentials a size_t .
Based on David's review.
2020-02-11 02:13:40 +02:00
Nick Mathewson
eb8841cd95 Merge branch 'bug33104_041' 2020-02-10 14:23:13 -05:00
Nick Mathewson
0c90fc37ec Merge branch 'maint-0.4.1' into maint-0.4.2 2020-02-10 14:17:10 -05:00
Nick Mathewson
71c49f7356 Merge branch 'maint-0.4.0' into maint-0.4.1 2020-02-10 14:17:10 -05:00
Nick Mathewson
d9144700af Merge branch 'maint-0.4.2' 2020-02-10 14:17:10 -05:00
Nick Mathewson
f3fa22bf1b Merge branch 'maint-0.3.5' into maint-0.4.0 2020-02-10 14:17:09 -05:00
Nick Mathewson
dab0b1c3a7 bump to 0.4.3.2-alpha 2020-02-10 13:37:59 -05:00
Nick Mathewson
99d044d553 Fix a Rust compilation warning; resolve bug 33212. 2020-02-10 13:32:09 -05:00
Nick Mathewson
d9e211ab70 Use semicolons after HT_PROTOTYPE and HT_GENERATE. 2020-02-10 12:54:43 -05:00
Nick Mathewson
a9cc4ce0eb ht.h: Require a semicolon after HT_PROTOTYPE and HT_GENERATE[2] 2020-02-10 12:54:06 -05:00
Nick Mathewson
b2c3cb1b26 Add tests for control_event_signal. 2020-02-10 11:38:36 -05:00
teor
5e6021bdae
stats: Fix a function comment typo in rephist.c
Comment-only change.
2020-02-10 15:58:25 +10:00
Nick Mathewson
fbc1eaa0af Try to shorten an #error in address.c 2020-02-06 17:21:34 -05:00
Nick Mathewson
1651f92c16 Break CONNECTION_TESTCAE_ARG across multiple lines 2020-02-06 17:21:34 -05:00
Nick Mathewson
8a5a1600cd Extract verbatim table in uname.c 2020-02-06 17:06:12 -05:00
Nick Mathewson
7036ed3471 Fix a couple more long warning lines
These are not a problem with 2-space indentation, but cocci will
start getting confused when clang-format wraps them with 4-space
indentation.
2020-02-06 16:25:56 -05:00
Nick Mathewson
6076adde25 circuitbuild: expect bug message that clang-format will generate.
clang-format wants to put no space here, so we need to tell the test
to expect a lack of a space.
2020-02-06 15:34:06 -05:00
Nick Mathewson
99a5aecbc7 Wrap columnar tables in "clang-format off/on"
These tables have aligned comments, so we don't want clang-format to
mess with them.
2020-02-06 14:54:04 -05:00
Nick Mathewson
06a6130666 Use parentheses to avoid mis-indentations of stringified macro args
clang-format sometimes thinks that "#name" should be written as
"#     name" if it appears at the start of a line.  Using () appears
to suppress this, while confusing Coccinelle.
2020-02-06 14:54:04 -05:00
Nick Mathewson
6104c407e0 maps: do not include _ as part of expected prefix in macros.
Doing this makes our macro usage a little clear IMO, and also avoids
having to use an unadorned "new" within a macro.  (Clang-format
seems to think that "new" means we're doing C++, and so it generates
some output that checkSpace.pl doesn't care for.)
2020-02-06 14:54:04 -05:00
Nick Mathewson
87b71a692a Remove senseless CHECK_PRINTF()s from util_bug.c
These belong in util_bug.h (and they already are there).

Their presence made clang-format misindent these functions in a way
that checkSpace.pl dislikes.
2020-02-06 14:54:04 -05:00
Nick Mathewson
efa5020a9c log_test_helpers: remove semicolons from end of macros
We want our code to require semicolons after use of these macros,
so that our code formatters and/or analysis tools don't get
confused.
2020-02-06 14:54:04 -05:00
Nick Mathewson
60f01da78e Use smaller CPP error messages
Clang-format wants to split these messages across multiple lines,
which confuses the heck out of coccinelle.
2020-02-06 14:54:04 -05:00
Nick Mathewson
e9b663beaf onion_queue.c: use TAILQ_HEAD less confusingly.
When we use macro inline, clang-format likes to break it in the
middle, which makes checkSpace get confused.
2020-02-06 14:33:05 -05:00
Nick Mathewson
9feeb4cf97 prob_distr: use "clang-format off" to avoid wide lines for URLs 2020-02-06 14:33:05 -05:00
Nick Mathewson
98fdc3e41a Use a compile-time assert in control_events.h
(The original idiom here led clang-format to generating a too-wide line.)
2020-02-06 14:33:05 -05:00
Nick Mathewson
21f45197a2 Merge branch 'bug33103' 2020-02-06 08:28:25 -05:00
David Goulet
a8456d2a11 test: Missing hs_init() in some hs_control tests
Without the hs_init(), the caches are not initialized and the previous commit
was causing an assert due to the missing caches.

But regardless of that, those tests were not initializing but were calling
hs_free_all().

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-06 12:54:54 +02:00
David Goulet
9278a24729 hs-v3: Remove descriptor when we remove client authorization
When the ONION_CLIENT_AUTH_REMOVE command is given to tor, now also remove the
descriptor associated with the client authorization credentials.

Fixes #33148

Signed-off-by: David Goulet <dgoulet@torproject.org>
2020-02-06 12:54:54 +02:00
Nick Mathewson
56c54a66bd Fix a memory leak in handling GETINFO.
Fixes bug 33103; bugfix on 0.4.3.1-alpha.
2020-02-05 19:18:32 -05:00
Nick Mathewson
1a375c3b19 Merge branch 'trove_2020_002_035' into trove_2020_002_041
Resolved Conflicts:
	src/feature/dirparse/parsecommon.c
2020-02-05 12:06:24 -05:00
Nick Mathewson
f160212ee8 When parsing tokens, reject early on spurious keys. 2020-02-05 11:57:31 -05:00
Nick Mathewson
9e1085c924 When parsing, reject >1024-bit RSA private keys sooner.
Private-key validation is fairly expensive for long keys in openssl,
so we need to avoid it sooner.
2020-02-05 11:11:35 -05:00
George Kadianakis
2a5e641cfe Merge branch 'tor-github/pr/1704' 2020-02-04 13:09:41 +01:00
George Kadianakis
6c749bf38c Merge branch 'tor-github/pr/1700' 2020-02-04 13:06:00 +01:00