Commit Graph

10483 Commits

Author SHA1 Message Date
Nick Mathewson
8f3e3279c1 Try to diagnose bug 12184
Check for consistency between the queued destroy cells and the marked
circuit IDs.  Check for consistency in the count of queued destroy
cells in several ways.  Check to see whether any of the marked circuit
IDs have somehow been marked longer than the channel has existed.
2014-06-14 11:00:44 -04:00
Nick Mathewson
95d47a7481 Merge remote-tracking branch 'public/bug12169_relay_check' 2014-06-04 15:30:43 -04:00
Nick Mathewson
0073c5b517 Merge remote-tracking branch 'andrea/bug10616' 2014-06-04 15:12:45 -04:00
Nick Mathewson
e74c360156 Merge remote-tracking branch 'public/bug12195' 2014-06-04 12:16:03 -04:00
Nick Mathewson
84ed086d48 Fix ancient code that only checked circ_id, not circ_id and chan
This code mis-handled the case where a circuit got the same circuit
ID in both directions.  I found three instances of it in the
codebase, by grepping for [pn]_circ_id.

Because of the issue in command_process_relay_cell(), this would
have made roughly one circuit in a million completely nonfunctional.

Fixes bug 12195.
2014-06-03 18:19:08 -04:00
Andrea Shepard
2de0281879 Squelch spurious LD_BUG message in connection_ap_handshake_socks_reply() 2014-06-03 14:37:49 -07:00
Nick Mathewson
dd0745d066 Don't try to fetch bridge descriptors when DisableNetwork is set
Patch from Roger; changes file by me.

Fixes 10405; bugfix on 0.2.3.9-alpha, where DisableNetwork was
introduced.
2014-06-02 02:17:28 -04:00
Nick Mathewson
723894f114 Merge remote-tracking branch 'public/bug12170_024_v2' 2014-06-02 00:47:51 -04:00
Nick Mathewson
ad8977e394 Avoid needless router_dir_info_has_changed from router_set_status
On some profiles of Andrea's from #11332, I found that a great deal
of time can still be attributed to functions called from
update_router_have_minimum_dir_info().  This is making our
digestmap, tor_memeq, and siphash functions take a much bigger
portion of runtime than they really should.

If we're calling update_router_have_minimum_dir_info() too often,
that's because we're calling router_dir_info_changed() too often.
And it looks like most of the callers of router_dir_info_changed()
are coming as tail-calls from router_set_status() as invoked by
channel_do_open_actions().

But we don't need to call router_dir_info_changed() so much!  (I'm
not quite sure we need to call it from here at all, but...) Surely
we don't need to call it from router_set_status when the router's
status has not actually changed.

This patch makes us call router_dir_info_changed() from
router_set_status only when we are changing the router's status.

Fix for bug 12170.  This is leftover from our fix back in 273ee3e81
in 0.1.2.1-alpha, where we started caching the value of
update_router_have_minimum_dir_info().
2014-06-02 00:45:15 -04:00
Nick Mathewson
d9564d5285 Use uint32 !=, not tor_memneq, for relay cell integrity checking
tor_memeq has started to show up on profiles, and this is one of the
most frequent callers of that function, appearing as it does on every
cell handled for entry or exit.

59f9097d5c introduced tor_memneq here;
it went into Tor 0.2.1.31.  Fixes part of 12169.
2014-06-01 14:05:10 -04:00
Nick Mathewson
14842de9a7 sandbox: Allow DirPortFrontPage unconditionally if it's set
fixes 12114; bug not in any release.

Improves fix for 12028
2014-05-27 19:21:11 -04:00
Nick Mathewson
824bebd409 sandbox: Correct fix for hs part of 12064
Bugfix on cfd0ee514c279bc6c7b; bug not in any released version of tor
2014-05-23 11:46:44 -04:00
Nick Mathewson
5de91d118d Merge branch 'bug11965_v2' 2014-05-23 11:23:00 -04:00
Nick Mathewson
802c063148 Postpone fetches based on should_delay_dir_fetch(), not DisableNetwork
Without this fix, when running with bridges, we would try fetching
directory info far too early, and have up to a 60 second delay if we
started with bridge descriptors available.

Fixes bug 11965. Fix on 0.2.3.6-alpha, arma thinks.
2014-05-23 11:22:35 -04:00
Nick Mathewson
cfd0ee514c sandbox: allow reading of hidden service configuration files.
fixes part of 12064
2014-05-22 20:39:10 -04:00
Nick Mathewson
85f49abfbe sandbox: refactor string-based option-unchanged tests to use a macro
There was too much code duplication in doing it the old way, and I
nearly made a copy-and-paste error in the last commit.
2014-05-22 20:00:22 -04:00
Nick Mathewson
ffc1fde01f sandbox: allow access to cookie files, approved-routers
fixes part of 12064
2014-05-22 19:56:56 -04:00
Michael Wolf
387f294d40 sandbox: allow access to various stats/*-stats files
Fix for 12064 part 1
2014-05-22 19:48:24 -04:00
Nick Mathewson
1a73e17801 Merge remote-tracking branch 'andrea/bug11476' 2014-05-22 16:27:29 -04:00
Andrea Shepard
170e0df741 Eliminate #ifdef ENABLE_MEMPOOLS in packed_cell_new/free() 2014-05-21 10:53:25 -07:00
Nick Mathewson
2609b939d6 fix a wide line 2014-05-20 15:22:27 -04:00
Nick Mathewson
c21377e7bc sandbox: support logfile rotation
Fixes bug 12032; bugfix on 0.2.5.1-alpha
2014-05-20 15:21:48 -04:00
Nick Mathewson
268a117cdf sandbox: tolerate reloading with DirPortFrontPage set
Also, don't tolerate changing DirPortFrontPage.

Fixes bug 12028; bugfix on 0.2.5.1-alpha.
2014-05-20 14:58:28 -04:00
Nick Mathewson
465982012c sandbox: Disallow options which would make us call exec()
None of the things we might exec() can possibly run under the
sanbox, so rather than crash later, we have to refuse to accept the
configuration nice and early.

The longer-term solution is to have an exec() helper, but wow is
that risky.

fixes 12043; bugfix on 0.2.5.1-alpha
2014-05-20 12:21:31 -04:00
Nick Mathewson
f87071f49e sandbox: Permit access to stats/dirreq-stats
This prevents a crash when rotating logs with dirreq-stats enabled

fixes 12035; bugfix on 0.2.5.1-alpha.
2014-05-20 12:06:08 -04:00
Nick Mathewson
0b2b5b7606 Oops; permit rename with the correct filename 2014-05-20 12:03:27 -04:00
Nick Mathewson
ace9063fb4 Fix a sentence that I never 2014-05-20 11:58:18 -04:00
Nick Mathewson
f6d3006363 Sandbox: allow access to stats/bridge-stats
Fix for 12041; bugfix on 0.2.5.1-alpha.
2014-05-20 11:57:29 -04:00
Roger Dingledine
767b18ea8e note a comment that nickm didn't finish 2014-05-17 00:02:41 -04:00
Nick Mathewson
2d21a8f4d6 Merge remote-tracking branch 'public/bug11469_024' 2014-05-15 13:35:08 -04:00
Nick Mathewson
1badef5cec Use DirPort for uploading descriptors.
When we converted the horrible set of options that previously
controlled "use ORPort or DirPort? Anonymously or Non-anonymouly?" to
a single 'indirection' argument, we missed
directory_post_to_dirservers.

The problematic code was introduced in 5cbeb6080, which went into
0.2.4.3-alpha.  This is a fix for bug 11469.
2014-05-14 21:49:57 -04:00
Nick Mathewson
9b4ac986cb Use tor_getpw{nam,uid} wrappers to fix bug 11946
When running with User set, we frequently try to look up our
information in the user database (e.g., /etc/passwd).  The seccomp2
sandbox setup doesn't let us open /etc/passwd, and probably
shouldn't.

To fix this, we have a pair of wrappers for getpwnam and getpwuid.
When a real call to getpwnam or getpwuid fails, they fall back to a
cached value, if the uid/gid matches.

(Granting access to /etc/passwd isn't possible with the way we
handle opening files through the sandbox.  It's not desirable either.)
2014-05-14 13:53:14 -04:00
Nick Mathewson
e12af2adb0 Add a pair of wrapper functions: tor_getpwnam() and tor_getpwuid()
We'll use these to deal with being unable to access the user DB
after we install the sandbox, to fix bug 11946.
2014-05-14 13:50:43 -04:00
Andrea Shepard
39d4e67be8 Add --disable-mempools configure option 2014-05-12 18:23:34 -07:00
dana koch
d6e6c63baf Quench clang's complaints with -Wshorten-64-to-32 when time_t is not long.
On OpenBSD 5.4, time_t is a 32-bit integer. These instances contain
implicit treatment of long and time_t as comparable types, so explicitly
cast to time_t.
2014-05-11 23:36:00 -04:00
Nick Mathewson
4eb3018f94 Move structures into (private) part of buffers.h so we can inspect them while testing 2014-05-08 12:40:40 -04:00
Nick Mathewson
5b861ae53f Merge remote-tracking branch 'public/bug11648' 2014-05-08 12:01:23 -04:00
Nick Mathewson
411c622906 Merge commit 'bb9b4c37f8e7f5cf78918f382e90d8b11ff42551' into maint-0.2.4 2014-05-07 23:11:32 -04:00
Nick Mathewson
683b80bf81 Merge remote-tracking branch 'public/bug11737_diagnostic' 2014-05-07 22:52:44 -04:00
Nick Mathewson
0de2625675 Merge remote-tracking branch 'public/bug8387_diagnostic' 2014-05-07 22:15:24 -04:00
Nick Mathewson
48b9c6fcc6 Better log message for 8387 diagnostic 2014-05-07 22:13:29 -04:00
Nick Mathewson
6d39c8d156 Always finalize a zlib stream of server descriptors.
Possible fix for bug 11648.
2014-05-07 10:23:08 -04:00
Nick Mathewson
499e77663e Basic tests for get_unique_circ_id_by_chan. 2014-05-07 02:57:50 -04:00
Nick Mathewson
4a740451ac Merge remote-tracking branch 'public/bug11750' 2014-05-06 20:44:41 -04:00
Nick Mathewson
5cea500ce7 Merge branch 'bug11743_option_b' 2014-05-06 20:40:40 -04:00
Nick Mathewson
2fa601c797 Future-proof "id" lines against proposal 220. 2014-05-06 17:10:59 -04:00
Nick Mathewson
f077bb55ab fix a copy-paste comment mistake 2014-05-06 17:09:16 -04:00
Nick Mathewson
c7549cb4cd Merge remote-tracking branch 'karsten/bug11742' 2014-05-06 13:54:37 -04:00
Nick Mathewson
a06044a485 Check HT_REP_IS_BAD_() when giving a bug-7164 warning.
This may let us know if we're hitting 7164 because of an
hte_hash-corruption situation proposed by "cypherpunks" in bug
11737.
2014-05-06 13:03:24 -04:00
Nick Mathewson
0ad607d604 Faster chan_circid_entry_hash implementation
Since this is critical-path, let's tune the value we pass to
csiphash a little so it fits into one whole round.
2014-05-06 12:27:18 -04:00