Christian Grothoff [Thu, 8 Oct 2015 17:19:24 +0000 (17:19 +0000)]
-NULL builder on disconnect
Christian Grothoff [Thu, 8 Oct 2015 17:09:42 +0000 (17:09 +0000)]
redefining core message timeout semantics and core-core-API messaging to address (hypothetical?) stalls which may explain #3863. This kind-of breaks the API in that the timeout now has a different semantic -- it's an advisory as to when the app would like the message transmitted, the transmission will no longer actually time out. However, that was the documented semantics before, just not the actual behavior. Most applications didn't rely on it, and tests still pass, so I didn't break too much...
Christian Grothoff [Thu, 8 Oct 2015 17:01:28 +0000 (17:01 +0000)]
-signed cmp, otherwise -1 will not trigger
Christian Grothoff [Thu, 8 Oct 2015 16:21:58 +0000 (16:21 +0000)]
use constant for overdue warnings
Christian Grothoff [Thu, 8 Oct 2015 16:20:50 +0000 (16:20 +0000)]
-fix indent
Christian Grothoff [Thu, 8 Oct 2015 16:20:42 +0000 (16:20 +0000)]
add constant for the delay after which we log warnings
Christian Grothoff [Thu, 8 Oct 2015 16:13:08 +0000 (16:13 +0000)]
improve doxygen
Christian Grothoff [Thu, 8 Oct 2015 16:12:48 +0000 (16:12 +0000)]
-make time API more intutitive
Christian Grothoff [Thu, 8 Oct 2015 12:18:11 +0000 (12:18 +0000)]
-ignore
Christian Grothoff [Thu, 8 Oct 2015 12:12:25 +0000 (12:12 +0000)]
-fix
Christian Grothoff [Thu, 8 Oct 2015 11:54:50 +0000 (11:54 +0000)]
-make code nicer
Christian Grothoff [Thu, 8 Oct 2015 11:51:39 +0000 (11:51 +0000)]
-fix ptr ar
Christian Grothoff [Thu, 8 Oct 2015 11:44:13 +0000 (11:44 +0000)]
-fix FTBFS
Christian Grothoff [Thu, 8 Oct 2015 11:37:17 +0000 (11:37 +0000)]
-fix npe
Christian Grothoff [Thu, 8 Oct 2015 11:30:19 +0000 (11:30 +0000)]
add logic for strata compression
Christian Grothoff [Thu, 8 Oct 2015 11:28:27 +0000 (11:28 +0000)]
-fix template
Christian Grothoff [Thu, 8 Oct 2015 11:28:07 +0000 (11:28 +0000)]
export compression/decompression functions
Florian Dold [Wed, 7 Oct 2015 22:01:03 +0000 (22:01 +0000)]
statistics in set profiler, temp. fix for salt
Christian Grothoff [Wed, 7 Oct 2015 20:31:35 +0000 (20:31 +0000)]
add capability to log per-service resource consumption via ARM
Florian Dold [Wed, 7 Oct 2015 18:05:23 +0000 (18:05 +0000)]
set statistics
David Barksdale [Wed, 7 Oct 2015 17:02:23 +0000 (17:02 +0000)]
Fix typo in dv
David Barksdale [Wed, 7 Oct 2015 16:36:35 +0000 (16:36 +0000)]
Fix make -j
Christian Grothoff [Wed, 7 Oct 2015 14:15:38 +0000 (14:15 +0000)]
determine network scope for ATS even if we do not yet have a session and only have an address
Christian Grothoff [Wed, 7 Oct 2015 12:55:53 +0000 (12:55 +0000)]
-more meaningful log msg
Christian Grothoff [Wed, 7 Oct 2015 12:55:45 +0000 (12:55 +0000)]
-newline, include error code
Christian Grothoff [Wed, 7 Oct 2015 12:39:54 +0000 (12:39 +0000)]
-no printf
Christian Grothoff [Wed, 7 Oct 2015 11:54:46 +0000 (11:54 +0000)]
add additional diagnostics to help localize #3971 invariant violation
Christian Grothoff [Wed, 7 Oct 2015 11:54:25 +0000 (11:54 +0000)]
-typo
Christian Grothoff [Wed, 7 Oct 2015 11:49:37 +0000 (11:49 +0000)]
-simplify
Christian Grothoff [Wed, 7 Oct 2015 11:48:56 +0000 (11:48 +0000)]
-ensure result is +- 1
Christian Grothoff [Wed, 7 Oct 2015 11:31:37 +0000 (11:31 +0000)]
-fix testcase to handle SUCCESS status reply, only recv UDP after STUN Request is out, etc.
Christian Grothoff [Wed, 7 Oct 2015 10:11:29 +0000 (10:11 +0000)]
enable clean up of NAT STUN state, prevents test_plugin_udp from segfaulting
Christian Grothoff [Wed, 7 Oct 2015 09:51:46 +0000 (09:51 +0000)]
-removing spectacular dead code
Christian Grothoff [Wed, 7 Oct 2015 09:51:33 +0000 (09:51 +0000)]
stylistic fixes
Christian Grothoff [Wed, 7 Oct 2015 09:20:44 +0000 (09:20 +0000)]
-do not leave PPT timeout tasks behind
Christian Grothoff [Wed, 7 Oct 2015 08:51:00 +0000 (08:51 +0000)]
-fix NPE
Christian Grothoff [Wed, 7 Oct 2015 08:25:40 +0000 (08:25 +0000)]
major bugfix: is_ready must only be set to GNUNET_NO if we actually did transmit something to transport, otherwise we cannot expect to get a SEND_OK back
Christian Grothoff [Wed, 7 Oct 2015 08:24:43 +0000 (08:24 +0000)]
-adding more logging to diagnose transport transmission delays
Bart Polot [Tue, 6 Oct 2015 19:13:15 +0000 (19:13 +0000)]
- ignores
Bart Polot [Tue, 6 Oct 2015 19:09:30 +0000 (19:09 +0000)]
- fix multiple gnunet-nat crashes
Bart Polot [Tue, 6 Oct 2015 19:09:29 +0000 (19:09 +0000)]
- fix use of uninitialized memory
Bart Polot [Tue, 6 Oct 2015 17:50:47 +0000 (17:50 +0000)]
- fix script to work with barebone versions of sh
Florian Dold [Tue, 6 Oct 2015 16:17:08 +0000 (16:17 +0000)]
fix contestation handling / blacklisting
Florian Dold [Tue, 6 Oct 2015 15:22:02 +0000 (15:22 +0000)]
towards handling byz. faults correctly
- blacklisting of peers in consensus
- restructuring
Bart Polot [Tue, 6 Oct 2015 14:14:15 +0000 (14:14 +0000)]
- ignore confirmation of connections that are being destroyed (tunnel shutting down, for instance)
Bart Polot [Tue, 6 Oct 2015 12:07:00 +0000 (12:07 +0000)]
- remove unused variable
Bart Polot [Tue, 6 Oct 2015 12:03:16 +0000 (12:03 +0000)]
- refactor test dropping, reduce test load
Bart Polot [Tue, 6 Oct 2015 12:03:14 +0000 (12:03 +0000)]
- Revamp traffic logging, small bugfixes, complete doxygen
Florian Dold [Tue, 6 Oct 2015 02:08:16 +0000 (02:08 +0000)]
consensus work in progress
- store referendum entries differently
- split referendum evaluation in specialized task functions
Florian Dold [Mon, 5 Oct 2015 23:16:17 +0000 (23:16 +0000)]
include element type in hash
Florian Dold [Mon, 5 Oct 2015 22:27:46 +0000 (22:27 +0000)]
consensus
- fix problem with evil peers and loopback set operations
- logging
- fix round numbering
Florian Dold [Mon, 5 Oct 2015 21:26:56 +0000 (21:26 +0000)]
work on consensus and set
- evil peers for consensus
- various fixes for consensus and set
Christian Grothoff [Mon, 5 Oct 2015 16:35:44 +0000 (16:35 +0000)]
adding DEBUG logic for #3863
Christian Grothoff [Mon, 5 Oct 2015 15:28:21 +0000 (15:28 +0000)]
fix #3986
Florian Dold [Sat, 3 Oct 2015 22:36:37 +0000 (22:36 +0000)]
- another fix to generation handling and lazy copying
Christian Grothoff [Sat, 3 Oct 2015 20:14:05 +0000 (20:14 +0000)]
-improving comments and coding style for libgnunethello library
Christian Grothoff [Sat, 3 Oct 2015 20:12:09 +0000 (20:12 +0000)]
improving comments and coding style for libgnunethello library
Christian Grothoff [Sat, 3 Oct 2015 16:50:24 +0000 (16:50 +0000)]
-bump year
Christian Grothoff [Sat, 3 Oct 2015 16:38:55 +0000 (16:38 +0000)]
find network interfaces also using 'ip' command (#3911)
Florian Dold [Fri, 2 Oct 2015 16:26:47 +0000 (16:26 +0000)]
- consensus test conf
Bart Polot [Fri, 2 Oct 2015 15:07:44 +0000 (15:07 +0000)]
- don't show WARNINGs on duplicate traffic
Bart Polot [Fri, 2 Oct 2015 15:07:43 +0000 (15:07 +0000)]
- only break for cadet errors
Christian Grothoff [Fri, 2 Oct 2015 14:52:08 +0000 (14:52 +0000)]
-properly verify size, fixes #3991
Bart Polot [Fri, 2 Oct 2015 03:48:54 +0000 (03:48 +0000)]
- remove debug
Bart Polot [Fri, 2 Oct 2015 03:37:55 +0000 (03:37 +0000)]
- adjust logging
Bart Polot [Fri, 2 Oct 2015 03:37:54 +0000 (03:37 +0000)]
- find the correct stored key, not always the first one
Bart Polot [Fri, 2 Oct 2015 03:37:53 +0000 (03:37 +0000)]
Don't reset axolotl state on receipt of a pseudo-duplicate AX key exchange.
Pseudo duplicate means same handshake key but different ratchet key. This
can happen if Alice's first KX message is lost, she receives Bob's KX and
send a payload message herself, advancing her ratchet. If Bob doesn't send
a payload message fast enough, or it gets lost, Alice will retransmit her
KX message, this time with a different ratchet key. If Bob only checks the
ratchet key (the handshake keys are discarded) it could reset the Axolotl
status and re-use keys for messages that got lost/delayed.
Bart Polot [Fri, 2 Oct 2015 03:37:52 +0000 (03:37 +0000)]
Fix try_old_keys, in order to use it for data with the same Header key but out-of-order message number
Bart Polot [Fri, 2 Oct 2015 03:37:50 +0000 (03:37 +0000)]
- added hexdump-like function to write binary data to log
Bart Polot [Fri, 2 Oct 2015 03:37:49 +0000 (03:37 +0000)]
- escape line to avoid false matches
Bart Polot [Fri, 2 Oct 2015 03:37:48 +0000 (03:37 +0000)]
- typo in variable name
Bart Polot [Fri, 2 Oct 2015 03:37:47 +0000 (03:37 +0000)]
- use correct message number in the logs
Bart Polot [Fri, 2 Oct 2015 03:37:46 +0000 (03:37 +0000)]
- cancel all transmissions so core_api doesn't complain
Bart Polot [Fri, 2 Oct 2015 03:37:45 +0000 (03:37 +0000)]
- log shutdown
Bart Polot [Fri, 2 Oct 2015 03:37:44 +0000 (03:37 +0000)]
- removed false break
Bart Polot [Fri, 2 Oct 2015 03:37:42 +0000 (03:37 +0000)]
Fixed usage of known broken paths. If a connection was destroyed
while handling a neighbor disconnection, cadet tried to reconnect
to peers using paths containing the disconnecting peer.
Bart Polot [Fri, 2 Oct 2015 03:37:41 +0000 (03:37 +0000)]
- cancel transmit requests before destroying channels, warn API users if not
Bart Polot [Fri, 2 Oct 2015 03:37:40 +0000 (03:37 +0000)]
- don't connect twice on core_connect, as it may cause problems since connections hashmap is not yet created
Bart Polot [Fri, 2 Oct 2015 03:37:39 +0000 (03:37 +0000)]
- fix processing of messages with no payload (keepalives)
Bart Polot [Fri, 2 Oct 2015 03:37:38 +0000 (03:37 +0000)]
- fix log
Bart Polot [Fri, 2 Oct 2015 03:37:37 +0000 (03:37 +0000)]
- additional test
Bart Polot [Fri, 2 Oct 2015 03:37:36 +0000 (03:37 +0000)]
- never invalidate direct paths, show warning instead
Bart Polot [Fri, 2 Oct 2015 03:37:35 +0000 (03:37 +0000)]
- use better variable names
Bart Polot [Fri, 2 Oct 2015 03:37:34 +0000 (03:37 +0000)]
- don't try to connect to peers that are neighbors already
Bart Polot [Fri, 2 Oct 2015 03:37:33 +0000 (03:37 +0000)]
- connect if there are less connections than the minimum, not more
Bart Polot [Fri, 2 Oct 2015 03:37:32 +0000 (03:37 +0000)]
- don't leave freed memory on the peer's path list, signal error and remove anyway
Bart Polot [Fri, 2 Oct 2015 03:37:31 +0000 (03:37 +0000)]
- fix logging
Bart Polot [Fri, 2 Oct 2015 03:37:30 +0000 (03:37 +0000)]
- use proper connection states for a tunnel
Bart Polot [Fri, 2 Oct 2015 03:37:29 +0000 (03:37 +0000)]
- reduce log level
Bart Polot [Fri, 2 Oct 2015 03:37:28 +0000 (03:37 +0000)]
- compiler warning
Bart Polot [Fri, 2 Oct 2015 03:37:27 +0000 (03:37 +0000)]
- use correct hash type
Bart Polot [Fri, 2 Oct 2015 03:37:26 +0000 (03:37 +0000)]
- style, change rounds
Bart Polot [Fri, 2 Oct 2015 03:37:25 +0000 (03:37 +0000)]
- don't stop looking for connections as long as the tunnel exists
Bart Polot [Fri, 2 Oct 2015 03:37:24 +0000 (03:37 +0000)]
- indicate retransmitted packets on conn log
Bart Polot [Fri, 2 Oct 2015 03:37:23 +0000 (03:37 +0000)]
- unify logging format
Bart Polot [Fri, 2 Oct 2015 03:37:21 +0000 (03:37 +0000)]
- typo
Bart Polot [Fri, 2 Oct 2015 03:37:20 +0000 (03:37 +0000)]
- don't call core if there is no traffic
Carlo von lynX [Wed, 30 Sep 2015 20:15:37 +0000 (20:15 +0000)]
DESTDIR fix in Makefiles needed by distro packagers
Gabor X Toth [Wed, 30 Sep 2015 18:50:15 +0000 (18:50 +0000)]
free fix
Christian Grothoff [Tue, 29 Sep 2015 08:39:15 +0000 (08:39 +0000)]
stop crashing if sshd is not running on loopback