oweals/gnunet.git
10 years ago-make sqlite3 hard requirement (#3341)
Christian Grothoff [Fri, 4 Apr 2014 15:02:07 +0000 (15:02 +0000)]
-make sqlite3 hard requirement (#3341)

10 years ago-fix #3355: disable warning
Christian Grothoff [Fri, 4 Apr 2014 14:55:35 +0000 (14:55 +0000)]
-fix #3355: disable warning

10 years agodefragmentation context not freed when session is deleted
Matthias Wachs [Thu, 3 Apr 2014 14:42:10 +0000 (14:42 +0000)]
defragmentation context not freed when session is deleted

10 years agoril may suggest 0, so this check is wrong
Matthias Wachs [Thu, 3 Apr 2014 11:05:52 +0000 (11:05 +0000)]
ril may suggest 0, so this check is wrong

10 years agofixing configuration for failing test
Matthias Wachs [Thu, 3 Apr 2014 09:44:21 +0000 (09:44 +0000)]
fixing configuration for failing test

10 years agodoxygen docu
Matthias Wachs [Thu, 3 Apr 2014 08:21:04 +0000 (08:21 +0000)]
doxygen docu

10 years ago- Remove trailing '/' for the directory path.
Sree Harsha Totakura [Tue, 1 Apr 2014 14:04:56 +0000 (14:04 +0000)]
- Remove trailing '/' for the directory path.

10 years ago- Log NSE histogram to file in addition to testbed-logger when HISTOGRAM_DIR
Sree Harsha Totakura [Tue, 1 Apr 2014 14:04:26 +0000 (14:04 +0000)]
- Log NSE histogram to file in addition to testbed-logger when HISTOGRAM_DIR
variable is present in configuration.

- Make logging to testbed-logger optional and only when testbed-logger service
is available.

- This is change is required to get NSE histograms on planetlab nodes.

10 years ago- remove ERROR log
Bart Polot [Tue, 1 Apr 2014 13:01:38 +0000 (13:01 +0000)]
- remove ERROR log

10 years ago-allow users direct core access
Christian Grothoff [Tue, 1 Apr 2014 12:11:33 +0000 (12:11 +0000)]
-allow users direct core access

10 years ago- fix uninitialized variable
Bart Polot [Tue, 1 Apr 2014 12:02:17 +0000 (12:02 +0000)]
- fix uninitialized variable

10 years ago-XDG environment variables cause more trouble than they are worth
Christian Grothoff [Tue, 1 Apr 2014 11:38:15 +0000 (11:38 +0000)]
-XDG environment variables cause more trouble than they are worth

10 years ago- log checksum error (for #3333)
Bart Polot [Tue, 1 Apr 2014 09:20:36 +0000 (09:20 +0000)]
- log checksum error (for #3333)

10 years ago- check handles on send_echo, fix #3349
Bart Polot [Tue, 1 Apr 2014 09:20:34 +0000 (09:20 +0000)]
- check handles on send_echo, fix #3349

10 years agominor changes for address conversion
Matthias Wachs [Tue, 1 Apr 2014 09:19:24 +0000 (09:19 +0000)]
minor changes for address conversion

10 years agoDo not account canceled messages as sent regarding flow control
Bart Polot [Tue, 1 Apr 2014 00:16:59 +0000 (00:16 +0000)]
Do not account canceled messages as sent regarding flow control

10 years agoCancel CORE transmit ready request when cancelling last transmittable message
Bart Polot [Tue, 1 Apr 2014 00:16:55 +0000 (00:16 +0000)]
Cancel CORE transmit ready request when cancelling last transmittable message

10 years ago- doc
Bart Polot [Tue, 1 Apr 2014 00:16:52 +0000 (00:16 +0000)]
- doc

10 years ago- shutdown dht last to allow tunnels to clean searches up
Bart Polot [Tue, 1 Apr 2014 00:16:49 +0000 (00:16 +0000)]
- shutdown dht last to allow tunnels to clean searches up

10 years ago- log
Bart Polot [Tue, 1 Apr 2014 00:16:46 +0000 (00:16 +0000)]
- log

10 years ago- remove explicit keepalive from queueing
Bart Polot [Tue, 1 Apr 2014 00:16:43 +0000 (00:16 +0000)]
- remove explicit keepalive from queueing

10 years ago- fix queue priority
Bart Polot [Tue, 1 Apr 2014 00:16:40 +0000 (00:16 +0000)]
- fix queue priority

10 years ago- use specific GMP_ function to stop dht
Bart Polot [Tue, 1 Apr 2014 00:16:37 +0000 (00:16 +0000)]
- use specific GMP_ function to stop dht

10 years ago- stop dht route search on tunnel destroy
Bart Polot [Tue, 1 Apr 2014 00:16:34 +0000 (00:16 +0000)]
- stop dht route search on tunnel destroy

10 years ago- peer info api
Bart Polot [Tue, 1 Apr 2014 00:16:31 +0000 (00:16 +0000)]
- peer info api

10 years ago- don't dereference NULL tunnels
Bart Polot [Tue, 1 Apr 2014 00:16:28 +0000 (00:16 +0000)]
- don't dereference NULL tunnels

10 years ago- implement client-side peer_id info request
Bart Polot [Tue, 1 Apr 2014 00:16:26 +0000 (00:16 +0000)]
- implement client-side peer_id info request

10 years ago- add peer option, move listen to free '-p'
Bart Polot [Tue, 1 Apr 2014 00:16:22 +0000 (00:16 +0000)]
- add peer option, move listen to free '-p'

10 years ago- use small mesh hashes for info api
Bart Polot [Tue, 1 Apr 2014 00:16:19 +0000 (00:16 +0000)]
- use small mesh hashes for info api

10 years ago- implement h2s
Bart Polot [Tue, 1 Apr 2014 00:16:15 +0000 (00:16 +0000)]
- implement h2s

10 years ago- move mesh hash to public api
Bart Polot [Tue, 1 Apr 2014 00:16:13 +0000 (00:16 +0000)]
- move mesh hash to public api

10 years ago- remove unnecessary break
Bart Polot [Tue, 1 Apr 2014 00:16:10 +0000 (00:16 +0000)]
- remove unnecessary break

10 years agoMissing extern for C++
Bruno Cabral [Mon, 31 Mar 2014 10:51:44 +0000 (10:51 +0000)]
Missing extern for C++

10 years agofix for connection counter
Matthias Wachs [Mon, 31 Mar 2014 09:38:08 +0000 (09:38 +0000)]
fix for connection counter

10 years agofix duplicate connect notifications
Matthias Wachs [Mon, 31 Mar 2014 09:35:55 +0000 (09:35 +0000)]
fix duplicate connect notifications

10 years agofixed check for session_ack handling
Matthias Wachs [Mon, 31 Mar 2014 09:03:01 +0000 (09:03 +0000)]
fixed check for session_ack handling

10 years agodebug message for assertion to figure out malformed address length
Matthias Wachs [Mon, 31 Mar 2014 09:02:34 +0000 (09:02 +0000)]
debug message for assertion to figure out malformed address length

10 years agofriend is a reserved work in C++. Rename to friend_id.
Bruno Cabral [Sun, 30 Mar 2014 13:42:47 +0000 (13:42 +0000)]
friend is a reserved work in C++. Rename to friend_id.

10 years agono need to be an error
Matthias Wachs [Sat, 29 Mar 2014 06:51:25 +0000 (06:51 +0000)]
no need to be an error

10 years agodebuggging message for "send ACK but no session found" issue
Matthias Wachs [Sat, 29 Mar 2014 06:44:04 +0000 (06:44 +0000)]
debuggging message for "send ACK but no session found" issue

10 years agodestroy session if connect fails
Matthias Wachs [Sat, 29 Mar 2014 05:45:46 +0000 (05:45 +0000)]
destroy session if connect fails

10 years agoFixed the overflow problems reported in issue 3350
Claudiu Olteanu [Fri, 28 Mar 2014 21:41:12 +0000 (21:41 +0000)]
Fixed the overflow problems reported in issue 3350

10 years agochanges to switch when connected
Matthias Wachs [Fri, 28 Mar 2014 16:41:42 +0000 (16:41 +0000)]
changes to switch when connected

10 years agocancel delayed disconnect task on free
Matthias Wachs [Fri, 28 Mar 2014 15:09:29 +0000 (15:09 +0000)]
cancel delayed disconnect task on free

10 years agochanges in ACK handling
Matthias Wachs [Fri, 28 Mar 2014 14:55:42 +0000 (14:55 +0000)]
changes in ACK handling

10 years agomake plugin fail if bind fails ...
Matthias Wachs [Fri, 28 Mar 2014 13:39:12 +0000 (13:39 +0000)]
make plugin fail if bind fails ...

10 years agore-add address on connect
Matthias Wachs [Fri, 28 Mar 2014 13:06:44 +0000 (13:06 +0000)]
re-add address on connect

10 years agofixing crash in unix
Matthias Wachs [Fri, 28 Mar 2014 12:42:39 +0000 (12:42 +0000)]
fixing crash in unix

10 years agoFramework for put/get/monitor
Supriti Singh [Fri, 28 Mar 2014 10:24:39 +0000 (10:24 +0000)]
Framework for put/get/monitor

10 years agoimproved printing
Matthias Wachs [Fri, 28 Mar 2014 08:41:34 +0000 (08:41 +0000)]
improved printing

10 years agofixing crash: cancel dead task 1004493731513019834!
Matthias Wachs [Fri, 28 Mar 2014 08:30:30 +0000 (08:30 +0000)]
fixing crash: cancel dead task 1004493731513019834!

10 years agoAPI header with disconnect call
Matthias Wachs [Thu, 27 Mar 2014 16:51:50 +0000 (16:51 +0000)]
API header with disconnect call

10 years agoadded functionality to use the CLI to disconnect peers
Matthias Wachs [Thu, 27 Mar 2014 16:51:32 +0000 (16:51 +0000)]
added functionality to use the CLI to disconnect peers
fixed DISCONNECT functionality

10 years agoWe do not need the address and session here
Matthias Wachs [Thu, 27 Mar 2014 15:03:17 +0000 (15:03 +0000)]
We do not need the address and session here

10 years agoimproved network lookup
Matthias Wachs [Thu, 27 Mar 2014 13:41:20 +0000 (13:41 +0000)]
improved network lookup

10 years agoone break to break it all:
Matthias Wachs [Thu, 27 Mar 2014 13:15:01 +0000 (13:15 +0000)]
one break to break it all:
disconnecting blacklist client was removed from first blacklist check only

10 years agoblacklist checks have to be cancelled when session is terminated, especially when...
Matthias Wachs [Thu, 27 Mar 2014 12:35:39 +0000 (12:35 +0000)]
blacklist checks have to be cancelled when session is terminated, especially when blacklist clients are active

10 years agoremove blacklist check in neighbours on CONNECT
Matthias Wachs [Thu, 27 Mar 2014 10:20:43 +0000 (10:20 +0000)]
remove blacklist check in neighbours on CONNECT

10 years agoreformat
Matthias Wachs [Thu, 27 Mar 2014 10:20:13 +0000 (10:20 +0000)]
reformat

10 years agosome testing code for debugging
Matthias Wachs [Thu, 27 Mar 2014 09:35:27 +0000 (09:35 +0000)]
some testing code for debugging

10 years agoAddress was not deleted if SEND_CONNECT in FAST RECONNECT failed
Matthias Wachs [Thu, 27 Mar 2014 09:05:19 +0000 (09:05 +0000)]
Address was not deleted if SEND_CONNECT in FAST RECONNECT failed
Therefore alternative inbound session was not suggested ...
This was a hard one...

10 years agoerror handling for failed CONNECT messages in CONNECT_SWITCHING
Matthias Wachs [Wed, 26 Mar 2014 16:29:55 +0000 (16:29 +0000)]
error handling for failed CONNECT messages in CONNECT_SWITCHING

10 years agowe do not need this address in neighbours: was already added when CONNECT was receive...
Matthias Wachs [Wed, 26 Mar 2014 15:04:29 +0000 (15:04 +0000)]
we do not need this address in neighbours: was already added when CONNECT was received in _transport.c before CONNECT was given to neighbours

10 years agoremove CONNECT_SWITCHING_BLACKLIST state
Matthias Wachs [Wed, 26 Mar 2014 14:59:36 +0000 (14:59 +0000)]
remove CONNECT_SWITCHING_BLACKLIST state

10 years agoremoving GNUNET_TRANSPORT_PS_CONNECT_RECV_BLACKLIST state
Matthias Wachs [Wed, 26 Mar 2014 14:16:41 +0000 (14:16 +0000)]
removing GNUNET_TRANSPORT_PS_CONNECT_RECV_BLACKLIST state

10 years agoremove debugging breaks
Matthias Wachs [Wed, 26 Mar 2014 13:07:03 +0000 (13:07 +0000)]
remove debugging breaks

10 years ago- Avoid dereferencing undefined pointer when the request status indicates an error.
Sree Harsha Totakura [Wed, 26 Mar 2014 11:08:09 +0000 (11:08 +0000)]
- Avoid dereferencing undefined pointer when the request status indicates an error.

10 years agoremove debug break
Matthias Wachs [Wed, 26 Mar 2014 11:02:57 +0000 (11:02 +0000)]
remove debug break

10 years agoremove state
Matthias Wachs [Wed, 26 Mar 2014 10:54:36 +0000 (10:54 +0000)]
remove state

10 years agoremoving blacklist state not required anymore
Matthias Wachs [Wed, 26 Mar 2014 10:54:04 +0000 (10:54 +0000)]
removing blacklist state not required anymore

10 years agofix compile warning
Matthias Wachs [Wed, 26 Mar 2014 09:40:07 +0000 (09:40 +0000)]
fix compile warning

10 years agoDo blacklist checks on CONNECT before giving CONNECT to neighbours.
Matthias Wachs [Tue, 25 Mar 2014 16:40:41 +0000 (16:40 +0000)]
Do blacklist checks on CONNECT before giving CONNECT to neighbours.
If peer is blacklisted we do not need to to anything, this simplifies the state machine:
If peer is blacklisted: CONNECT is not given to neighbours
If address is blacklisted: address is not given to ATS and will therefore not be suggested
So neighbour can use this information without additional blacklist checks

10 years agochecking incoming connection behaviour:
Matthias Wachs [Mon, 24 Mar 2014 16:43:41 +0000 (16:43 +0000)]
checking incoming connection behaviour:
- removing blacklist checks for incoming checks
- adjusting blacklist check on connect to report to ATS
- incoming blacklist check states did not set timeouts
- timeouts not set for PS_CONNECT_RECV_BLACKLIST_INBOUND in handle_connect

10 years agoFIX:
Matthias Wachs [Mon, 24 Mar 2014 15:26:46 +0000 (15:26 +0000)]
FIX:
In ATS fast reconnect the timeout was not set correctly if reconnect was successful and neighbour was disconnected in master task after 1 second

10 years agochecking incoming connections
Matthias Wachs [Mon, 24 Mar 2014 15:06:30 +0000 (15:06 +0000)]
checking incoming connections
major fixes:
- do not disconnect neighbor on CONNECT_SENT timeout, but:
destroy address and expect new address from ATS
- fixes related to mantis #3263 when peers reconnect:
when peers receive CONNECT in PS_DISCONNECT, disconnect was completed but
-- no address requested
-- ack timestamp not saved
-- ack state was not set to expect a SESSION_ACK message
-> reconnect failed!

minor change:
- adding a GST_ACK_State enumeration to make code easier understandable

10 years agoMerged update_successor and update_predecessor with finger table add
Supriti Singh [Mon, 24 Mar 2014 12:28:30 +0000 (12:28 +0000)]
Merged update_successor and update_predecessor with finger table add

10 years agotask was not marked as canceled when canceled
Matthias Wachs [Mon, 24 Mar 2014 10:22:07 +0000 (10:22 +0000)]
task was not marked as canceled when canceled
fixing mantis #0003345

10 years ago- fix
Bart Polot [Sat, 22 Mar 2014 22:54:51 +0000 (22:54 +0000)]
- fix

10 years ago- fix
Bart Polot [Sat, 22 Mar 2014 22:53:49 +0000 (22:53 +0000)]
- fix

10 years ago- optimizations for runtime
Bart Polot [Sat, 22 Mar 2014 22:53:48 +0000 (22:53 +0000)]
- optimizations for runtime

10 years ago- add more links
Bart Polot [Sat, 22 Mar 2014 22:46:04 +0000 (22:46 +0000)]
- add more links

10 years ago- start test if warmup "fails"
Bart Polot [Sat, 22 Mar 2014 22:46:02 +0000 (22:46 +0000)]
- start test if warmup "fails"

10 years ago- fix handling of duplicate incoming channel create with respect to queued retransmis...
Bart Polot [Sat, 22 Mar 2014 22:37:13 +0000 (22:37 +0000)]
- fix handling of duplicate incoming channel create with respect to queued retransmissions

10 years ago- profiler changes
Bart Polot [Sat, 22 Mar 2014 20:56:45 +0000 (20:56 +0000)]
- profiler changes

10 years ago- dont assert
Bart Polot [Sat, 22 Mar 2014 20:56:41 +0000 (20:56 +0000)]
- dont assert

10 years ago- debug
Bart Polot [Sat, 22 Mar 2014 19:27:16 +0000 (19:27 +0000)]
- debug

10 years ago- pass warmup parameter
Bart Polot [Sat, 22 Mar 2014 16:54:40 +0000 (16:54 +0000)]
- pass warmup parameter

10 years ago- fix
Bart Polot [Sat, 22 Mar 2014 16:54:38 +0000 (16:54 +0000)]
- fix

10 years ago- cancel SYNACK retry task when forced to SYNACK by an incoming SYN
Bart Polot [Sat, 22 Mar 2014 02:58:45 +0000 (02:58 +0000)]
- cancel SYNACK retry task when forced to SYNACK by an incoming SYN

10 years ago- don-t re-scheduler, destroy immediately (comes from scheduler)
Bart Polot [Sat, 22 Mar 2014 02:58:43 +0000 (02:58 +0000)]
- don-t re-scheduler, destroy immediately (comes from scheduler)

10 years ago- fix mask
Bart Polot [Sat, 22 Mar 2014 02:58:42 +0000 (02:58 +0000)]
- fix mask

10 years ago- set first channel number properly
Bart Polot [Sat, 22 Mar 2014 02:58:40 +0000 (02:58 +0000)]
- set first channel number properly

10 years ago- send ACK on duplicate channel_create
Bart Polot [Sat, 22 Mar 2014 02:58:39 +0000 (02:58 +0000)]
- send ACK on duplicate channel_create

10 years ago- fix peer to notify
Bart Polot [Sat, 22 Mar 2014 02:58:37 +0000 (02:58 +0000)]
- fix peer to notify

10 years ago- don't delay tests
Bart Polot [Sat, 22 Mar 2014 02:58:36 +0000 (02:58 +0000)]
- don't delay tests

10 years ago- invalidate instead of destroying, to avoid reusing right away (from stale DHT info...
Bart Polot [Sat, 22 Mar 2014 02:58:34 +0000 (02:58 +0000)]
- invalidate instead of destroying, to avoid reusing right away (from stale DHT info, for instance)

10 years ago- elimiate all paths
Bart Polot [Sat, 22 Mar 2014 02:58:33 +0000 (02:58 +0000)]
- elimiate all paths

10 years ago- eliminate broken paths
Bart Polot [Sat, 22 Mar 2014 02:58:32 +0000 (02:58 +0000)]
- eliminate broken paths

10 years ago- use proper type
Bart Polot [Sat, 22 Mar 2014 02:58:30 +0000 (02:58 +0000)]
- use proper type

10 years ago- fix infinite resend loop
Bart Polot [Sat, 22 Mar 2014 02:58:28 +0000 (02:58 +0000)]
- fix infinite resend loop