kenmirrors/freeswitch
1
0
Fork 0
mirror of https://github.com/signalwire/freeswitch.git synced 2025-02-24 10:31:13 +00:00
Code Issues Projects Releases Wiki Activity
17,749 Commits 55 Branches 148 Tags
Commit Graph

2191 Commits

Author SHA1 Message Date
Brian West
87edbed6bb FS-535: be more careful and catch ipv6 edge case 2010-11-22 15:32:23 -06:00
Brian West
22047f9337 FS-2862: fix seg more 2010-11-22 15:24:22 -06:00
Brian West
cf398e1a44 FS-535: tested but please test MORE. 2010-11-22 14:59:47 -06:00
Brian West
90c156243c FS-2862: fix seg 2010-11-22 10:10:26 -06:00
Brian West
6dfd39c5f3 log ip too 2010-11-21 12:01:48 -06:00
Anthony Minessale
5973679e35 fix passing of callee-id 2010-11-19 14:47:45 -06:00
Anthony Minessale
6c4f49a888 apparently some sip device vendors did not read the RFC (who knew?) adding verbose_sdp=true var to add needless a= lines for standard iana codecs that explicitly do not require them 2010-11-19 13:46:14 -06:00
Mathieu Rene
d2fd5ba0aa fix regression 2010-11-19 12:07:19 -05:00
Mathieu Rene
de417e99f0 Implement switch_queue_pop_timeout() and refactor sofia_profile_worker_thread_run() to use it so it doesn't wake up too often 2010-11-18 21:00:35 -05:00
Anthony Minessale
04e57577b3 FS-2801 2010-11-15 11:22:34 -06:00
Anthony Minessale
b160942bc0 test 2010-11-12 12:03:40 -06:00
Anthony Minessale
f079f898cb declinatio mortuus obfirmo! 2010-11-11 10:49:02 -06:00
Anthony Minessale
b278dd2379 add manual_rtp_bugs to profile and chan var and 3 new RTP bugs SEND_LINEAR_TIMESTAMPS|START_SEQ_AT_ZERO|NEVER_SEND_MARKER 
RTP_BUG_SEND_LINEAR_TIMESTAMPS = (1 << 3),

	  Our friends at Sonus get real mad when the timestamps are not in perfect sequence even during periods of silence.
	  With this flag, we will only increment the timestamp when write packets even if they are eons apart.

	RTP_BUG_START_SEQ_AT_ZERO = (1 << 4),

	  Our friends at Sonus also get real mad if the sequence number does not start at 0.
	  Typically, we set this to a random starting value for your saftey.
	  This is a security risk you take upon yourself when you enable this flag.

	RTP_BUG_NEVER_SEND_MARKER = (1 << 5),

	  Our friends at Sonus are on a roll, They also get easily dumbfounded by marker bits.
	  This flag will never send any. Sheesh....
 2010-11-10 16:58:36 -06:00
Anthony Minessale
10119e9e88 FS-2824 2010-11-08 10:13:35 -06:00
Michael Jerris
40ac860aaa fix missing name and potential segfault in gateway status 2010-11-05 20:24:31 -04:00
Anthony Minessale
d205755ae0 typo 2010-11-04 14:38:43 -05:00
Anthony Minessale
2043d5a671 fix display of timeout 2010-11-03 12:22:01 -05:00
Anthony Minessale
e5b891eed1 add separate reg timeout from retry sec 2010-11-03 10:58:32 -05:00
Michael Jerris
fc4d290cb7 add support for NDLB-force-rport=safe param that does force-rport behavior only on endpoints we know are safe to do so on. This is a dirty hack to try to work with certain endpoints behind sonicwall which does not use the same port when it does nat, when the devices do not support rport, while not breaking devices that acutally use different ports that force-rport will break 2010-11-03 11:53:39 -04:00
Anthony Minessale
df52cc0952 fix registration overlap timeout snafu 2010-11-02 17:58:24 -05:00
Anthony Minessale
97d80d924a same as last one in another place 2010-11-02 11:50:04 -05:00
Anthony Minessale
35676e7e04 parse static route in sip uri in notify by event 2010-11-01 18:47:09 -05:00
Anthony Minessale
1970ec1d81 FS-2810 2010-11-01 10:03:10 -05:00
Anthony Minessale
6737ae52d9 FS-2801 2010-10-28 15:34:56 -05:00
Anthony Minessale
19325c4369 fix race in codec failure condition, then fix bug in sdp parsing (likely a regression from recent codec changes) to never have the problem in the first place so you are double-protected 2010-10-27 16:37:35 -05:00
Anthony Minessale
8bb4d0d413 add the stuff drk_ keeps begging for 2010-10-27 14:09:38 -05:00
Anthony Minessale
99ada6bdba add events 2010-10-27 12:00:30 -05:00
Anthony Minessale
91230e01f9 used || instead of 'or' in sql stmt 2010-10-27 11:46:39 -05:00
Anthony Minessale
6ae933a2e9 FS-2800 2010-10-27 11:10:21 -05:00
Anthony Minessale
e10bc0a965 allow {dtmf_type=none} to work in oubound dial strings 2010-10-26 15:43:14 -05:00
Anthony Minessale
59d3b84def profile param ignore-183nosdp, chanvar sip_ignore_183nosdp FS-1978 2010-10-25 11:52:43 -05:00
Anthony Minessale
7d7d66342f reverse perspective of dialog events on non-registered entities 2010-10-22 15:14:14 -05:00
Anthony Minessale
83c8c1254b this should have been moved into agressive nat detection not commented 2010-10-22 11:22:17 -05:00
Anthony Minessale
fd1736b38f 1) Add force-publish-expires to set custom presence update expires delta (-1 means endless) 
2) Check how many users are registered when receiving a PUBLISH AND Multiple Registrations is enabled:
	if there is more than just 1 AND you are sending a offline message: skip publishing it to everyone
	to prevent clients from thinking themselves has gone offline.
 2010-10-21 12:55:07 -05:00
Anthony Minessale
23180c18e1 changes from devlist 2010-10-20 10:17:57 -05:00
Anthony Minessale
ad74326b3f fix presence update from dp to apply to pidf presence 2010-10-19 13:58:30 -05:00
Anthony Minessale
36851a90b7 dont update display to ring when call is hungup in pidf presence 2010-10-19 13:33:17 -05:00
Anthony Minessale
4e0399d05f adjust sql stmts in presence to allow even non-registered entities to be tracked 2010-10-18 15:30:29 -05:00
Anthony Minessale
6772c795e0 FS-2787 2010-10-15 17:59:00 -05:00
Anthony Minessale
afaf1fac05 ilbc tweak 2010-10-15 17:53:38 -05:00
Anthony Minessale
dfa78985b4 Change codec behaviour 
channel_variable: sdp_m_per_ptime
Adds a new m= line for each distinct ptime in codec list.

When this variable is not set:
	When mixing codecs with various ptime in a codec list, they will now be allowed to co-exist in the sdp but it will send no ptime attr.
		This means the ptime preferences on the offer will be ignored when mixing codecs with various ptimes.
	When receiving a codec list with no ptime attr, the ptime will be chosen from local preference instead of assuming 20ms
		This means if offer contains PCMU with not ptime and FS has PCMU@40i

Dynamic payloads will now start at 98 and increment per additional dynamic codec per call.
	So now you can add CELT@32000h,CELT@48000h and each one will be auto-assigned a dynamic paylaod type.
 2010-10-13 19:28:20 -05:00
Anthony Minessale
54dcb64a4d typo 2010-10-13 16:21:53 -05:00
Anthony Minessale
93c2ed941a silent recovery 2010-10-13 16:17:43 -05:00
Brian West
42f534fa6a FS-2773 2010-10-12 11:41:28 -05:00
Anthony Minessale
14361c0907 FS-620 2010-10-12 09:32:15 -05:00
Anthony Minessale
f13fa0c1a5 FS-2763 2010-10-08 15:38:01 -05:00
Anthony Minessale
9537197b4d add some goodies 2010-10-07 18:30:47 -05:00
Anthony Minessale
3cead32f13 presence again 2010-10-07 15:52:20 -05:00
Anthony Minessale
430fb98050 update watchdog code 2010-10-06 16:05:13 -05:00
Anthony Minessale
6c4cb07bac fix memory leak caused by regression from FS-2747 2010-10-06 16:05:13 -05:00