Commit Graph

15596 Commits

Author SHA1 Message Date
Anthony Minessale 804ef7709d change from sqlite hash to newly added one 2014-03-09 00:37:17 +05:00
Anthony Minessale 32cce8027e Add hashtable code from openzap/freetdm to FS core 2014-03-09 00:37:17 +05:00
Travis Cross 010740d664 Correct sizeof argument in mod_conference
This corrects a memset introduced in commit
bd4a0d8cbc.

The sizeof would have only returned the size of a pointer.
2014-03-08 01:56:18 +00:00
Michael Jerris e7c521b515 more subdir references 2014-03-07 20:50:28 -05:00
Michael Jerris 17d3b88a65 fix on newer automake 2014-03-07 20:31:20 -05:00
Travis Cross e6ec9b3c61 Add automake subdir-options for modules
The automake project is apparently changing behavior in their next
major version and warning everyone who relies on subdir-options to add
it explicitly.
2014-03-08 00:32:02 +00:00
Michael Jerris bcd9f49fbe move applications to use automake 2014-03-07 18:36:12 -05:00
Brian West 07272e864e FS-6321 --resolve
Copy URI params from Refer-To header into custom header in subsequent INVITE

sip_h_X-FS-Refer-Params is set to show up in the INVITE as X-FS-Refer-Params
2014-03-07 12:56:58 -06:00
Anthony Minessale f87ae15fc6 don't refuse invites on established sessions due to session limits 2014-03-07 23:46:26 +05:00
Anthony Minessale 8d2c6b354e According to https://code.google.com/p/webrtc/issues/detail?id=2768 ; The Chrome WebRTC engine reserves payload 98 and 99, IKR? So, to avoid taking a nasty spill down the stairs and subjecting ourselves to further school absences, we'll just start our payload space at 102 when making WebRTC calls....... 2014-03-07 20:24:16 +05:00
Chris Rienzo f9f36993e8 FS-6282 mod_rayo: fix memory leak in previous commit 2014-03-06 23:32:56 -05:00
Chris Rienzo 5138aede9b FS-6282 mod_rayo: allow outbound call JID to be assigned by client 2014-03-06 23:14:28 -05:00
Anthony Minessale b8e4a66dd2 another crypto regression 2014-03-07 08:34:39 +05:00
Anthony Minessale 0da8c6331d don't kick in nat mode on polycom tcp unless its not in the local network 2014-03-07 08:34:26 +05:00
Anthony Minessale 5aa955b5c9 also means forbidden on recovering calls 2014-03-07 07:11:32 +05:00
Anthony Minessale 87e0dda3d3 no var set on outbound meands forbidden and on inbound it means optional 2014-03-06 20:05:02 -06:00
Anthony Minessale fcef3ad4b1 FS-6319 --resolve 2014-03-06 19:37:11 -06:00
Anthony Minessale 46c5268e09 FS-6319 2014-03-07 06:35:02 +05:00
Anthony Minessale 5375d8b643 add on to last commit 2014-03-07 06:34:32 +05:00
Anthony Minessale bd4a0d8cbc add a way to tell mod_conference when the rate of the channel has changed due to a codec change so it can reset the resampler and codecs internally 2014-03-07 05:17:47 +05:00
Anthony Minessale a491df05f1 declinatio mortuus obfirmo! 2014-03-07 03:35:36 +05:00
Anthony Minessale 390e6713cc part of last patch 2014-03-07 02:59:09 +05:00
Anthony Minessale e9847afe22 feed all packets to jitterbuffer when enabled to absorb bursts and improve smoothing and delay protection 2014-03-07 02:48:56 +05:00
Brian West f7be96396b add missing flags after refactor 2014-03-06 10:49:43 -06:00
Travis Cross 7cde2adcb7 Fix minor edge case in switch_split_user_domain
If the input started with 'sip:sips:' it would have been incorrectly
parsed.
2014-03-06 06:03:27 +00:00
Anthony Minessale 656cb2ac03 add optional rtp_secure_media_suites variable clobbered by rtp_secure_media with mandatory|optional:<suites> 2014-03-06 08:17:22 +05:00
Anthony Minessale 1d733235a5 remove unused stuff from last commit 2014-03-06 07:41:39 +05:00
Anthony Minessale e5b291514c FS-5755
rtp_secure_media=mandatory
rtp_secure_media=optional
rtp_secure_media=mandatory:AES_CM_256_HMAC_SHA1_80,AES_CM_256_HMAC_SHA1_32
rtp_secure_media=optional:AES_CM_256_HMAC_SHA1_80
rtp_secure_media=forbidden

true implies mandatory
false implies forbidden
not set implies optional

rtp_secure_media_inbound or rtp_secure_media_outbound take precedence and are treated the same way based on leg direction
2014-03-06 07:34:47 +05:00
Anthony Minessale d3121d930e switch_false currently returns false on NULL 2014-03-05 17:36:35 -06:00
Anthony Minessale 7cb91467e0 FS-5814 --resolve 2014-03-06 00:02:40 +05:00
Travis Cross 74775d4397 Revert conference "tool" misfeature
This was added as part of a mass copyright header update in commit
6e7d5d089.  That's obviously not the right way to add features, so
we're reverting this.

If this feature is actually desired, it should be added in its own
commit, properly described in the commit message, and documented.

(The commit added a "tool" flag that could be applied to a conference
participant to mess with that person by disrupting his or her audio.)

This partially reverts commit 6e7d5d0897.

This feature earlier tried to sneak in under the guise of a whitespace
cleanup in commit a000749e70 which
Anthony reverted at commit a24f9aa8bc.

Let's not play these games.
2014-03-05 03:29:10 +00:00
Chris Rienzo 286d2aef29 FS-6304 mod_rayo- fix race condition on outbound calls 2014-03-04 22:12:43 -05:00
Giovanni Maruzzelli 4d8866a7cf gsmopen: added driver_usb_dongle directory, for building a working and stable 'option' modem serial driver for 2.6.32 kernels (eg: Proxmox, OpenVZ) 2014-03-05 02:06:03 +01:00
Anthony Minessale 6ae038add3 FS-5755 84c0680153 is wrong. It made optional enforce crypto. 2014-03-04 19:07:54 -06:00
William King fd38a255f8 FS-6167 --resolve 2014-03-04 13:41:34 -08:00
William King 3d461d7cde FS-1327 --resolve 2014-03-04 13:09:51 -08:00
Marc Olivier Chouinard 84c0680153 FS-5755 Fix regression if rtp_secure_media=false, it will force encryption. 2014-03-04 09:42:17 -05:00
Travis Cross 411a76020a Improve channel variable name to srtp_allow_idle_gaps
This was momentarily called force_send_silence_when_idle, but that was
non-obvious as you had to set that value to true to be able to not
send silence when idle.  This name describes the purpose much better.
2014-03-04 01:51:04 +00:00
Travis Cross 680bc46768 Avoid repeating ourselves in generating silence
We were handling the "send silence but not comfort noise" case in both
silence_stream_file_read and switch_generate_sln_silence.  This
changes the former to rely on the latter.
2014-03-04 00:16:43 +00:00
Travis Cross 5a7ea956b9 Add force_send_silence_when_idle channel variable
If set to true, this prevents us from overriding the value of
send_silence_when_idle.  When that is unset or set to zero and SRTP is
engaged, we typically override the value because many devices can't
handle gaps in the SRTP stream.

This variable is mostly for testing whether particular devices can
handle this behavior.  Use at your own risk.
2014-03-04 00:09:02 +00:00
Chris Rienzo e650939b25 FS-6296 --resolve mod_rayo: fixed crash on <prompt> bad request 2014-03-03 19:01:28 -05:00
Travis Cross 20da552564 Preserve value of send_silence_when_idle if possible
In commit 55d01d3def we set
send_silence_when_idle to -1 rather than 400 when SRTP is engaged.
But this left no way to enable white noise silence when desired.

When SRTP is engaged we can't simply not send RTP because it breaks
too many devices.  So we need to prevent send_silence_when_idle from
being unset or being set to zero.  This change allows it to be set to
other values so as to feed white noise rather than all zeros into the
codec.
2014-03-03 23:43:29 +00:00
Travis Cross 11ca1a2b2e Fix handling of send_silence_when_idle==0 in switch_ivr_sleep
When the channel variable send_silence_when_idle was set to zero,
switch_ivr_sleep was calling SWITCH_IVR_VERIFY_SILENCE_DIVISOR on it
anyway, causing it to be set to 400.  The only way to get the behavior
of not sending silence when idle was to unset the variable completely.

This corrects the behavior such that setting the value to zero has the
same effect as leaving it unset.
2014-03-03 23:21:58 +00:00
Anthony Minessale 6ef3f7bde7 add timeout <seconds> to mod_curl api call 2014-03-03 22:58:45 +05:00
Michael Jerris 07399e213f fix missing type definitions 2014-03-03 08:26:54 -05:00
Michael Jerris ae216daf02 fix warning abount comment inside comment 2014-03-03 08:26:54 -05:00
Travis Cross ecd6dfc612 Output newline after json output in mod_json_cdr
In UNIX, text files by definition end with a newline.
2014-03-03 01:13:39 +00:00
Travis Cross 95e4163ab7 Handle too-short write(3)s in mod_json_cdr
write(3) can write fewer bytes than was requested for any number of
reasons.  The correct behavior is to retry unless there is an error.

If there is an error, try to unlink the file; no sense in leaving
corrupted data laying around.
2014-03-03 01:12:15 +00:00
Travis Cross 75a00bd954 Fix memory leak in mod_json_cdr 2014-03-02 22:02:07 +00:00
Travis Cross 164d6a7bf5 Optimize switch_split_user_domain a bit
This avoids searching the string repeatedly with strchr.
2014-03-02 09:43:14 +00:00