Russell Bryant [Thu, 29 Mar 2007 17:43:52 +0000 (17:43 +0000)]
When building a response to a subscription, the "from" must be the full Jabber
ID. This fixes some problems where jabber users are not able to add their
Asterisk account to their user list, since they are unable to get Asterisk
to approve their subscription. (issue #8210, reported by caspy, and verified
by bradtem)
Keep a global array of variables indicating whether certain conference rooms are in use. This ensures that two people going into a new dynamic conference when the 'e' option is set don't go into the same conference room. (issue #8835 reported by eliel)
Russell Bryant [Thu, 29 Mar 2007 17:17:41 +0000 (17:17 +0000)]
Merged revisions 59357 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2
........
r59357 | russell | 2007-03-29 12:14:33 -0500 (Thu, 29 Mar 2007) | 5 lines
If an error occurs when reading from an RTP socket, and the error code does not
indicate that we should try again, then return NULL instead of a "null frame".
This will prevent Asterisk from trying over and over again, and eventually
causing the system to crash. (issue #8285, john)
Russell Bryant [Thu, 29 Mar 2007 16:55:39 +0000 (16:55 +0000)]
When the IAX2 read callback gets called, return NULL instead of a "null frame".
This will cause Asterisk to hangup the call instead of keep trying whatever it
was doing. Under normal conditions, this function would *never* be called.
However, the author of this patch says an error will occur that will cause it
to get called every 100 thousand calls or so. When this does happen, it puts
the channel in a loop that eventually brings down the system. So, hangup up
the call is certainly a better alternative. (issue #8286, john)
Issue 9415 - No point to getting a diagnostic field if we aren't doing anything with the information.
(Plus, it tends to crash the Postgres ODBC driver.)
Russell Bryant [Tue, 27 Mar 2007 23:02:12 +0000 (23:02 +0000)]
Fix app_directory when ODBC_STORAGE is being used. The Makefile did not
properly ensure that this information got copied from what was selected
for app_voicemail. (issue #9224)
Steve Murphy [Tue, 27 Mar 2007 18:16:32 +0000 (18:16 +0000)]
via 9373 (duplicate context in AEL crashes asterisk), kpfleming pointed on asterisk-dev, that DECLINE in this case the proper thing to do. This change now has it doing the proper thing.
Russell Bryant [Tue, 27 Mar 2007 18:05:46 +0000 (18:05 +0000)]
Merged revisions 59258 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2
........
r59258 | russell | 2007-03-27 13:04:02 -0500 (Tue, 27 Mar 2007) | 4 lines
Fix the use of the "sourceaddress" option when "bindaddr" is set to 0.0.0.0
instead of having each interface explicitly listed.
(issue #7874, patch by stevens)
Russell Bryant [Tue, 27 Mar 2007 16:20:53 +0000 (16:20 +0000)]
Convert the RTPQOS function to just be additional parameter of the CHANNEL
function. This way, it will be possible for other RTP based channel drivers
to expose this information in the future.
Steve Murphy [Mon, 26 Mar 2007 21:41:32 +0000 (21:41 +0000)]
fix for 9373 (duplicate context in AEL crashes asterisk). I turned a duplicate context from a WARNING to an ERROR. Now you get a module load failure, and asterisk just exits. That's better than a crash, right\?
Russell Bryant [Mon, 26 Mar 2007 17:45:55 +0000 (17:45 +0000)]
The AUDIORTPQOS and VIDEORTPQOS variables are not fully functional in some
because they get set in sip_hangup. So, there are common situations where
the variables will not be available in the dialplan at all. So, this patch
provides an alternate method for getting to this information by introducing
AUDIORTPQOS and VIDEORTPQOS dialplan functions.
(issue #9370, patch by Corydon76, with some testing by blitzrage)
Nadi Sarrar [Mon, 26 Mar 2007 15:25:53 +0000 (15:25 +0000)]
* mISDN >= 1.2 provides a dsp pipeline for i.e. echo cancellation modules, make chan_misdn use it.
* add a check for linux/mISDNdsp.h to configure.ac and update the autogenerated files: 'configure', 'autoconfig.h.in'
(the 'configure' script was not in sync with the latest configure.ac, so the diff is a bit bigger than expected).
Joshua Colp [Wed, 21 Mar 2007 03:25:48 +0000 (03:25 +0000)]
Until we can do media level parsing for sendrecv/etc just use the first value found. This crept up when a phone was offered audio+video and returned an inactive video stream. chan_sip thought the phone said to put the person on hold but that was totally wrong. (issue #9319 reported by benbrown)
added method standard_dec for dialing out on groups, to avoid conflicts, which caused issues with some ISDN providers
........
r58850 | crichter | 2007-03-13 13:58:32 +0100 (Di, 13 Mär 2007) | 1 line
fixed the crypt_keys stuff
........
r59062 | crichter | 2007-03-20 10:18:06 +0100 (Di, 20 Mär 2007) | 1 line
avoid sending a disconnect when we already received one.
........
r59063 | crichter | 2007-03-20 10:23:22 +0100 (Di, 20 Mär 2007) | 1 line
Russell Bryant [Thu, 15 Mar 2007 22:25:12 +0000 (22:25 +0000)]
Merge changes from svn/asterisk/team/russell/LaTeX_docs.
* Convert most of the doc directory into a single LaTeX formatted document
so that we can generate a PDF, HTML, or other formats from this
information.
* Add a CLI command to dump the application documentation into LaTeX format
which will only be include if the configure script is run with
--enable-dev-mode.
* The PDF turned out to be close to 1 MB, so it is not included. However, you
can simply run "make asterisk.pdf" to generate it yourself. We may include
it in release tarballs or have automatically generated ones on the web site,
but that has yet to be decided.
Russell Bryant [Wed, 14 Mar 2007 19:18:08 +0000 (19:18 +0000)]
Some people like to put "limitonpeer" instead of "limitonpeers" in their
configuration. While we're at it, support "limitonpeerz" and
"limitonpeerssssss". (inspired by issue #9172)
Russell Bryant [Wed, 14 Mar 2007 16:33:01 +0000 (16:33 +0000)]
By default, don't attempt to do any CallerID handling at all with SLA because
it is known to not work properly in some situations. However, add an option to
enable it for those that would like to use it anyway.
The short story behind this is that to properly handle CallerID with SLA, we
need the ability to change the CallerID on an existing call, and we are not
ready to handle that.
Russell Bryant [Tue, 13 Mar 2007 23:19:51 +0000 (23:19 +0000)]
Ensure that the blinky lights show that the trunk stopped ringing when the
trunk hangs up before a station has answered it.
(issue #9234, reported by francesco_r)
Joshua Colp [Mon, 12 Mar 2007 01:21:12 +0000 (01:21 +0000)]
Allow RFC2833 compensation to compensate for even stupider implementations by queueing up the end frame at the start, not the actual end. (issue #8963 reported by AndrewZ)
Joshua Colp [Mon, 12 Mar 2007 00:51:16 +0000 (00:51 +0000)]
Add matchexterniplocally setting which only substitutes your externip/externhost setting if it matches the localnet setting. I know of at least two people who need opposite settings, so I made it an option! (issue #8821 reported by kokoskarokoska)
Russell Bryant [Sat, 10 Mar 2007 18:11:11 +0000 (18:11 +0000)]
Fix a few more places in chan_iax2 where the ast_frame used for receiving a
frame was not properly initialized.
- Interpolating a frame when the jitterbuffer is in use
- decrypting a frame when IAX2 encryption is on
- frames in an IAX2 trunk
Russell Bryant [Fri, 9 Mar 2007 23:59:10 +0000 (23:59 +0000)]
Merge some updates to the SLA documentation. I plan to keep working on this
to explain all of the expected behavior with call handling, configuration
details for specific phones, and other things. However, I got tired of doing
it in plain text, so I switched to using LaTeX. I have included the PDF version.
I haven't been able to get a nice looking plain text version out of it yet, but
I'm not terribly concerned since this is supposed to be more of the manual,
while the plain text sample configuration file is the reference.
If we are unable to lookup the host in a c line we have to abort, otherwise the previous data is gone and we will (potentially) have no data when all is said and done.
Russell Bryant [Thu, 8 Mar 2007 22:15:15 +0000 (22:15 +0000)]
Hang up the channel that put the call on hold in the event processing thread to
avoid a race condition. Also, if the station originated the call that it is
putting on hold, don't hang up the trunk if it was the only station on the call
and it is hanging up due to hold and not a normal hangup.
Russell Bryant [Thu, 8 Mar 2007 01:01:46 +0000 (01:01 +0000)]
If we receive ZT_EVENT_REMOVED, destroy the specified channel.
(issue #7256, tzafrir)
Also, update the configure script to make sure that we don't try to build
chan_zap if the installed version of zaptel does not include ZT_EVENT_REMOVED.
Russell Bryant [Wed, 7 Mar 2007 18:19:19 +0000 (18:19 +0000)]
(This bug was reported to me by Kinsey Moore)
Merged revisions 58242 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2
........
r58242 | russell | 2007-03-07 12:17:07 -0600 (Wed, 07 Mar 2007) | 7 lines
Fix a problem where the Asterisk channel name could be that of the wrong IAX2
user for a call. This is because the first step of choosing this name is to
look for an IAX2 peer that happens to have the same IP/port number that this
call is coming from and assuming that is it. However, this is not always
correct. So, I have made it change this name after authentication happens
since at that point, we have an exact match.
Russell Bryant [Wed, 7 Mar 2007 00:25:19 +0000 (00:25 +0000)]
Merged revisions 58164 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2
........
r58164 | russell | 2007-03-06 18:20:13 -0600 (Tue, 06 Mar 2007) | 4 lines
If the channels acquired using the manager Redirect action are not up, then
don't attempt to do anything with them. It could lead to weird behavior,
including crashes. (issue #8977)
Steve Murphy [Tue, 6 Mar 2007 23:10:14 +0000 (23:10 +0000)]
Merged revisions 58115 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2
........
r58115 | murf | 2007-03-06 15:52:52 -0700 (Tue, 06 Mar 2007) | 1 line
Fix for 9220: Eyebeam cannot renew subscriptions for presence info. Reason: re-SUBSCRIBE requests don't include Accept headers, which the rfc says are optional (to put it tersely), (it uses MAY), and luckily, the sip_pvt struct has the format info stored, so we simply leave it if the format is set, and the accept header null.
........
projects / thirdparty / asterisk.git / log
