asterisk/asterisk.git
12 years agosend "423 Interval too small" Response to Subscribe with Expires less that min allowed
David Vossel [Wed, 21 Jul 2010 18:52:14 +0000 (18:52 +0000)]
send "423 Interval too small" Response to Subscribe with Expires less that min allowed

[RFC3265]3.1.6.1....
   The notifier MAY also check that the duration in the "Expires" header
   is not too small.  If and only if the expiration interval is greater
   than zero AND smaller than one hour AND less than a notifier-
   configured minimum, the notifier MAY return a "423 Interval too
   small" error which contains a "Min-Expires" header field.  The "Min-
   Expires" header field is described in SIP [1].

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278536 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix invalid test for rxisoffhook in FXO channels
Tzafrir Cohen [Wed, 21 Jul 2010 17:44:20 +0000 (17:44 +0000)]
Fix invalid test for rxisoffhook in FXO channels

This fixes some cases of no outgoing calls on FXO before an incoming call.

Remove an unnecessary testing of an "off-hook" bit from DAHDI for FXO
(KS/GS) channels.In some cases the bit would not be initialized properly
before the first inbound call and thus prevent an outgoing call.

If those tests are actually required by anybody, they should define
DAHDI_CHECK_HOOKSTATE in channels/sig_analog.c .

(closes issue #14577)
Reported by: jkroon
Patches:
      asterisk_chan_dahdi_hookstate_fix_trunk_new.diff uploaded by frawd (license 610)
Tested by: frawd

Review: https://reviewboard.asterisk.org/r/699/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278501 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoUse poll() instead of select() in res_timing_pthread to avoid stack corruption.
Russell Bryant [Wed, 21 Jul 2010 16:15:00 +0000 (16:15 +0000)]
Use poll() instead of select() in res_timing_pthread to avoid stack corruption.

This code did not properly check FD_SETSIZE to ensure that it did not try to
select() on fds that were too large.  Switching to poll() removes the limitation
on the maximum fd value.

(closes issue #15915)
Reported by: keiron

(closes issue #17187)
Reported by: Eddie Edwards

(closes issue #16494)
Reported by: Hubguru

(closes issue #15731)
Reported by: flop

(closes issue #12917)
Reported by: falves11

(closes issue #14920)
Reported by: vrban

(closes issue #17199)
Reported by: aleksey2000

(closes issue #15406)
Reported by: kowalma

(closes issue #17438)
Reported by: dcabot

(closes issue #17325)
Reported by: glwgoes

(closes issue #17118)
Reported by: erikje

possibly other issues, too ...

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278465 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoEnsure realtime conferences are treated the same as static conferences when trying...
Tilghman Lesher [Wed, 21 Jul 2010 15:56:05 +0000 (15:56 +0000)]
Ensure realtime conferences are treated the same as static conferences when trying to find an empty one.

Also, parse the useropts properly, when retrieving from realtime, and add them
to the existing flags.

(closes issue #17502)
 Reported by: kenji
 Patches:
       20100720__issue17502.diff.txt uploaded by tilghman (license 14)
 Tested by: kenji

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278463 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoProperly show the current page being transfered for 'fax show session'
Matthew Nicholson [Wed, 21 Jul 2010 15:54:29 +0000 (15:54 +0000)]
Properly show the current page being transfered for 'fax show session'

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278462 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoProperly set the port number for UDPTL media sessions.
Matthew Nicholson [Wed, 21 Jul 2010 15:51:24 +0000 (15:51 +0000)]
Properly set the port number for UDPTL media sessions.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278461 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDon't print failure status when the remote end hangs up, it may not be an actual...
Matthew Nicholson [Wed, 21 Jul 2010 13:03:01 +0000 (13:03 +0000)]
Don't print failure status when the remote end hangs up, it may not be an actual failure.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278426 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoUpdate documentation for 'comebacktoorigin' in featuers.conf.
Russell Bryant [Wed, 21 Jul 2010 13:02:46 +0000 (13:02 +0000)]
Update documentation for 'comebacktoorigin' in featuers.conf.

The documentation for this option did not match the code.  Fix that along with
some minor cleanups to the code along the way.  Document a slight change in
behavior (to something that was previously undocumented) in UPGRADE.txt.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278425 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoChange order so that it more closely matches the related SIP command.
Tilghman Lesher [Wed, 21 Jul 2010 06:45:06 +0000 (06:45 +0000)]
Change order so that it more closely matches the related SIP command.

(closes issue #17648)
 Reported by: GMLudo

Review: https://reviewboard.asterisk.org/r/789/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278393 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoinclude stat.h for everybody, needed for device2chan
Jeff Peeler [Wed, 21 Jul 2010 03:53:19 +0000 (03:53 +0000)]
include stat.h for everybody, needed for device2chan

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278361 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoSeparate queue_log arguments into separate fields, and allow the text file to be...
Tilghman Lesher [Tue, 20 Jul 2010 23:23:25 +0000 (23:23 +0000)]
Separate queue_log arguments into separate fields, and allow the text file to be used, even when realtime is used.

(closes issue #17082)
 Reported by: coolmig
 Patches:
       20100720__issue17082.diff.txt uploaded by tilghman (license 14)
 Tested by: coolmig

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278307 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 278261 via svnmerge from
Tilghman Lesher [Tue, 20 Jul 2010 22:40:19 +0000 (22:40 +0000)]
Merged revisions 278261 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r278261 | tilghman | 2010-07-20 17:23:13 -0500 (Tue, 20 Jul 2010) | 7 lines

  Delete IMAP messages in reverse order, to ensure reordering after each expunge does not cause deletion of the wrong message.

  (closes issue #16350)
   Reported by: noahisaac
   Patches:
         20100623__issue16350.diff.txt uploaded by tilghman (license 14)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278275 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoReference correct struct member for unlikely event PRI_EVENT_CONFIG_ERR.
Richard Mudgett [Tue, 20 Jul 2010 22:38:13 +0000 (22:38 +0000)]
Reference correct struct member for unlikely event PRI_EVENT_CONFIG_ERR.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278274 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 278167 via svnmerge from
Tilghman Lesher [Tue, 20 Jul 2010 22:26:23 +0000 (22:26 +0000)]
Merged revisions 278167 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r278167 | tilghman | 2010-07-20 15:59:06 -0500 (Tue, 20 Jul 2010) | 4 lines

  Do not queue up DTMF frames while a call is on hold.

  (Fixes ABE-2110)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278272 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agofixes sip CANCEL race condition
David Vossel [Tue, 20 Jul 2010 21:41:21 +0000 (21:41 +0000)]
fixes sip CANCEL race condition

If Asterisk sends a 4xx error and the other side sends a CANCEl
before receiving the 4xx and responding with the ACK, Asterisk
will process the CANCEL and send a 487 Request Terminated as
a new final response to the INVITE.  Since we are issuing a new
final response to the INVITE, the old one must be pretend_acked
else it will keep retransmitting.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278234 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoThis commit contains several changes to the way output channel variables are handled.
Matthew Nicholson [Tue, 20 Jul 2010 21:01:26 +0000 (21:01 +0000)]
This commit contains several changes to the way output channel variables are handled.

FAX output channel variables will now match the values reported by FAXOPT() and should be set in all failure and success cases.

This commit also contains a few modifications to the way FAXOPT() variables are populated in a few spots and fixes for some reference count leaks of the session details structure in some failure cases.

Also found and fixed more cases where FAXOPT(status) may not have gotten set.

FAX-214
FAX-203

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278168 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd load priority order, such that preload becomes unnecessary in most cases
Tilghman Lesher [Tue, 20 Jul 2010 19:35:02 +0000 (19:35 +0000)]
Add load priority order, such that preload becomes unnecessary in most cases

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278132 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd a package to install_prereq.
Russell Bryant [Tue, 20 Jul 2010 18:11:08 +0000 (18:11 +0000)]
Add a package to install_prereq.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278096 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoOnly call ast_channel_cc_params_init() if allocating a channel succeeds.
Russell Bryant [Tue, 20 Jul 2010 17:22:36 +0000 (17:22 +0000)]
Only call ast_channel_cc_params_init() if allocating a channel succeeds.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278051 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 278023 via svnmerge from
Tilghman Lesher [Tue, 20 Jul 2010 16:50:11 +0000 (16:50 +0000)]
Merged revisions 278023 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r278023 | tilghman | 2010-07-20 11:37:18 -0500 (Tue, 20 Jul 2010) | 7 lines

  Off-by-one error

  (closes issue #16506)
   Reported by: nik600
   Patches:
         20100629__issue16506.diff.txt uploaded by tilghman (license 14)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@278024 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277906 via svnmerge from
Jean Galarneau [Mon, 19 Jul 2010 21:07:08 +0000 (21:07 +0000)]
Merged revisions 277906 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277906 | jeang | 2010-07-19 15:16:36 -0500 (Mon, 19 Jul 2010) | 7 lines

  Avoid trying to pickup a parked extension before the park operation is completed.

  A crash could occur if the extension is picked up while the parking extension is
  being announced. Testing pu->notquiteyet while searching for a parked extension
  resolves this crash.

  (ABE-2418)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277945 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix port setting of external address in SIP.
Mark Michelson [Mon, 19 Jul 2010 17:16:23 +0000 (17:16 +0000)]
Fix port setting of external address in SIP.

There are two changes here:

1. Since the externip setting can now have a port attached
to it, calling it "externip" is misleading. The option is now
documented and parsed as "externaddr." This also extends to the
"matchexterniplocally" setting. It is now documented and parsed
as "matchexternaddrlocally." The old names for the options may
still be used, but they are no longer used in the sip.conf.sample
file.

2. If no port is set for the externaddr, and UDP is the transport
to be used, then we will set the port of the externaddr to that of
the udpbindaddr. This was how things worked prior to the IPv6 merge,
so this is a regression fix.

(closes issue #17665)
Reported by: mmichelson
Patches:
      17665.diff#2 uploaded by pprindeville (license 347)
Tested by: pprindeville

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277873 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRemove the fe80:1234::1234 test case from test_acl.c
Mark Michelson [Mon, 19 Jul 2010 17:10:00 +0000 (17:10 +0000)]
Remove the fe80:1234::1234 test case from test_acl.c

The ACL test was failing on Mac OS X because it would
convert the above invalid link-local address into
fe80::1234 while reporting no error from getaddrinfo().
Linux does not do this.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277872 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix regression with distinctive ring detection.
Jeff Peeler [Mon, 19 Jul 2010 14:39:07 +0000 (14:39 +0000)]
Fix regression with distinctive ring detection.

The issue here is that passing an array to a function prohibits the ARRAY_LEN
macro from returning the real size. To avoid this the size is now defined and
use of ARRAY_LEN is avoided.

(closes issue #15718)
Reported by: alecdavis
Patches:
      bug15718.patch uploaded by jpeeler (license 325)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277837 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMake ACLs IPv6-capable.
Mark Michelson [Mon, 19 Jul 2010 14:17:16 +0000 (14:17 +0000)]
Make ACLs IPv6-capable.

ACLs can now be configured to match IPv6 networks. This is only
relevant for ACLs in chan_sip for now since other channel drivers
do not support IPv6 addressing. However, once those channel drivers
are outfitted to support IPv6 addressing, the ACLs will already be
ready for IPv6 support.

https://reviewboard.asterisk.org/r/791

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277814 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277738 via svnmerge from
Tilghman Lesher [Sat, 17 Jul 2010 17:42:32 +0000 (17:42 +0000)]
Merged revisions 277738 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277738 | tilghman | 2010-07-17 11:59:11 -0500 (Sat, 17 Jul 2010) | 5 lines

  Remove uclibc cross-compile triplet, as uclibc has a working fork()... it's only uclinux that does not.

  (closes issue #17616)
   Reported by: pprindeville
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277775 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277568 via svnmerge from
Tilghman Lesher [Sat, 17 Jul 2010 17:39:28 +0000 (17:39 +0000)]
Merged revisions 277568 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277568 | tilghman | 2010-07-16 16:54:29 -0500 (Fri, 16 Jul 2010) | 8 lines

  Since we split values at the semicolon, we should store values with a semicolon as an encoded value.

  (closes issue #17369)
   Reported by: gkservice
   Patches:
         20100625__issue17369.diff.txt uploaded by tilghman (license 14)
   Tested by: tilghman
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277773 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAllow xmllint to be used for XML docs validation.
Russell Bryant [Sat, 17 Jul 2010 13:10:47 +0000 (13:10 +0000)]
Allow xmllint to be used for XML docs validation.

xmllint seems to be more commonly available since it comes with libxml2.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277703 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoUpdate res_fax.c to be a good xml citizen.
Bradley Latus [Sat, 17 Jul 2010 00:03:37 +0000 (00:03 +0000)]
Update res_fax.c to be a good xml citizen.
(closes issues #17667)
 Reported by: snuffy

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277667 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277625 via svnmerge from
Tim Ringenbach [Fri, 16 Jul 2010 23:23:15 +0000 (23:23 +0000)]
Merged revisions 277625 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277625 | tringenbach | 2010-07-16 17:43:39 -0500 (Fri, 16 Jul 2010) | 9 lines

  Save and restore AST_FLAG_BRIDGE_HANGUP_DONT on attended transfer.

  ast_bridge_call() clears AST_FLAG_BRIDGE_HANGUP_DONT. But during an attended
  transfer, ast_bridge_call() is called for a second bridge on the same channel,
  and it clears that flag, which still needs to get set for when the original
  ast_bridge_call() gets control back and checks it.

  Review: https://reviewboard.asterisk.org/r/741
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277657 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277497 via svnmerge from
Matthew Nicholson [Fri, 16 Jul 2010 21:24:45 +0000 (21:24 +0000)]
Merged revisions 277497 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277497 | mnicholson | 2010-07-16 16:18:38 -0500 (Fri, 16 Jul 2010) | 4 lines

  Default to no udptl error correction so that error correction will be disabled in the event that the remote end indicates that they do not support the error correction mode we requested.

  FAX-128
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277530 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix reporting estimated queue hold time.
Jeff Peeler [Fri, 16 Jul 2010 21:16:08 +0000 (21:16 +0000)]
Fix reporting estimated queue hold time.

Just say the number of seconds (after minutes) rather than doing some incorrect
calculation with respect to minutes.

(closes issue #17498)
Reported by: corruptor
Patches:
      holdesecs_bug.diff uploaded by corruptor (license 253)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277488 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFinally, a method that really fixes the assertions in chan_iax2.c related to cancelli...
Tilghman Lesher [Fri, 16 Jul 2010 20:35:28 +0000 (20:35 +0000)]
Finally, a method that really fixes the assertions in chan_iax2.c related to cancelling lagid.

No, replacing usleep(1) with sched_yield() did not have an effect.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277484 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277419 via svnmerge from
Richard Mudgett [Fri, 16 Jul 2010 20:27:51 +0000 (20:27 +0000)]
Merged revisions 277419 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277419 | rmudgett | 2010-07-16 15:18:54 -0500 (Fri, 16 Jul 2010) | 15 lines

  priexclusive in chan_dahdi.conf ignored when reloading dahdi module

  During a reload, the priexclusive and outsignalling parameters are not
  read in from the config file as intended.  Unfortunately, they get set to
  defaults as a result.  This patch makes sure that they do not get set to
  defaults during a reload.

  (closes issue #17441)
  Reported by: mtryfoss
  Patches:
        issue17441_v1.4.patch uploaded by rmudgett (license 664)
        issue17441_v1.6.2.patch uploaded by rmudgett (license 664)
        issue17441_trunk.patch uploaded by rmudgett (license 664)
  Tested by: rmudgett
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277467 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd documentation for MOH realtime fields
Tilghman Lesher [Fri, 16 Jul 2010 20:25:11 +0000 (20:25 +0000)]
Add documentation for MOH realtime fields

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277452 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoupdated devicestate test for device state changes
Matthew Nicholson [Fri, 16 Jul 2010 19:32:10 +0000 (19:32 +0000)]
updated devicestate test for device state changes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277409 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd missing handling for ringing state for use with queue empty options.
Jeff Peeler [Fri, 16 Jul 2010 19:22:49 +0000 (19:22 +0000)]
Add missing handling for ringing state for use with queue empty options.

(closes issue #17471)
Reported by: jazzy
Patches:
      app_queue.c.diff uploaded by jazzy (license 1056)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277366 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277327 via svnmerge from
Matthew Nicholson [Fri, 16 Jul 2010 18:31:08 +0000 (18:31 +0000)]
Merged revisions 277327 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277327 | mnicholson | 2010-07-16 13:30:22 -0500 (Fri, 16 Jul 2010) | 8 lines

  Interpret device state AST_DEVICE_UNKNOWN as extension state AST_EXTENSION_NOT_INUSE.

  (closes issue #16035)
  Reported by: francesco_r
  Patches:
        pbx.c.patch uploaded by viniciusfontes (license 978)
  Tested by: francesco_r, agx, lawbar
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277331 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277261 via svnmerge from
Tilghman Lesher [Fri, 16 Jul 2010 18:14:05 +0000 (18:14 +0000)]
Merged revisions 277261 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277261 | tilghman | 2010-07-16 13:04:11 -0500 (Fri, 16 Jul 2010) | 5 lines

  If variable gotten is not set, will segfault on Solaris.

  (closes issue #17636)
   Reported by: bklang
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277263 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoPrint f->subclass.integer instead of f->subclass.
Matthew Nicholson [Fri, 16 Jul 2010 18:05:01 +0000 (18:05 +0000)]
Print f->subclass.integer instead of f->subclass.

(fix build breakage introduced in r277250)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277262 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277247 via svnmerge from
Matthew Nicholson [Fri, 16 Jul 2010 17:30:39 +0000 (17:30 +0000)]
Merged revisions 277247 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277247 | mnicholson | 2010-07-16 12:29:57 -0500 (Fri, 16 Jul 2010) | 4 lines

  For pass through DTMF tones, measure the actual duration between the begin and end packets on the wire.  If it is detected to be less than AST_MIN_DTMF_DURATION, trigger dtmf emulation.

  AST-362
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277250 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 277182 via svnmerge from
Paul Belanger [Fri, 16 Jul 2010 17:13:46 +0000 (17:13 +0000)]
Merged revisions 277182 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r277182 | pabelanger | 2010-07-16 13:10:36 -0400 (Fri, 16 Jul 2010) | 8 lines

  Total analysis time error with SIP and silence suppression

  When using app_amd with SIP providers that have silence
  suppression on, the iTotalTime count increases exponentially.

  (closes issue #17656)
  Reported by: juls
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277183 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix up some weird indentation problems in reqresp_parser.c
Mark Michelson [Fri, 16 Jul 2010 16:25:01 +0000 (16:25 +0000)]
Fix up some weird indentation problems in reqresp_parser.c

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277175 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAvoid crashing when installing a duplicate translation path with a lower cost.
Sean Bright [Fri, 16 Jul 2010 15:20:40 +0000 (15:20 +0000)]
Avoid crashing when installing a duplicate translation path with a lower cost.

(closes issue #17092)
Reported by: moy
Patches:
      translate.rev254273.patch uploaded by moy (license 222)
Tested by: moy

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277143 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd Despegar.com (my main sponsor) to the CREDITS file.
Eliel C. Sardanons [Fri, 16 Jul 2010 13:40:30 +0000 (13:40 +0000)]
Add Despegar.com (my main sponsor) to the CREDITS file.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277103 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFormatting changes
Olle Johansson [Fri, 16 Jul 2010 13:32:22 +0000 (13:32 +0000)]
Formatting changes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277102 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFormatting fixes
Olle Johansson [Fri, 16 Jul 2010 13:10:24 +0000 (13:10 +0000)]
Formatting fixes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277065 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoClarify syntax changes
Olle Johansson [Fri, 16 Jul 2010 12:13:45 +0000 (12:13 +0000)]
Clarify syntax changes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277028 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdding a few more to the list of CREDITS
Olle Johansson [Fri, 16 Jul 2010 11:45:05 +0000 (11:45 +0000)]
Adding a few more to the list of CREDITS

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@277027 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFormatting changes (guideline corrections)
Olle Johansson [Fri, 16 Jul 2010 10:31:42 +0000 (10:31 +0000)]
Formatting changes (guideline corrections)

Found a unused bag of curly brackets under my table. I always wondered where
they had gone. They where indeed needed in chan_sip.c

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276989 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdding a few more credits
Olle Johansson [Fri, 16 Jul 2010 10:08:45 +0000 (10:08 +0000)]
Adding a few more credits

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276952 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd ability to configure the Max-Forwards header in the dialplan, as well as in
Olle Johansson [Fri, 16 Jul 2010 10:00:58 +0000 (10:00 +0000)]
Add ability to configure the Max-Forwards header in the dialplan, as well as in
sip.conf configuration for the channel and for devices.

The Max-Forwards header is used to prevent loops in a SIP network. Each intermediary,
like SIP proxys and SBCs, decrement this counter and detects when it reaches zero,
at which point the SIP request is nicely killed in a SIP-friendly way.

Review: https://reviewboard.asterisk.org/r/778/

Thanks to dvossel for the review and good advice.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276951 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd a dialplan function to check if a queue exists: QUEUE_EXISTS
Olle Johansson [Fri, 16 Jul 2010 09:25:48 +0000 (09:25 +0000)]
Add a dialplan function to check if a queue exists: QUEUE_EXISTS

Review: https://reviewboard.asterisk.org/r/777/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276950 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAnd yet one more
Tilghman Lesher [Fri, 16 Jul 2010 06:04:22 +0000 (06:04 +0000)]
And yet one more

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276911 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years ago"Item may be used uninitialized in this function."
Tilghman Lesher [Fri, 16 Jul 2010 05:59:11 +0000 (05:59 +0000)]
"Item may be used uninitialized in this function."

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276910 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix reversed logic of if statement.
Mark Michelson [Fri, 16 Jul 2010 05:42:24 +0000 (05:42 +0000)]
Fix reversed logic of if statement.

Found based on message from Philip Prindeville on the
Asterisk Developers mailing list.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276909 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDetect the --dynamic-list flag a bit better
Tilghman Lesher [Fri, 16 Jul 2010 05:38:06 +0000 (05:38 +0000)]
Detect the --dynamic-list flag a bit better

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276908 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix build on FreeBSD
Tilghman Lesher [Fri, 16 Jul 2010 04:45:33 +0000 (04:45 +0000)]
Fix build on FreeBSD

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276871 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix trunk build for Mac OS X 10.6
Tilghman Lesher [Fri, 16 Jul 2010 04:23:02 +0000 (04:23 +0000)]
Fix trunk build for Mac OS X 10.6

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276870 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAllow ipaddress to contain the maximum IPv6 address.
Tilghman Lesher [Fri, 16 Jul 2010 04:18:58 +0000 (04:18 +0000)]
Allow ipaddress to contain the maximum IPv6 address.

Also, update meetme to the full list of supported fields.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276869 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoQuote AC_SUBST within m4_ifval, so it does not get prematurely expanded.
Tilghman Lesher [Thu, 15 Jul 2010 23:25:09 +0000 (23:25 +0000)]
Quote AC_SUBST within m4_ifval, so it does not get prematurely expanded.

(closes issue #17654)
 Reported by: pprindeville
 Patches:
       issue17654.diff uploaded by qwell (license 4)
 Tested by: qwell, pprindeville

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276830 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoCorrect not setting the bindport before attempting to open the socket.
Jeff Peeler [Thu, 15 Jul 2010 20:21:03 +0000 (20:21 +0000)]
Correct not setting the bindport before attempting to open the socket.

Related to changes from 276571, I was accidentally testing with a port set in
my configuration causing me to miss this. Also moved the TCP handling as well
to occur before build_peer is called.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276788 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDefine LLONG_MAX on systems that do not have it.
Tilghman Lesher [Thu, 15 Jul 2010 19:46:57 +0000 (19:46 +0000)]
Define LLONG_MAX on systems that do not have it.

(closes issue #17644)
 Reported by: pprindeville

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276769 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix linking asterisk on CentOS 5, which is using gcc 4.1.1. Gcc 4.1.2 has the real...
Tilghman Lesher [Thu, 15 Jul 2010 18:44:20 +0000 (18:44 +0000)]
Fix linking asterisk on CentOS 5, which is using gcc 4.1.1.  Gcc 4.1.2 has the real fix.

Review: https://reviewboard.asterisk.org/r/790/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276731 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 276652 via svnmerge from
Jeff Peeler [Thu, 15 Jul 2010 13:51:11 +0000 (13:51 +0000)]
Merged revisions 276652 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r276652 | jpeeler | 2010-07-15 08:48:58 -0500 (Thu, 15 Jul 2010) | 2 lines

  In a perfect world, the frame source would never be NULL. In the meantime, don't crash when it is.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276653 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd lua5.1 to the handy dandy list of packages.
Russell Bryant [Thu, 15 Jul 2010 12:21:10 +0000 (12:21 +0000)]
Add lua5.1 to the handy dandy list of packages.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276616 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix MWI notification transmission problems over SIP.
Jeff Peeler [Wed, 14 Jul 2010 22:58:24 +0000 (22:58 +0000)]
Fix MWI notification transmission problems over SIP.

MWI updates were not being sent if no messages were found in the event cache.
This was corrected since a phone may need to clear its MWI status configured
previously from another mailbox.

Upon module or sip reload, MWI updates could not be sent due to the sipsock
socket not being set early enough in reload_config. The code handling the
descriptor assignment and such has simply been moved before the call to
build_peer.

Issuing a sip reload cleared the IP address of the peer, but skipped checking
the database for registration information. The database is now checked both
for sip reload and actually reloading the module.

If a transmission occurs before the do_monitor thread has started, do not
attempt to send a signal to it.

(closes issue #17398)
Reported by: ip-rob

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276571 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix errors where incorrect address information was printed.
Mark Michelson [Wed, 14 Jul 2010 22:32:29 +0000 (22:32 +0000)]
Fix errors where incorrect address information was printed.

ast_sockaddr_stringiy_fmt (which is call by all ast_sockaddr_stringify* functions)
uses thread-local storage for storing the string that it creates. In cases where
ast_sockaddr_stringify_fmt was being called twice within the same statement, the
result of one call would be overwritten by the result of the other call. This
usually was happening in printf-like statements and was resulting in the same
stringified addressed being printed twice instead of two separate addresses.

I have fixed this by using ast_strdupa on the result of stringify functions if
they are used twice within the same statement. As far as I could tell, there were
no instances where a pointer to the result of such a call were saved anywhere, so
this is the only situation I could see where this error could occur.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276570 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMake compile again.
Richard Mudgett [Wed, 14 Jul 2010 21:29:32 +0000 (21:29 +0000)]
Make compile again.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276531 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoOops, merge reverted this fix.
Tilghman Lesher [Wed, 14 Jul 2010 21:11:09 +0000 (21:11 +0000)]
Oops, merge reverted this fix.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276493 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRemove the old stub files, preferring the optional_api method.
Tilghman Lesher [Wed, 14 Jul 2010 20:48:59 +0000 (20:48 +0000)]
Remove the old stub files, preferring the optional_api method.

(closes issue #17475)
 Reported by: tilghman

Review: https://reviewboard.asterisk.org/r/695/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276490 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDon't try to call an embedded module's backup_globals() function until
Kevin P. Fleming [Wed, 14 Jul 2010 20:15:48 +0000 (20:15 +0000)]
Don't try to call an embedded module's backup_globals() function until
after confirming it exists.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276441 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agohandle special case were "200 Ok" to pending INVITE never receives ACK
David Vossel [Wed, 14 Jul 2010 19:51:08 +0000 (19:51 +0000)]
handle special case were "200 Ok" to pending INVITE never receives ACK

Unlike most responses, the 200 Ok to a pending INVITE Request is
acknowledged by an ACK Request.  If the ACK Request for this Response is not received
the previous behavior was to immediately destroy the dialog and hangup
the channel. Now in an effort to be more RFC compliant, instead of immediately
destroying the dialog during this special case, termination is done with a BYE Request
as the dialog is technically confirmed when the 200 Ok is sent even if the ACK is
never received.  The behavior of immediately hanging up the channel remains.
This only affects how dialog termination proceeds for this one special case.

RFC 3261 section 13.3.1.4
"If the server retransmits the 2xx response for 64*T1 seconds without receiving
an ACK, the dialog is confirmed, but the session SHOULD be terminated.  This is
accomplished with a BYE, as described in Section 15."

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276439 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoExpand the caller ANI field to an ast_party_id
Richard Mudgett [Wed, 14 Jul 2010 16:58:03 +0000 (16:58 +0000)]
Expand the caller ANI field to an ast_party_id

Expand the ani field in ast_party_caller and ast_party_connected_line to
an ast_party_id.

This is an extension to the ast_callerid restructuring patch in review:
https://reviewboard.asterisk.org/r/702/

Review: https://reviewboard.asterisk.org/r/744/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276393 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agocollapse debug code in retrans_pkt into separate lines
David Vossel [Wed, 14 Jul 2010 16:40:42 +0000 (16:40 +0000)]
collapse debug code in retrans_pkt into separate lines

I've been working in this function a bunch lately, and
these huge debug strings are getting annoying.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276392 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMake compile again.
Richard Mudgett [Wed, 14 Jul 2010 16:39:18 +0000 (16:39 +0000)]
Make compile again.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276391 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDo not skip sending MWI for a peer if an address is defined. Really just a merge...
Jeff Peeler [Wed, 14 Jul 2010 16:36:02 +0000 (16:36 +0000)]
Do not skip sending MWI for a peer if an address is defined. Really just a merge mistake from IPv6

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276389 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix documentation for pgsql cel and cdr, and slightly improve pgsql_cel.
Tim Ringenbach [Wed, 14 Jul 2010 16:09:11 +0000 (16:09 +0000)]
Fix documentation for pgsql cel and cdr, and slightly improve pgsql_cel.

Change the documented pgsql schema to use "timestamp" instead of "time",
as the latter is only a time without a date.

Added some missing columns for cel's pgsql schema, and corrected spelling
on some others. Updated cel's uniqueid size to be the same as the cdr.
Added id column to cel's pgsql schema and updated code to allow unknown
columns to get their default value instead of forcing 0 or empty string.

Added microseconds to the timestamp cel logs to pgsql.

Review: https://reviewboard.asterisk.org/r/734

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276349 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoast_callerid restructuring
Richard Mudgett [Wed, 14 Jul 2010 15:48:36 +0000 (15:48 +0000)]
ast_callerid restructuring

The purpose of this patch is to eliminate struct ast_callerid since it has
turned into a miscellaneous collection of various party information.

Eliminate struct ast_callerid and replace it with the following struct
organization:

struct ast_party_name {
char *str;
int char_set;
int presentation;
unsigned char valid;
};
struct ast_party_number {
char *str;
int plan;
int presentation;
unsigned char valid;
};
struct ast_party_subaddress {
char *str;
int type;
unsigned char odd_even_indicator;
unsigned char valid;
};
struct ast_party_id {
struct ast_party_name name;
struct ast_party_number number;
struct ast_party_subaddress subaddress;
char *tag;
};
struct ast_party_dialed {
struct {
char *str;
int plan;
} number;
struct ast_party_subaddress subaddress;
int transit_network_select;
};
struct ast_party_caller {
struct ast_party_id id;
char *ani;
int ani2;
};

The new organization adds some new information as well.

* The party name and number now have their own presentation value that can
be manipulated independently.  ISDN supplies the presentation value for
the name and number at different times with the possibility that they
could be different.

* The party name and number now have a valid flag.  Before this change the
name or number string could be empty if the presentation were restricted.
Most channel drivers assume that the name or number is then simply not
available instead of indicating that the name or number was restricted.

* The party name now has a character set value.  SIP and Q.SIG have the
ability to indicate what character set a name string is using so it could
be presented properly.

* The dialed party now has a numbering plan value that could be useful to
have available.

The various channel drivers will need to be updated to support the new
core features as needed.  They have simply been converted to supply
current functionality at this time.

The following items of note were either corrected or enhanced:

* The CONNECTEDLINE() and REDIRECTING() dialplan functions were
consolidated into func_callerid.c to share party id handling code.

* CALLERPRES() is now deprecated because the name and number have their
own presentation values.

* Fixed app_alarmreceiver.c write_metadata().  The workstring[] could
contain garbage.  It also can only contain the caller id number so using
ast_callerid_parse() on it is silly.  There was also a typo in the
CALLERNAME if test.

* Fixed app_rpt.c using ast_callerid_parse() on the channel's caller id
number string.  ast_callerid_parse() alters the given buffer which in this
case is the channel's caller id number string.  Then using
ast_shrink_phone_number() could alter it even more.

* Fixed caller ID name and number memory leak in chan_usbradio.c.

* Fixed uninitialized char arrays cid_num[] and cid_name[] in
sig_analog.c.

* Protected access to a caller channel with lock in chan_sip.c.

* Clarified intent of code in app_meetme.c sla_ring_station() and
dial_trunk().  Also made save all caller ID data instead of just the name
and number strings.

* Simplified cdr.c set_one_cid().  It hand coded the ast_callerid_merge()
function.

* Corrected some weirdness with app_privacy.c's use of caller
presentation.

Review: https://reviewboard.asterisk.org/r/702/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276347 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 276267 via svnmerge from
Leif Madsen [Wed, 14 Jul 2010 11:51:48 +0000 (11:51 +0000)]
Merged revisions 276267 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r276267 | lmadsen | 2010-07-14 06:49:01 -0500 (Wed, 14 Jul 2010) | 1 line

  Update documentation for voicemail.conf externpass option.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276268 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agochan_sip: RFC compliant retransmission timeout
David Vossel [Tue, 13 Jul 2010 22:18:38 +0000 (22:18 +0000)]
chan_sip: RFC compliant retransmission timeout

Retransmission of packets should not be based on how many packets were
sent, but instead on a timeout period.  Depending on whether or not the
packet is for a INVITE or NON-INVITE transaction, the number of packets
sent during the retransmission timeout period will be different, so
timing out based on the number of packets sent is not accurate.

This patch fixes this by removing the retransmit limit and only stopping
retransmission after a timeout period is reached.  By default this
timeout period is 64*(Timer T1) for both INVITE and non-INVITE
transactions.  For more information on sip timer values refer to
RFC3261 Appendix A.

Review: https://reviewboard.asterisk.org/r/749/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276219 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRevert early destruction of RTP sessions
Terry Wilson [Tue, 13 Jul 2010 21:42:42 +0000 (21:42 +0000)]
Revert early destruction of RTP sessions

Some code improperly assumes that the sessions are still there, so revert the
change until I can find all of them and fix them.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276206 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRecorded merge of revisions 276126 via svnmerge from
Russell Bryant [Tue, 13 Jul 2010 19:15:47 +0000 (19:15 +0000)]
Recorded merge of revisions 276126 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r276126 | russell | 2010-07-13 14:14:54 -0500 (Tue, 13 Jul 2010) | 2 lines

  Only reset a CDR that exists.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276127 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 276123 via svnmerge from
Russell Bryant [Tue, 13 Jul 2010 19:09:42 +0000 (19:09 +0000)]
Merged revisions 276123 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r276123 | russell | 2010-07-13 14:06:53 -0500 (Tue, 13 Jul 2010) | 2 lines

  Use chan->cdr instead of chan_cdr (just like peer->cdr instead of peer_cdr in the last commit).
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276124 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoOops, XML documentation fix.
Tilghman Lesher [Tue, 13 Jul 2010 19:05:17 +0000 (19:05 +0000)]
Oops, XML documentation fix.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276122 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoIt really cannot fail in the places below, but the stupid compiler doesn't know that.
Tilghman Lesher [Tue, 13 Jul 2010 19:00:02 +0000 (19:00 +0000)]
It really cannot fail in the places below, but the stupid compiler doesn't know that.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276120 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoWeird compiler error on Bamboo.
Tilghman Lesher [Tue, 13 Jul 2010 18:41:59 +0000 (18:41 +0000)]
Weird compiler error on Bamboo.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276118 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFILE() now supports line-mode and writing (altering) files.
Tilghman Lesher [Tue, 13 Jul 2010 18:31:41 +0000 (18:31 +0000)]
FILE() now supports line-mode and writing (altering) files.

(closes issue #16461)
 Reported by: skyman
 Patches:
       20100622__issue16461.diff.txt uploaded by tilghman (license 14)
 Tested by: tilghman

Review: https://reviewboard.asterisk.org/r/737/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276114 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 275773 via svnmerge from
Jeff Peeler [Tue, 13 Jul 2010 17:37:40 +0000 (17:37 +0000)]
Merged revisions 275773 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r275773 | jpeeler | 2010-07-12 15:34:51 -0500 (Mon, 12 Jul 2010) | 12 lines

  Make user removals and traversals thread safe in meetme.

  Race conditions present in meetme involving the user list where a lack of
  locking has the potential for a user to be removed during a traversal or as in
  the case of the reporter after checking if the list is empty could cause a
  crash. Fixing this was done by convering the userlist to an ao2 container.

  (closes issue #17390)
  Reported by: Vince

  Review: https://reviewboard.asterisk.org/r/746/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@276074 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDestroy RTP fds when we schedule final dialog destruction
Terry Wilson [Tue, 13 Jul 2010 17:11:37 +0000 (17:11 +0000)]
Destroy RTP fds when we schedule final dialog destruction

Since we are only keeping the dialog around for retransmissions at this point
and there is no possibility that we are still handling RTP, go ahead and
destroy the RTP sessions. Keeping them alive for 32 past when they are used
is unnecessary and can lead to problems with having too many open file
descriptors, etc.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275998 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 275994 via svnmerge from
Russell Bryant [Tue, 13 Jul 2010 16:53:44 +0000 (16:53 +0000)]
Merged revisions 275994 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r275994 | russell | 2010-07-13 11:51:18 -0500 (Tue, 13 Jul 2010) | 14 lines

  Access peer->cdr directly instead of through a saved off reference.

  At this point in the code, it is possible that peer_cdr may be invalid.
  Specifically, in the blind transfer code, CDRs are swapped between channels.
  So, peer_cdr is no longer == peer->cdr.

  The scenario that exposed a crash in this code was a blind transfer that hit
  the system call limit, causing the transferee channel to get destroyed after
  the transfer attempt failed.  Even if it succeeds and this code doesn't crash,
  this code was still trying to reset a CDR on a channel that was now owned by
  a different thread, which is a BadThing(tm).

  (ABE-2417)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275995 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 275909 via svnmerge from
Tilghman Lesher [Tue, 13 Jul 2010 14:48:40 +0000 (14:48 +0000)]
Merged revisions 275909 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r275909 | tilghman | 2010-07-13 09:47:30 -0500 (Tue, 13 Jul 2010) | 2 lines

  Move SQL scripts into their own database-specific directories.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275910 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd example script for use with the externpasscheck voicemail.conf option.
Russell Bryant [Tue, 13 Jul 2010 11:41:54 +0000 (11:41 +0000)]
Add example script for use with the externpasscheck voicemail.conf option.

(closes issue #17628)
Reported by: lmadsen
Tested by: russell, lmadsen

Review: https://reviewboard.asterisk.org/r/774/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275863 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDon't try to ref authpeer when it isn't set
Terry Wilson [Mon, 12 Jul 2010 23:27:42 +0000 (23:27 +0000)]
Don't try to ref authpeer when it isn't set

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275816 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd which ITU spec specifies the numbering plan.
Richard Mudgett [Mon, 12 Jul 2010 17:54:46 +0000 (17:54 +0000)]
Add which ITU spec specifies the numbering plan.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275725 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 275665 via svnmerge from
Jeff Peeler [Mon, 12 Jul 2010 17:21:01 +0000 (17:21 +0000)]
Merged revisions 275665 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r275665 | jpeeler | 2010-07-12 11:58:39 -0500 (Mon, 12 Jul 2010) | 11 lines

  Change ast_write to not stop generator when called from ast_prod.

  For SIP channels configured with the progressinband option on, the ringback was
  being immediately stopped. This problem was due to ast_prod being moved for a
  deadlock fix in 259858. Prodding the channel after setting up the generator
  triggered the check in ast_write to stop the generator. The fix here should
  write the frame the same as was done before the call to ast_prod was moved.

  (closes issue #17372)
  Reported by: tech_admin
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275682 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agocdr_pgsql does not detect when a table is found.
Leif Madsen [Mon, 12 Jul 2010 15:37:01 +0000 (15:37 +0000)]
cdr_pgsql does not detect when a table is found.

This change adds an ERROR message to let you know when a failure exists to
get the columns from the pgsql database, which typically means that the
table does not exist.

(closes issue #17478)
Reported by: kobaz
Patches:
      cdr_pgsql.patch uploaded by kobaz (license 834)
Tested by: kobaz, russell, lmadsen

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275626 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAllow netsock2.c to compile on systems that do not define AI_NUMERICSERV.
Mark Michelson [Mon, 12 Jul 2010 14:55:23 +0000 (14:55 +0000)]
Allow netsock2.c to compile on systems that do not define AI_NUMERICSERV.

(closes issue #17617)
Reported by: pprindeville
Patches:
      asterisk-trunk-bugid17617.patch uploaded by pprindeville (license 347)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275587 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdded support for indirect work mode.
TransNexus OSP Development [Mon, 12 Jul 2010 04:16:18 +0000 (04:16 +0000)]
Added support for indirect work mode.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275551 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoWhen creating a conference for a unit test, it is not mandatory to open a
Eliel C. Sardanons [Sat, 10 Jul 2010 20:49:30 +0000 (20:49 +0000)]
When creating a conference for a unit test, it is not mandatory to open a
dahdi pseudo channel, so if we fail doing it, continue creating the conference.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@275509 65c4cc65-6c06-0410-ace0-fbb531ad65f3