asterisk/asterisk.git
8 years agoMerged revisions 305131 via svnmerge from
Jason Parker [Mon, 31 Jan 2011 21:01:28 +0000 (21:01 +0000)]
Merged revisions 305131 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r305131 | qwell | 2011-01-31 15:00:25 -0600 (Mon, 31 Jan 2011) | 16 lines

  Merged revisions 305130 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r305130 | qwell | 2011-01-31 14:59:37 -0600 (Mon, 31 Jan 2011) | 9 lines

    Merged revisions 305129 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r305129 | qwell | 2011-01-31 14:56:25 -0600 (Mon, 31 Jan 2011) | 2 lines

      Set file descriptors to -1 on creation, so that we don't see weirdness later.
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@305132 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAsterisk HTTP response Content-type
Andrew Latham [Mon, 31 Jan 2011 13:57:53 +0000 (13:57 +0000)]
Asterisk HTTP response Content-type

Address content type for BSD and other platforms

(closes issue #18456)
Reported by: alexo
Patches:
    asterisk18_http.patch uploaded by alexo (license 1175)
Tested by: alexo

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@305084 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 305040 via svnmerge from
Tilghman Lesher [Mon, 31 Jan 2011 07:52:48 +0000 (07:52 +0000)]
Merged revisions 305040 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r305040 | tilghman | 2011-01-31 01:51:40 -0600 (Mon, 31 Jan 2011) | 2 lines

  Use the non-specific API aliases, to avoid a problem with building the utils directory.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@305041 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304985 via svnmerge from
Tilghman Lesher [Mon, 31 Jan 2011 07:28:06 +0000 (07:28 +0000)]
Merged revisions 304985 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304985 | tilghman | 2011-01-31 01:27:13 -0600 (Mon, 31 Jan 2011) | 16 lines

  Merged revisions 304978 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304978 | tilghman | 2011-01-31 01:25:14 -0600 (Mon, 31 Jan 2011) | 9 lines

    Merged revisions 304952 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r304952 | tilghman | 2011-01-31 00:54:45 -0600 (Mon, 31 Jan 2011) | 2 lines

      Fix compilation when ODBC_STORAGE is defined.
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304989 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304950 via svnmerge from
Tilghman Lesher [Mon, 31 Jan 2011 06:50:49 +0000 (06:50 +0000)]
Merged revisions 304950 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r304950 | tilghman | 2011-01-31 00:41:36 -0600 (Mon, 31 Jan 2011) | 18 lines

  Change mutex tracking so that it only consumes memory in the core mutex object when it's actually being used.

  This reduces the overall size of a mutex which was 3016 bytes before this back
  down to 216 bytes (this is on 64-bit Linux with a glibc-implemented mutex).
  The exactness of the numbers here may vary slightly based upon how mutexes are
  implemented on a platform, but the long and short of it is that prior to this
  commit, chan_iax2 held down 98MB of memory on a 64-bit system for nothing more
  than a table of 32767 locks.  After this commit, the same table occupies a mere
  7MB of memory.

  (closes issue #18194)
   Reported by: job
   Patches:
         20110124__issue18194.diff.txt uploaded by tilghman (license 14)
   Tested by: tilghman

  Review: https://reviewboard.asterisk.org/r/1066
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304951 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd Function and Application Relationships to documentation
Andrew Latham [Sun, 30 Jan 2011 00:22:59 +0000 (00:22 +0000)]
Add Function and Application Relationships to documentation

Add and extend the see-also sections to the documentation for applications
and functions in an effort to expand the online documentation of the wiki.
Also check for and update any links to moved documentation in the doc folder.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304913 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304866 via svnmerge from
Sean Bright [Sat, 29 Jan 2011 23:10:06 +0000 (23:10 +0000)]
Merged revisions 304866 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304866 | seanbright | 2011-01-29 18:07:18 -0500 (Sat, 29 Jan 2011) | 14 lines

  Merged revisions 304865 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304865 | seanbright | 2011-01-29 18:05:25 -0500 (Sat, 29 Jan 2011) | 7 lines

    Plug some memory leaks in the LDAP realtime driver.

    (closes issue #18435)
    Reported by: zaltar
    Patches:
          res_config_ldap.patch uploaded by zaltar (license 1148)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304867 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304777 via svnmerge from
Sean Bright [Sat, 29 Jan 2011 18:10:34 +0000 (18:10 +0000)]
Merged revisions 304777 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304777 | seanbright | 2011-01-29 13:09:37 -0500 (Sat, 29 Jan 2011) | 22 lines

  Merged revisions 304776 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304776 | seanbright | 2011-01-29 13:08:14 -0500 (Sat, 29 Jan 2011) | 15 lines

    If we fail to allocate our announcement objects, make sure we don't leak objects.

    The majority of this patch was committed already in r304726 and r304729.

    (issue #18225)
    Reported by: kenji

    (issue #18444)
    Reported by: junky

    (closes issue #18343)
    Reported by: kobaz
    Patches:
          meetme-refs.diff uploaded by kobaz (license 834)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304778 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304774 via svnmerge from
Sean Bright [Sat, 29 Jan 2011 17:57:01 +0000 (17:57 +0000)]
Merged revisions 304774 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304774 | seanbright | 2011-01-29 12:54:43 -0500 (Sat, 29 Jan 2011) | 16 lines

  Merged revisions 304773 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304773 | seanbright | 2011-01-29 12:51:28 -0500 (Sat, 29 Jan 2011) | 9 lines

    When we pass the S() or L() options to MeetMe, make sure that we honor C as well.

    Without this patch, if the user was kicked from the conference via the S() or L()
    mechanism, we would just hang up on them even if we also passed C (continue in
    dialplan when kicked).  With this patch we honor the C flag in those cases.

    (closes issue #17317)
    Reported by: var
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304775 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304730 via svnmerge from
Sean Bright [Sat, 29 Jan 2011 17:34:22 +0000 (17:34 +0000)]
Merged revisions 304730 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304730 | seanbright | 2011-01-29 12:15:27 -0500 (Sat, 29 Jan 2011) | 22 lines

  Merged revisions 304729 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304729 | seanbright | 2011-01-29 12:01:51 -0500 (Sat, 29 Jan 2011) | 15 lines

    Make sure that we unref the correct object when ejecting the most recent caller.

    Currently, when we kick the last user to enter, we decrement our own reference
    count which results in a crash when we kick another user or when we exit the
    conference ourselves.

    This will fix #18225 in 1.8 and trunk, but that particular bug does not exist in
    1.6.2.

    (closes issue #18225)
    Reported by: kenji
    Patches:
          issue18225.patch uploaded by seanbright (license 71)
    Tested by: seanbright
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304772 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304727 via svnmerge from
Sean Bright [Sat, 29 Jan 2011 16:31:17 +0000 (16:31 +0000)]
Merged revisions 304727 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304727 | seanbright | 2011-01-29 11:28:27 -0500 (Sat, 29 Jan 2011) | 16 lines

  Merged revisions 304726 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304726 | seanbright | 2011-01-29 11:26:57 -0500 (Sat, 29 Jan 2011) | 9 lines

    Fix user reference leak in MeetMe.

    We were unlinking the user from the conferences user container, but not
    decrementing the reference count of the user as well, resulting in a leak.

    (closes issue #18444)
    Reported by: junky
    Tested by: seanbright
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304728 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304683 via svnmerge from
Sean Bright [Fri, 28 Jan 2011 22:59:27 +0000 (22:59 +0000)]
Merged revisions 304683 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304683 | seanbright | 2011-01-28 17:54:23 -0500 (Fri, 28 Jan 2011) | 16 lines

  Merged revisions 304659,304682 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304659 | seanbright | 2011-01-28 16:22:09 -0500 (Fri, 28 Jan 2011) | 5 lines

    Don't leak references if we can't create a pseudo channel for mixing in MeetMe.

    If there was a problem allocating a pseudo channel when building our meetme, we
    weren't destroying our user container or destroying the mutexes that we created.
  ........
    r304682 | seanbright | 2011-01-28 17:38:05 -0500 (Fri, 28 Jan 2011) | 2 lines

    Revert part of the previous commit that snuck in.
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304684 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304638 via svnmerge from
Sean Bright [Fri, 28 Jan 2011 20:19:57 +0000 (20:19 +0000)]
Merged revisions 304638 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r304638 | seanbright | 2011-01-28 15:19:08 -0500 (Fri, 28 Jan 2011) | 11 lines

  Restore some conditionals that we lost in r277814.

  There are some cases where ast_append_ha() is called with a NULL instead of a
  valid int pointer.  So if we get a NULL, don't try to dereference it.

  (closes issue #18162)
  Reported by: imcdona
  Patches:
        issue0018162.patch uploaded by pabelanger (license 224)
  Tested by: enegaard
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304639 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoPatch that fixes the "realtime show pgsql cache" command crash when giving a
Brett Bryant [Thu, 27 Jan 2011 20:09:33 +0000 (20:09 +0000)]
Patch that fixes the "realtime show pgsql cache" command crash when giving a
table name, because of the use of an uninitialized variable. Fixes an error
introduced in r300882.

(closes issue #18605)
Reported by: romain_proformatique
Patches:
      res_config_pgsql_fix.patch uploaded by romain proformatique (license 975)
Tested by: romain_proformatique

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304600 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoFix bug with 'F' option for ReceiveFAX and SendFAX.
Kevin P. Fleming [Thu, 27 Jan 2011 20:07:05 +0000 (20:07 +0000)]
Fix bug with 'F' option for ReceiveFAX and SendFAX.

Skipping the call to set_t38_fax_caps() caused the FAX session
details to not be marked as supporting audio FAX either... the
function's name is a bit misleading. This patch restores the
single bit of non-T.38 behavior from that function when audio
mode is forced.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304599 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304554 via svnmerge from
Richard Mudgett [Thu, 27 Jan 2011 19:12:32 +0000 (19:12 +0000)]
Merged revisions 304554 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r304554 | rmudgett | 2011-01-27 13:08:14 -0600 (Thu, 27 Jan 2011) | 4 lines

  Warning message if CALLCOMPLETION(cc_callback_macro or cc_agent_dialstring) are empty.

  Test if the value pointer is not NULL instead of not ast_strlen_zero().
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304555 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304466 via svnmerge from
Jason Parker [Thu, 27 Jan 2011 17:03:44 +0000 (17:03 +0000)]
Merged revisions 304466 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304466 | qwell | 2011-01-27 11:03:01 -0600 (Thu, 27 Jan 2011) | 23 lines

  Merged revisions 304465 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304465 | qwell | 2011-01-27 11:01:24 -0600 (Thu, 27 Jan 2011) | 16 lines

    Merged revisions 304464 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r304464 | qwell | 2011-01-27 10:57:46 -0600 (Thu, 27 Jan 2011) | 9 lines

      Fix default prefix=/usr regression on non-Linux systems.

      This partially reverts a change made in branches/1.4/ r267759, which will
      cause issue #17013 to be reopened.  This issue was pointed out by a user
      on #asterisk, who helpfully discovered that paths were being set incorrectly.

      To truly understand what was wrong, one should run:
          svn diff --force -c<this revision> configure
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304467 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304462 via svnmerge from
Jason Parker [Thu, 27 Jan 2011 16:49:38 +0000 (16:49 +0000)]
Merged revisions 304462 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304462 | qwell | 2011-01-27 10:48:44 -0600 (Thu, 27 Jan 2011) | 16 lines

  Merged revisions 304461 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304461 | qwell | 2011-01-27 10:48:00 -0600 (Thu, 27 Jan 2011) | 9 lines

    Merged revisions 304460 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r304460 | qwell | 2011-01-27 10:47:03 -0600 (Thu, 27 Jan 2011) | 1 line

      Rerun bootstrap.sh with no changes, so that it is more obvious what my next commit changes.
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304463 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoRename the SendFAX/ReceiveFAX 'force audio' option.
Kevin P. Fleming [Thu, 27 Jan 2011 15:57:52 +0000 (15:57 +0000)]
Rename the SendFAX/ReceiveFAX 'force audio' option.

The recently added option to disable usage of T.38 for a single
session should have been named 'F' for 'force audio', since that
is really what the user is asking to happen (and it's a positive
option instead of a negative option that way).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304422 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged from revision 304341
Richard Mudgett [Thu, 27 Jan 2011 00:06:27 +0000 (00:06 +0000)]
Merged from revision 304341
https://origsvn.digium.com/svn/asterisk/be/branches/C.3-bier

..........
  r304341 | rmudgett | 2011-01-26 16:38:39 -0600 (Wed, 26 Jan 2011) | 7 lines

  Add connected line chan_dahdi.conf pricpndialplan option.

  * Added from_channel value to prilocaldialplan option.

  JIRA ABE-2731
  JIRA SWP-2842
..........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304385 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd option to followme to delay answer until ready to bridge call.
Jeff Peeler [Wed, 26 Jan 2011 23:41:55 +0000 (23:41 +0000)]
Add option to followme to delay answer until ready to bridge call.

Followme answers an incoming call if it hasn't already been answered and starts
MOH. Some poorly designed autodialers see the answer and start playing their
message to the hold music. The 'N' option has been added to indicate ringing and
not answer until the call is accepted.

(closes issue #18479)
Reported by: ianc
Patches:
      trunk_followme.diff uploaded by ianc (license 998)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304384 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd ability to disable T.38 usage for specific SendFAX/ReceiveFAX sessions.
Kevin P. Fleming [Wed, 26 Jan 2011 22:39:07 +0000 (22:39 +0000)]
Add ability to disable T.38 usage for specific SendFAX/ReceiveFAX sessions.

Sometimes during troubleshooting it can be useful to disable T.38 usage in order
to narrow down a problem. This patch adds an 'n' option to SendFAX and ReceiveFAX
so that can be done without having to disable T.38 usage entirely for the peer
that Asterisk is communicating with.

(inspired by trying to assist Bryant Zimmerman on asterisk-users)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304342 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304339 via svnmerge from
Jeff Peeler [Wed, 26 Jan 2011 22:27:51 +0000 (22:27 +0000)]
Merged revisions 304339 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304339 | jpeeler | 2011-01-26 16:27:30 -0600 (Wed, 26 Jan 2011) | 9 lines

  Merged revisions 304338 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304338 | jpeeler | 2011-01-26 16:26:37 -0600 (Wed, 26 Jan 2011) | 2 lines

    Change delimiter used internally for GOTO_ON_BLINDXFR to commas to match 76703.
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304340 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304250 via svnmerge from
Mark Michelson [Wed, 26 Jan 2011 21:03:44 +0000 (21:03 +0000)]
Merged revisions 304250 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.6.2

................
  r304250 | mmichelson | 2011-01-26 15:02:10 -0600 (Wed, 26 Jan 2011) | 9 lines

  Merged revisions 304242 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.4

  ........
    r304242 | mmichelson | 2011-01-26 14:38:37 -0600 (Wed, 26 Jan 2011) | 3 lines

    Get rid of unused 'verbose' field in ast_udptl
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304252 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304245 via svnmerge from
Matthew Nicholson [Wed, 26 Jan 2011 20:44:47 +0000 (20:44 +0000)]
Merged revisions 304245 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304245 | mnicholson | 2011-01-26 14:43:27 -0600 (Wed, 26 Jan 2011) | 20 lines

  Merged revisions 304244 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304244 | mnicholson | 2011-01-26 14:42:16 -0600 (Wed, 26 Jan 2011) | 13 lines

    Merged revisions 304241 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r304241 | mnicholson | 2011-01-26 14:38:22 -0600 (Wed, 26 Jan 2011) | 6 lines

      This patch modifies chan_sip to route responses to the address the request came from.  It also modifies chan_sip to respect the maddr parameter in the Via header.

      ABE-2664

      Review: https://reviewboard.asterisk.org/r/1059/
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304246 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304186 via svnmerge from
Sean Bright [Wed, 26 Jan 2011 20:25:24 +0000 (20:25 +0000)]
Merged revisions 304186 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304186 | seanbright | 2011-01-26 15:23:48 -0500 (Wed, 26 Jan 2011) | 16 lines

  Merged revisions 304181 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304181 | seanbright | 2011-01-26 15:22:47 -0500 (Wed, 26 Jan 2011) | 9 lines

    Merged revisions 304159 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r304159 | seanbright | 2011-01-26 15:18:29 -0500 (Wed, 26 Jan 2011) | 1 line

      Make sure the sample queues.conf is properly commented.
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304195 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303907 via svnmerge from
Matthew Nicholson [Wed, 26 Jan 2011 19:58:14 +0000 (19:58 +0000)]
Merged revisions 303907 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r303907 | mnicholson | 2011-01-25 14:56:12 -0600 (Tue, 25 Jan 2011) | 2 lines

  Reimplemented fax session reservation to reverse the ABI breakage introduced in r297486.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304152 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304150 via svnmerge from
Richard Mudgett [Wed, 26 Jan 2011 19:40:26 +0000 (19:40 +0000)]
Merged revisions 304150 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304150 | rmudgett | 2011-01-26 13:39:35 -0600 (Wed, 26 Jan 2011) | 16 lines

  Merged revisions 304149 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304149 | rmudgett | 2011-01-26 13:38:38 -0600 (Wed, 26 Jan 2011) | 9 lines

    Merged revisions 304148 from
    https://origsvn.digium.com/svn/asterisk/be/branches/C.3-bier

    ..........
      r304148 | rmudgett | 2011-01-26 13:23:46 -0600 (Wed, 26 Jan 2011) | 2 lines

      Update documentation for DAHDISendCallreroutingFacility() application.
    ..........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304151 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304097 via svnmerge from
Sean Bright [Wed, 26 Jan 2011 01:27:39 +0000 (01:27 +0000)]
Merged revisions 304097 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304097 | seanbright | 2011-01-25 20:26:26 -0500 (Tue, 25 Jan 2011) | 19 lines

  Merged revisions 304096 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r304096 | seanbright | 2011-01-25 20:24:58 -0500 (Tue, 25 Jan 2011) | 12 lines

    Per the man page, setvbuf() must be called before any other operation on an open file.

    We use setvbuf() to associate a buffer with a stream, but we have already written
    to the open file.  This works (by chance) on Linux, but fails on other platforms,
    such as OpenSolaris.

    (closes issue #16610)
    Reported by: bklang
    Patches:
          setvbuf.patch uploaded by crjw (license 963)
    Tested by: bklang, asgaroth, efutch
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304098 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 304007 via svnmerge from
Richard Mudgett [Tue, 25 Jan 2011 23:31:40 +0000 (23:31 +0000)]
Merged revisions 304007 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r304007 | rmudgett | 2011-01-25 17:28:25 -0600 (Tue, 25 Jan 2011) | 22 lines

  Merged revisions 304006 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r304006 | rmudgett | 2011-01-25 17:25:32 -0600 (Tue, 25 Jan 2011) | 15 lines

    Merged revisions 304005 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r304005 | rmudgett | 2011-01-25 17:21:09 -0600 (Tue, 25 Jan 2011) | 8 lines

      DTMF attended transfers sometimes fail for no apparent reason.

      The loop in feature_request_and_dial() can exit when Party C has answered
      without processing an AST_CONTROL_ANSWER.  Also sometimes an
      AST_CONTROL_ANSWER never happens even though Party C has answered.

      Don't hangup Party C if he is up or we receive an AST_CONTROL_ANSWER.
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@304008 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303962 via svnmerge from
Terry Wilson [Tue, 25 Jan 2011 22:15:41 +0000 (22:15 +0000)]
Merged revisions 303962 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303962 | twilson | 2011-01-25 16:09:01 -0600 (Tue, 25 Jan 2011) | 30 lines

  Merged revisions 303960 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303960 | twilson | 2011-01-25 16:02:42 -0600 (Tue, 25 Jan 2011) | 23 lines

    Merged revisions 303906 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303906 | twilson | 2011-01-25 14:50:59 -0600 (Tue, 25 Jan 2011) | 16 lines

      Guard against retransmitting BYEs indefinitely

      In the case of an attended transfer (A calls B, A atxfers to C) where
      A becomes unreachable before replying to Asterisk's BYE, Asterisk can
      sometimes retransmit the BYE indefinitely. This is because
      __sip_autodestruct tests p->refer && !ast_test_flag(&p->flags[0],
      SIP_ALREADYGONE and will then transmit a BYE. When this BYE times out,
      it will not ever be marked as ALREADYGONE, so when __sip_autodestruct
      is called again, we end up starting the cycle over.

      This patch adds a call to sip_alreadygone(pkt->owner) in retrans_pkt
      in the case of a BYE that has timed out. This should prevent Asterisk
      from trying to transmit new BYE messages in the future.

      Review: https://reviewboard.asterisk.org/r/1077/
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303963 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303860 via svnmerge from
Tilghman Lesher [Tue, 25 Jan 2011 18:56:23 +0000 (18:56 +0000)]
Merged revisions 303860 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303860 | tilghman | 2011-01-25 12:55:27 -0600 (Tue, 25 Jan 2011) | 12 lines

  Merged revisions 303858 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r303858 | tilghman | 2011-01-25 12:41:26 -0600 (Tue, 25 Jan 2011) | 5 lines

    Fix "sip show user <tab>", so that it actually shows results, instead of just completing the last entry.

    (closes issue #16675)
    Reported by: pj
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303861 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303771 via svnmerge from
Richard Mudgett [Tue, 25 Jan 2011 17:58:00 +0000 (17:58 +0000)]
Merged revisions 303771 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303771 | rmudgett | 2011-01-25 11:49:20 -0600 (Tue, 25 Jan 2011) | 54 lines

  Merged revisions 303769 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303769 | rmudgett | 2011-01-25 11:42:42 -0600 (Tue, 25 Jan 2011) | 47 lines

    Merged revisions 303765 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303765 | rmudgett | 2011-01-25 11:36:50 -0600 (Tue, 25 Jan 2011) | 40 lines

      Sending out unnecessary PROCEEDING messages breaks overlap dialing.

      Issue #16789 was a good idea.  Unfortunately, it breaks overlap dialing
      through Asterisk.  There is not enough information available at this point
      to know if dialing is complete.  The ast_exists_extension(),
      ast_matchmore_extension(), and ast_canmatch_extension() calls are not
      adequate to detect a dial through extension pattern of "_9!".

      Workaround is to use the dialplan Proceeding() application early in
      non-dial through extensions.

      * Effectively revert issue #16789.

      * Allow outgoing overlap dialing to hear dialtone and other early media.
      A PROGRESS "inband-information is now available" message is now sent after
      the SETUP_ACKNOWLEDGE message for non-digital calls.  An
      AST_CONTROL_PROGRESS is now generated for incoming SETUP_ACKNOWLEDGE
      messages for non-digital calls.

      * Handling of the AST_CONTROL_CONGESTION in chan_dahdi/sig_pri was
      inconsistent with the cause codes.

      * Added better protection from sending out of sequence messages by
      combining several flags into a single enum value representing call
      progress level.

      * Added diagnostic messages for deferred overlap digits handling corner
      cases.

      (closes issue #17085)
      Reported by: shawkris

      (closes issue #18509)
      Reported by: wimpy
      Patches:
            issue18509_early_media_v1.8_v3.patch uploaded by rmudgett (license 664)
            Expanded upon issue18509_early_media_v1.8_v3.patch to include analog
            and SS7 because of backporting requirements.
      Tested by: wimpy, rmudgett
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303772 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303678 via svnmerge from
Jeff Peeler [Tue, 25 Jan 2011 17:05:56 +0000 (17:05 +0000)]
Merged revisions 303678 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303678 | jpeeler | 2011-01-25 11:02:38 -0600 (Tue, 25 Jan 2011) | 33 lines

  Merged revisions 303677 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303677 | jpeeler | 2011-01-25 10:59:28 -0600 (Tue, 25 Jan 2011) | 26 lines

    Merged revisions 303676 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303676 | jpeeler | 2011-01-25 10:58:29 -0600 (Tue, 25 Jan 2011) | 20 lines

      Fix voicemail sequencing for file based storage.

      A previous change was made to account for when the number of voicemail messages
      exceeds the max limit to be handled properly, but it caused gaps in the messages
      to not be properly handled. This has now been resolved.

      In later non 1.4 branches, it appears that resequencing wasn't even occurring
      due from what appears and accidental code removal.

      (closes issue #18498)
      Reported by: JJCinAZ
      Patches:
            bug18498v2.patch uploaded by jpeeler (license 325)

      (closes issue #18486)
      Reported by: bluefox
      Patches:
            bug18486.patch uploaded by jpeeler (license 325)
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303679 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoUse unsigned char in comparison for UTF8 check to quiet a compiler warning.
Matthew Nicholson [Tue, 25 Jan 2011 15:52:42 +0000 (15:52 +0000)]
Use unsigned char in comparison for UTF8 check to quiet a compiler warning.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303638 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303549 via svnmerge from
Russell Bryant [Mon, 24 Jan 2011 20:57:28 +0000 (20:57 +0000)]
Merged revisions 303549 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303549 | russell | 2011-01-24 14:51:37 -0600 (Mon, 24 Jan 2011) | 45 lines

  Merged revisions 303548 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303548 | russell | 2011-01-24 14:49:53 -0600 (Mon, 24 Jan 2011) | 38 lines

    Merged revisions 303546 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303546 | russell | 2011-01-24 14:32:21 -0600 (Mon, 24 Jan 2011) | 31 lines

      Fix channel redirect out of MeetMe() and other issues with channel softhangup.

      Mantis issue #18585 reports that a channel redirect out of MeetMe() stopped
      working properly.  This issue includes a patch that resolves the issue by
      removing a call to ast_check_hangup() from app_meetme.c.  I left that in my
      patch, as it doesn't need to be there.  However, the rest of the patch fixes
      this problem with or without the change to app_meetme.

      The key difference between what happens before and after this patch is the
      effect of the END_OF_Q control frame.  After END_OF_Q is hit in ast_read(),
      ast_read() will return NULL.  With the ast_check_hangup() removed, app_meetme
      sees this which causes it to exit as intended.  Checking ast_check_hangup()
      caused app_meetme to exit earlier in the process, and the target of the
      redirect saw the condition where ast_read() returned NULL.

      Removing ast_check_hangup() works around the issue in app_meetme, but doesn't
      solve the issue if another application did the same thing.  There are also
      other edge cases where if an application finishes at the same time that a
      redirect happens, the target of the redirect will think that the channel hung
      up.  So, I made some changes in pbx.c to resolve it at a deeper level.  There
      are already places that unset the SOFTHANGUP_ASYNCGOTO flag in an attempt to
      abort the hangup process.  My patch extends this to remove the END_OF_Q frame
      from the channel's read queue, making the "abort hangup" more complete.  This
      same technique was used in every place where a softhangup flag was cleared.

      (closes issue #18585)
      Reported by: oej
      Tested by: oej, wedhorn, russell

      Review: https://reviewboard.asterisk.org/r/1082/
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303551 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd gsm-devel as a package to install on redhat based systems.
Russell Bryant [Mon, 24 Jan 2011 20:41:17 +0000 (20:41 +0000)]
Add gsm-devel as a package to install on redhat based systems.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303547 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAccording to section 19.1.2 of RFC 3261:
Matthew Nicholson [Mon, 24 Jan 2011 18:59:22 +0000 (18:59 +0000)]
According to section 19.1.2 of RFC 3261:

  For each component, the set of valid BNF expansions defines exactly
  which characters may appear unescaped.  All other characters MUST be
  escaped.

This patch modifies ast_uri_encode() to encode strings in line with this recommendation.  This patch also adds an ast_escape_quoted() function which escapes '"' and '\' characters in quoted strings in accordance with section 25.1 of RFC 3261.  The ast_uri_encode() function has also been modified to take an ast_flags struct describing the set of rules it should use when escaping characters to allow for it to escape SIP URIs in addition to HTTP URIs and other types of URIs or variations of those two URI types in the future.

The ast_uri_decode() function has also been modified to accept an ast_flags struct describing the set of rules to use when decoding to enable decoding '+' as ' ' in legacy http URLs.

The unit tests for these functions have also been updated.

ABE-2705

Review: https://reviewboard.asterisk.org/r/1081/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303509 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303467 via svnmerge from
Jason Parker [Mon, 24 Jan 2011 17:21:12 +0000 (17:21 +0000)]
Merged revisions 303467 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303467 | qwell | 2011-01-24 11:20:03 -0600 (Mon, 24 Jan 2011) | 22 lines

  Merged revisions 303285 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303285 | qwell | 2011-01-21 15:48:09 -0600 (Fri, 21 Jan 2011) | 15 lines

    Merged revisions 303284 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303284 | qwell | 2011-01-21 15:45:34 -0600 (Fri, 21 Jan 2011) | 8 lines

      Reset configuration before parsing users.conf.

      Some values configured in chan_dahdi.conf were able to leak in to users.conf
      configuration.  This was surprising users, and potentially setting non-sane
      "defaults".

      ASTNOW-125
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303468 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoRevert default compiler change.
Russell Bryant [Sat, 22 Jan 2011 04:13:15 +0000 (04:13 +0000)]
Revert default compiler change.

If someone wishes to do so, it is trivial to set your own default when running
the configure script.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303418 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoTemporarily revert r303288
Jason Parker [Fri, 21 Jan 2011 23:11:34 +0000 (23:11 +0000)]
Temporarily revert r303288

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303376 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303286 via svnmerge from
Jason Parker [Fri, 21 Jan 2011 21:51:06 +0000 (21:51 +0000)]
Merged revisions 303286 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303286 | qwell | 2011-01-21 15:50:11 -0600 (Fri, 21 Jan 2011) | 22 lines

  Merged revisions 303285 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303285 | qwell | 2011-01-21 15:48:09 -0600 (Fri, 21 Jan 2011) | 15 lines

    Merged revisions 303284 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303284 | qwell | 2011-01-21 15:45:34 -0600 (Fri, 21 Jan 2011) | 8 lines

      Reset configuration before parsing users.conf.

      Some values configured in chan_dahdi.conf were able to leak in to users.conf
      configuration.  This was surprising users, and potentially setting non-sane
      "defaults".

      ASTNOW-125
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303288 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoReally use llvm-gcc, when available.
Tilghman Lesher [Fri, 21 Jan 2011 09:09:54 +0000 (09:09 +0000)]
Really use llvm-gcc, when available.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303235 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd DB_KEYS.
Tilghman Lesher [Fri, 21 Jan 2011 08:13:18 +0000 (08:13 +0000)]
Add DB_KEYS.

Discussion on #asterisk on 2011-01-19:
(02:07:03 PM) boch: i wonder how to cycle all entries in a tree
(02:07:11 PM) leifmadsen: use While()
(02:07:17 PM) leifmadsen: you need to know the tree structure already though
(02:07:36 PM) boch: what you mean?
(02:09:02 PM) leifmadsen: you need to know the structure prior to looping, because you can't just return the structure from the dialplan
(02:09:43 PM) leifmadsen: the only way I can think of doing that is via something like writing the output of:  asterisk -rx "database show" to a file, then looping through that to know the structure of the database and check everything
(02:09:59 PM) leifmadsen: but at that point you're better off just using either a relational database or an external script
(02:10:13 PM) boch: for example i need to know all entries in the tree
(02:10:15 PM) boch: got it
(02:10:20 PM) leifmadsen: exactly
(02:10:22 PM) leifmadsen: that's the problem
(02:10:22 PM) boch: thank you
(02:13:09 PM) mateu: yeah, i'm surprised there isn't something from the dialplan like 'database show family' so one can get all keys in a family to loop over.
(02:15:35 PM) leifmadsen: database shows everything
(02:16:22 PM) mateu: i mean something from the dial plan that mimics 'database show <family>'
(02:16:41 PM) leifmadsen: guess no one has found that important enough to program :)
(02:16:52 PM) leifmadsen: at that point you should probably just use a relational database...
(02:17:10 PM) mateu: i dunno
(02:17:16 PM) mateu: seems pretty basic to me.
(02:17:16 PM) leifmadsen: me either
(02:17:19 PM) leifmadsen: sure does
(02:17:24 PM) leifmadsen: no one has programmed it though
(02:17:28 PM) ***leifmadsen shrugs
(02:17:43 PM) mateu: ok, well at least we know how it currently stands.  thanks leifmadsen
(02:28:52 PM) Corydon76-home: leifmadsen: something like HASHKEYS() ?
(02:30:11 PM) leifmadsen: Corydon76-home: ummm, I was thinking more like DUNDI_QUERY() and DUNDI_RESULT()
(02:30:31 PM) leifmadsen: although HASHKEYS() might work
(02:30:58 PM) leifmadsen: actually ya, looking at it, similar to HASHKEYS()
(02:31:01 PM) leifmadsen: DBKEYS() I guess?
(02:31:45 PM) Corydon76-home: So with no argument, retrieves families, with an argument, retrieves keys of that family?
(02:34:02 PM) leifmadsen: ya
(02:34:16 PM) leifmadsen: how would you iterate through layers of them?
(02:34:30 PM) leifmadsen: i.e. family/key/key/key ?
(02:34:43 PM) Corydon76-home: Essentially, yes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303198 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303153 via svnmerge from
Richard Mudgett [Thu, 20 Jan 2011 20:35:50 +0000 (20:35 +0000)]
Merged revisions 303153 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303153 | rmudgett | 2011-01-20 14:31:20 -0600 (Thu, 20 Jan 2011) | 22 lines

  Merged revision 303098 from
  https://origsvn.digium.com/svn/asterisk/be/branches/C.3-bier

  ..........
    r303098 | rmudgett | 2011-01-20 12:11:45 -0600 (Thu, 20 Jan 2011) | 15 lines

    CC_INTERFACES does not get built correctly with local channels.

    If local channels are used with CCSS, CC_INTERFACES gets garbage prepended
    to it so the CC recall fails.  Also CC_INTERFACES gets "&(null)" appended
    to it.

    * Initialize the buffer to eliminate the prepended garbage.

    * Filter out the empty interface strings to eliminate the latter.

    * Added a diagnostic message if the CC_INTERFACES is ever empty.

    JIRA ABE-2740
    JIRA SWP-2848
  ..........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303154 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303107 via svnmerge from
Shaun Ruffell [Thu, 20 Jan 2011 19:58:54 +0000 (19:58 +0000)]
Merged revisions 303107 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303107 | sruffell | 2011-01-20 13:57:31 -0600 (Thu, 20 Jan 2011) | 23 lines

  Merged revisions 303106 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r303106 | sruffell | 2011-01-20 13:56:34 -0600 (Thu, 20 Jan 2011) | 15 lines

    main/features: Use POLLPRI when waiting for events on parked channels.

    This change resolves a regression in the 1.6.2 when converting from
    select to poll.  The DAHDI timers use POLLPRI to indicate that the timer
    fired, but features was not waiting for that flag.  The result was no
    audio for MOH when a call was parked and res_timing_dahdi was in use.

    This patch is slightly modified from the one on the mantis issue.  It does
    not set an exception on the channel if the POLLPRI flag is set.

    (closes issue #18262)
    Reported by: francesco_r
    Patches:
          patch_park_moh-trunk-2.txt uploaded by cjacobsen (license 1029)
          Tested by: francesco_r, rfrantik, one47
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303108 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 303009 via svnmerge from
Jeff Peeler [Thu, 20 Jan 2011 17:14:01 +0000 (17:14 +0000)]
Merged revisions 303009 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r303009 | jpeeler | 2011-01-20 11:10:32 -0600 (Thu, 20 Jan 2011) | 21 lines

  Merged revisions 303008 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r303008 | jpeeler | 2011-01-20 11:07:44 -0600 (Thu, 20 Jan 2011) | 14 lines

    Merged revisions 303007 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r303007 | jpeeler | 2011-01-20 11:04:08 -0600 (Thu, 20 Jan 2011) | 8 lines

      Add new queue strategy to preserve behavior for when queue members moved to ao2.

      Add queue strategy called "rrordered" to mimic old behavior from when queue
      members were stored in a linked list.

      ABE-2707
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@303011 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302921 via svnmerge from
Russell Bryant [Thu, 20 Jan 2011 16:12:35 +0000 (16:12 +0000)]
Merged revisions 302921 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302921 | russell | 2011-01-20 10:12:15 -0600 (Thu, 20 Jan 2011) | 9 lines

  Merged revisions 302920 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302920 | russell | 2011-01-20 10:11:58 -0600 (Thu, 20 Jan 2011) | 2 lines

    Resolve a compiler warning.
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302922 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302918 via svnmerge from
Leif Madsen [Thu, 20 Jan 2011 15:46:24 +0000 (15:46 +0000)]
Merged revisions 302918 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302918 | lmadsen | 2011-01-20 09:45:39 -0600 (Thu, 20 Jan 2011) | 16 lines

  Merged revisions 302917 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302917 | lmadsen | 2011-01-20 09:42:05 -0600 (Thu, 20 Jan 2011) | 8 lines

    Option L() is milliseconds, not seconds.
    > Change the verbose output of option L() to say milliseconds and not seconds
    > as the value is in milliseconds.
    >
    > (closes issue #18264)
    > Reported by: jacco
    > Patches:
    >       app_dial_patch.txt uploaded by lmadsen (license 10)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302919 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoOn systems which have LLVM, use that compiler. Should result in a massive speed...
Tilghman Lesher [Thu, 20 Jan 2011 09:07:27 +0000 (09:07 +0000)]
On systems which have LLVM, use that compiler.  Should result in a massive speed increase.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302879 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302837 via svnmerge from
Russell Bryant [Wed, 19 Jan 2011 23:57:27 +0000 (23:57 +0000)]
Merged revisions 302837 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302837 | russell | 2011-01-19 17:56:48 -0600 (Wed, 19 Jan 2011) | 2 lines

  Only check container count if it exists.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302838 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoClarify a source comment about configuration template categories.
Sean Bright [Wed, 19 Jan 2011 23:53:44 +0000 (23:53 +0000)]
Clarify a source comment about configuration template categories.

(closes issue #18578)
Reported by: astmiv
Patches:
      asterisk.main.config.2.patch uploaded by astmiv (license 1189)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302836 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302834 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 23:49:54 +0000 (23:49 +0000)]
Merged revisions 302834 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302834 | seanbright | 2011-01-19 18:49:00 -0500 (Wed, 19 Jan 2011) | 14 lines

  Merged revisions 302833 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302833 | seanbright | 2011-01-19 18:47:22 -0500 (Wed, 19 Jan 2011) | 7 lines

    Support greetingsfolder as documented in voicemail.conf.sample.

    (closes issue #17870)
    Reported by: edhorton
    Patches:
          __20100816-app_voicemail-greetingsfolder-support.txt uploaded by lmadsen (license 10)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302835 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302831 via svnmerge from
Paul Belanger [Wed, 19 Jan 2011 23:33:42 +0000 (23:33 +0000)]
Merged revisions 302831 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302831 | pabelanger | 2011-01-19 18:29:45 -0500 (Wed, 19 Jan 2011) | 2 lines

  Add binutils-dev for BETTER_BACKTRACES
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302832 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302789 via svnmerge from
Russell Bryant [Wed, 19 Jan 2011 23:07:22 +0000 (23:07 +0000)]
Merged revisions 302789 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302789 | russell | 2011-01-19 17:06:46 -0600 (Wed, 19 Jan 2011) | 11 lines

  Merged revisions 302788 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302788 | russell | 2011-01-19 17:06:14 -0600 (Wed, 19 Jan 2011) | 4 lines

    Turn a noisy verbose message into a debug message.

    This can drown your console if you're using the AMI over HTTP.
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302790 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302785 via svnmerge from
Russell Bryant [Wed, 19 Jan 2011 22:36:30 +0000 (22:36 +0000)]
Merged revisions 302785 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302785 | russell | 2011-01-19 16:35:15 -0600 (Wed, 19 Jan 2011) | 15 lines

  Resolve a memory leak with the manager interface is disabled.

  The intent of this check as it stands in previous versions of Asterisk was to
  check if there are any active sessions.  If there were no sessions, then the
  function would return immediately and not bother with queueing up the manager
  event to be processed.  Since the conversion of storing sessions in an astobj2
  container, this check will always pass.  I changed it to go back to checking
  what was intended.

  The side effect of this was that if the AMI is disabled, the manager event
  queue is populated anyway, but the code that runs to clear out the queue
  never runs.  A producer with no consumer is a bad thing.

  Reported internally by kmorgan.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302786 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302713 via svnmerge from
Richard Mudgett [Wed, 19 Jan 2011 21:35:28 +0000 (21:35 +0000)]
Merged revisions 302713 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302713 | rmudgett | 2011-01-19 15:29:22 -0600 (Wed, 19 Jan 2011) | 29 lines

  Merged revisions 302693 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r302693 | rmudgett | 2011-01-19 15:25:41 -0600 (Wed, 19 Jan 2011) | 22 lines

    Merged revisions 302671 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r302671 | rmudgett | 2011-01-19 15:21:56 -0600 (Wed, 19 Jan 2011) | 15 lines

      DTMF transfer plays the wrong sounds for wrong number or other call failure.

      * Set the default for features.conf.sample xferfailsound option to "beeperr"
      as documented instead of "pbx-invalid" and corrected the use of it in DTMF
      blind transfer (#1).

      * Improved DTMF blind transfer handling of wrong numbers.

      Most of the concerns in this issue were taken care of by the patch for
      issue 17999: Issues with DTMF triggered attended transfers.

      (closes issue #18379)
      Reported by: gincantalupo
      Tested by: rmudgett
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302732 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302680 via svnmerge from
Tilghman Lesher [Wed, 19 Jan 2011 21:24:25 +0000 (21:24 +0000)]
Merged revisions 302680 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302680 | tilghman | 2011-01-19 15:23:31 -0600 (Wed, 19 Jan 2011) | 16 lines

  Merged revisions 302675 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r302675 | tilghman | 2011-01-19 15:22:45 -0600 (Wed, 19 Jan 2011) | 9 lines

    Merged revisions 302663 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r302663 | tilghman | 2011-01-19 15:20:28 -0600 (Wed, 19 Jan 2011) | 2 lines

      Add some API documentation
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302686 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302634 via svnmerge from
Tilghman Lesher [Wed, 19 Jan 2011 20:33:30 +0000 (20:33 +0000)]
Merged revisions 302634 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302634 | tilghman | 2011-01-19 14:24:57 -0600 (Wed, 19 Jan 2011) | 22 lines

  Merged revisions 302599 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302599 | tilghman | 2011-01-19 14:13:24 -0600 (Wed, 19 Jan 2011) | 15 lines

    Kill zombies.

    When we ast_safe_fork() with a non-zero argument, we're expected to reap our
    own zombies.  On a zero argument, however, the zombies are only reaped when
    there aren't any non-zero forked children alive.  At other times, we
    accumulate zombies.  This code is forward ported from res_agi in 1.4, so that
    forked children are always reaped, thus preventing an accumulation of zombie
    processes.

    (closes issue #18515)
    Reported by: ernied
    Patches:
          20101221__issue18515.diff.txt uploaded by tilghman (license 14)
    Tested by: ernied
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302644 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302600 via svnmerge from
Jason Parker [Wed, 19 Jan 2011 20:15:54 +0000 (20:15 +0000)]
Merged revisions 302600 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302600 | qwell | 2011-01-19 14:14:40 -0600 (Wed, 19 Jan 2011) | 1 line

  Fix typo pointed out on asterisk-users list.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302601 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302555 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 19:04:25 +0000 (19:04 +0000)]
Merged revisions 302555 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302555 | seanbright | 2011-01-19 14:03:32 -0500 (Wed, 19 Jan 2011) | 14 lines

  Merged revisions 302554 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302554 | seanbright | 2011-01-19 14:02:29 -0500 (Wed, 19 Jan 2011) | 7 lines

    Don't call strlen() when we only need to look at the next character or two.

    (closes issue #18042)
    Reported by: wdoekes
    Patches:
          astsvn-inefficient-ast-uri-decode.patch uploaded by wdoekes (license 717)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302556 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302552 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 18:55:43 +0000 (18:55 +0000)]
Merged revisions 302552 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302552 | seanbright | 2011-01-19 13:54:47 -0500 (Wed, 19 Jan 2011) | 14 lines

  Merged revisions 302551 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302551 | seanbright | 2011-01-19 13:54:03 -0500 (Wed, 19 Jan 2011) | 7 lines

    Remove an extraneous \r\n at the end of a parking manager events.

    (closes issue #18363)
    Reported by: clegall_proformatique
    Patches:
          asterisk_1.8_295998_parking_manager_events_format.patch uploaded by clegall proformatique (license 1139)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302553 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302549 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 18:45:44 +0000 (18:45 +0000)]
Merged revisions 302549 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302549 | seanbright | 2011-01-19 13:43:11 -0500 (Wed, 19 Jan 2011) | 17 lines

  Merged revisions 302548 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302548 | seanbright | 2011-01-19 13:37:09 -0500 (Wed, 19 Jan 2011) | 10 lines

    Properly handle partial reads from fgets() when handling AGIs.

    When fgets() failed with EAGAIN, we were continually decrementing the available
    space left in our buffer, resulting in botched command handling.

    (closes issue #16032)
    Reported by: notahat
    Patches:
          agi_buffer_patch2.diff uploaded by fnordian (license 110)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302550 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302505 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 17:59:18 +0000 (17:59 +0000)]
Merged revisions 302505 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302505 | seanbright | 2011-01-19 12:58:11 -0500 (Wed, 19 Jan 2011) | 14 lines

  Merged revisions 302504 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302504 | seanbright | 2011-01-19 12:56:32 -0500 (Wed, 19 Jan 2011) | 7 lines

    Make sure that h_length is set when we short-circuit out of ast_gethostbyname.

    (closes issue #16135)
    Reported by: thedavidfactor
    Patches:
          utils.patch uploaded by thedavidfactor (license 903)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302507 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302462 via svnmerge from
Paul Belanger [Wed, 19 Jan 2011 17:15:40 +0000 (17:15 +0000)]
Merged revisions 302462 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302462 | pabelanger | 2011-01-19 12:09:35 -0500 (Wed, 19 Jan 2011) | 9 lines

  Merged revisions 302461 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302461 | pabelanger | 2011-01-19 12:08:01 -0500 (Wed, 19 Jan 2011) | 2 lines

    Handle 'Resource temporarily unavailable' error more gracefully.
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302463 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302417 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 15:54:22 +0000 (15:54 +0000)]
Merged revisions 302417 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302417 | seanbright | 2011-01-19 10:53:20 -0500 (Wed, 19 Jan 2011) | 16 lines

  Merged revisions 302416 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302416 | seanbright | 2011-01-19 10:52:44 -0500 (Wed, 19 Jan 2011) | 9 lines

    Remove references to priorityjumping from the sample extensions.conf.

    Priority jumping was removed from pbx_config in r68970.

    (closes issue #18622)
    Reported by: kshumard
    Patches:
          extensions.conf.sample.patch uploaded by kshumard (license 92)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302418 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302414 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 15:46:56 +0000 (15:46 +0000)]
Merged revisions 302414 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302414 | seanbright | 2011-01-19 10:45:17 -0500 (Wed, 19 Jan 2011) | 7 lines

  Initialize an uninitialized variable.

  (closes issue #18640)
  Reported by: jcovert
  Patches:
        chan_sip.c.patch uploaded by jcovert (license 551)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302415 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302412 via svnmerge from
Sean Bright [Wed, 19 Jan 2011 15:34:07 +0000 (15:34 +0000)]
Merged revisions 302412 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302412 | seanbright | 2011-01-19 10:31:39 -0500 (Wed, 19 Jan 2011) | 10 lines

  Use appropriate type for requested format in chan_local.

  We were passing and storing the requested format as an int instead of format_t
  resulting in truncation.

  (closes issue #18238)
  Reported by: whizemen
  Patches:
        0018238_speex16.patch uploaded by whizemen (license 1143)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302413 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302318 via svnmerge from
Richard Mudgett [Tue, 18 Jan 2011 22:06:55 +0000 (22:06 +0000)]
Merged revisions 302318 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302318 | rmudgett | 2011-01-18 16:04:14 -0600 (Tue, 18 Jan 2011) | 1 line

  Use the expanded format type instead of plain int.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302319 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302314 via svnmerge from
Matthew Nicholson [Tue, 18 Jan 2011 21:44:49 +0000 (21:44 +0000)]
Merged revisions 302314 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302314 | mnicholson | 2011-01-18 15:43:21 -0600 (Tue, 18 Jan 2011) | 18 lines

  Merged revisions 302313 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r302313 | mnicholson | 2011-01-18 15:40:03 -0600 (Tue, 18 Jan 2011) | 11 lines

    Merged revisions 302311 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r302311 | mnicholson | 2011-01-18 15:35:03 -0600 (Tue, 18 Jan 2011) | 4 lines

      URI encode the user part of the contact header.

      ABE-2705
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302315 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302266 via svnmerge from
Jeff Peeler [Tue, 18 Jan 2011 20:40:59 +0000 (20:40 +0000)]
Merged revisions 302266 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302266 | jpeeler | 2011-01-18 14:19:57 -0600 (Tue, 18 Jan 2011) | 34 lines

  Merged revisions 302265 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r302265 | jpeeler | 2011-01-18 14:13:52 -0600 (Tue, 18 Jan 2011) | 27 lines

    Convert device state callbacks to ao2 objects to fix a deadlock in chan_sip.

    Lock scenario presented here:
    Thread 1
     holds ast_rdlock_contexts &conlock
     holds handle_statechange hints
     holds handle_statechange hint
      waiting for cb_extensionstate
       Locked Here: chan_sip.c line 7428 (find_call)
    Thread 2
     holds handle_request_do &netlock
     holds find_call sip_pvt_ptr
      waiting for ast_rdlock_contexts &conlock
       Locked Here: pbx.c line 9911 (ast_rdlock_contexts)

    Chan_sip has an established locking order of locking the sip_pvt and then
    getting the context lock. So the as stated by the summary, the operations in
    thread 2 have been modified to no longer require the context lock.

    (closes issue #18310)
    Reported by: one47
    Patches:
          statecbs_ao2.mk2.patch uploaded by one47 (license 23),
          modified by me

    Review: https://reviewboard.asterisk.org/r/1072/
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302270 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302267 via svnmerge from
Russell Bryant [Tue, 18 Jan 2011 20:21:29 +0000 (20:21 +0000)]
Merged revisions 302267 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302267 | russell | 2011-01-18 14:19:57 -0600 (Tue, 18 Jan 2011) | 5 lines

  Don't enable AO2_DEBUG by default if AST_DEVMODE is on.

  AO2_DEBUG is not important and is causing a false compiler warning to be
  generated on my Ubuntu Natty dev box.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302268 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302174 via svnmerge from
Richard Mudgett [Tue, 18 Jan 2011 18:17:01 +0000 (18:17 +0000)]
Merged revisions 302174 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r302174 | rmudgett | 2011-01-18 12:11:43 -0600 (Tue, 18 Jan 2011) | 102 lines

  Merged revisions 302173 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r302173 | rmudgett | 2011-01-18 12:07:15 -0600 (Tue, 18 Jan 2011) | 95 lines

    Merged revisions 302172 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r302172 | rmudgett | 2011-01-18 12:04:36 -0600 (Tue, 18 Jan 2011) | 88 lines

      Issues with DTMF triggered attended transfers.

      Issue #17999
      1) A calls B. B answers.
      2) B using DTMF dial *2 (code in features.conf for attended transfer).
      3) A hears MOH. B dial number C
      4) C ringing. A hears MOH.
      5) B hangup. A still hears MOH. C ringing.
      6) A hangup. C still ringing until "atxfernoanswertimeout" expires.
      For v1.4 C will ring forever until C answers the dead line. (Issue #17096)

      Problem: When A and B hangup, C is still ringing.

      Issue #18395
      SIP call limit of B is 1
      1. A call B, B answered
      2. B *2(atxfer) call C
      3. B hangup, C ringing
      4. Timeout waiting for C to answer
      5. Recall to B fails because B has reached its call limit.

      Because B reached its call limit, it cannot do anything until the transfer
      it started completes.

      Issue #17273
      Same scenario as issue 18395 but party B is an FXS port.  Party B cannot
      do anything until the transfer it started completes.  If B goes back off
      hook before C answers, B hears ringback instead of the expected dialtone.

      **********
      Note for the issue #17273 and #18395 fix:

      DTMF attended transfer works within the channel bridge.  Unfortunately,
      when either party A or B in the channel bridge hangs up, that channel is
      not completely hung up until the transfer completes.  This is a real
      problem depending upon the channel technology involved.

      For chan_dahdi, the channel is crippled until the hangup is complete.
      Either the channel is not useable (analog) or the protocol disconnect
      messages are held up (PRI/BRI/SS7) and the media is not released.

      For chan_sip, a call limit of one is going to block that endpoint from any
      further calls until the hangup is complete.

      For party A this is a minor problem.  The party A channel will only be in
      this condition while party B is dialing and when party B and C are
      conferring.  The conversation between party B and C is expected to be a
      short one.  Party B is either asking a question of party C or announcing
      party A.  Also party A does not have much incentive to hangup at this
      point.

      For party B this can be a major problem during a blonde transfer.  (A
      blonde transfer is our term for an attended transfer that is converted
      into a blind transfer.  :)) Party B could be the operator.  When party B
      hangs up, he assumes that he is out of the original call entirely.  The
      party B channel will be in this condition while party C is ringing, while
      attempting to recall party B, and while waiting between call attempts.

      WARNING:
      The ATXFER_NULL_TECH conditional is a hack to fix the problem.  It will
      replace the party B channel technology with a NULL channel driver to
      complete hanging up the party B channel technology.  The consequences of
      this code is that the 'h' extension will not be able to access any channel
      technology specific information like SIP statistics for the call.

      ATXFER_NULL_TECH is not defined by default.
      **********

      (closes issue #17999)
      Reported by: iskatel
      Tested by: rmudgett
      JIRA SWP-2246

      (closes issue #17096)
      Reported by: gelo
      Tested by: rmudgett
      JIRA SWP-1192

      (closes issue #18395)
      Reported by: shihchuan
      Tested by: rmudgett

      (closes issue #17273)
      Reported by: grecco
      Tested by: rmudgett

      Review: https://reviewboard.asterisk.org/r/1047/
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302178 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 293493 via svnmerge from
Terry Wilson [Mon, 17 Jan 2011 16:38:21 +0000 (16:38 +0000)]
Merged revisions 293493 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r293493 | twilson | 2010-11-01 09:58:00 -0500 (Mon, 01 Nov 2010) | 14 lines

  Only offer codecs both sides support for directmedia

  When using directmedia, Asterisk needs to limit the codecs offered to just
  the ones that both sides recognize, otherwise they may end up sending audio
  that the other side doesn't understand.

  (closes issue #17403)
  Reported by: one47
  Patches:
        sip_codecs_simplified4 uploaded by one47 (license 23)
  Tested by: one47, falves11

  Review: https://reviewboard.asterisk.org/r/967/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302048 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 302005 via svnmerge from
Terry Wilson [Mon, 17 Jan 2011 15:06:10 +0000 (15:06 +0000)]
Merged revisions 302005 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r302005 | twilson | 2011-01-17 09:04:59 -0600 (Mon, 17 Jan 2011) | 2 lines

  Document "encryption" option in sip.conf.sample
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@302006 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301946 via svnmerge from
Richard Mudgett [Fri, 14 Jan 2011 21:13:08 +0000 (21:13 +0000)]
Merged revisions 301946 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301946 | rmudgett | 2011-01-14 15:09:57 -0600 (Fri, 14 Jan 2011) | 13 lines

  Deadlock between dahdi_request() and pri_dchannel() processing an incomming call.

  The sig_pri_new_ast_channel() is called with the channel private lock held
  when pri_dchannel() calls it and no channel private lock held when
  dahdi_request() calls it.  The use of pri_grab() in
  sig_pri_new_ast_channel() could leave the channel private lock held when
  it returns if the lock was not held before calling it.

  Make sig_pri_new_ast_channel() just lock the PRI span lock instead of
  using pri_grab().  It is safe to do this because dahdi_request() does not
  have the channel private lock and the deadlock potential with the PRI span
  lock is only between pri_dchannel() and other threads.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301947 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301851 via svnmerge from
Brett Bryant [Fri, 14 Jan 2011 20:18:26 +0000 (20:18 +0000)]
Merged revisions 301851 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301851 | bbryant | 2011-01-14 15:11:55 -0500 (Fri, 14 Jan 2011) | 6 lines

  Changing previous revisions 301845/301847 to use ast_sockaddr_setnull() instead
  of setting the field manually to avoid uninitialized data.

  Review: https://reviewboard.asterisk.org/r/1076/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301858 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd relationships to function documentation.
Andrew Latham [Fri, 14 Jan 2011 20:07:02 +0000 (20:07 +0000)]
Add relationships to function documentation.

Fix amatuer type mistake

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301850 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301845 via svnmerge from
Brett Bryant [Fri, 14 Jan 2011 19:44:11 +0000 (19:44 +0000)]
Merged revisions 301845 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301845 | bbryant | 2011-01-14 14:35:23 -0500 (Fri, 14 Jan 2011) | 9 lines

  Fix for a consistent MulticastRTP channel driver crash due to use of unitilized
  data.

  (closes issue #18290)
  (closes issue #18602)
  Reported by: voipgate, wybecom

  Review: https://reviewboard.asterisk.org/r/1076/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301847 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd relationships to function documentation.
Andrew Latham [Fri, 14 Jan 2011 19:39:22 +0000 (19:39 +0000)]
Add relationships to function documentation.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301846 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301790 via svnmerge from
Jeff Peeler [Fri, 14 Jan 2011 17:34:28 +0000 (17:34 +0000)]
Merged revisions 301790 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301790 | jpeeler | 2011-01-14 11:32:52 -0600 (Fri, 14 Jan 2011) | 42 lines

  Resolve deadlock involving REFER.

  Two fixes:
  1) One must always have the private unlocked before calling
  pbx_builtin_setvar_helper to not invalidate locking order since it locks the
  channel.
  2) Unlock the channel before calling pbx_find_extension, which starts and stops
  autoservice during the lookup. The problem scenario as illustrated by the
  reporter:

  Thread: do_monitor
  -----------------------
  handle_request_do
   handle_incoming
    handle_request_refer
     ast_parking_ext_valid
      pbx_find_extension
       ast_autoservice_stop
        while (chan_list_state == as_chan_list_state) { usleep(1000); }

  Thread: autoservice_run
  -----------------------
  autoservice_run
   chan = ast_waitfor_n
    ast_waitfor_nandfds
     ast_waitfor_nandfds_classic / simple / complex (depending on your system)
      ast_channel_lock(c[x]);

  handle_request_do and schedule_process_request_queue locks the owner
  if it exists. The autoservice thread is waiting for the channel lock, which
  wasn't ever released since the do_monitor thread was waiting for autoservice
  operations to complete. Solved by unlocking the channel but keeping a reference
  to guarantee safety.

  (closes issue #18403)
  Reported by: jthurman
  Patches:
        20110103-blind_deadlock.diff uploaded by jthurman (license 614)
        issue18403.patch uploaded by jpeeler (license 325)
  Tested by: jthurman
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301791 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301731 via svnmerge from
Leif Madsen [Thu, 13 Jan 2011 17:02:34 +0000 (17:02 +0000)]
Merged revisions 301731 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301731 | lmadsen | 2011-01-13 11:01:43 -0600 (Thu, 13 Jan 2011) | 15 lines

  Merged revisions 301730 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301730 | lmadsen | 2011-01-13 11:01:11 -0600 (Thu, 13 Jan 2011) | 7 lines

    Add static entry for split Polycom 332 firmware.

    (closes issue #18607)
    Reported by: cjacobsen
    Patches:
          polycom_331.diff uploaded by cjacobsen (license 1029)
    Tested by: lathama
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301732 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdd dialplan variables for asterisk.conf directories
Paul Belanger [Thu, 13 Jan 2011 16:27:22 +0000 (16:27 +0000)]
Add dialplan variables for asterisk.conf directories

Review: https://reviewboard.asterisk.org/r/1075/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301729 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301683 via svnmerge from
Terry Wilson [Wed, 12 Jan 2011 21:24:18 +0000 (21:24 +0000)]
Merged revisions 301683 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301683 | twilson | 2011-01-12 15:19:48 -0600 (Wed, 12 Jan 2011) | 15 lines

  Merged revisions 301682 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301682 | twilson | 2011-01-12 15:05:02 -0600 (Wed, 12 Jan 2011) | 9 lines

    Don't reject all SUBSCRIBE auth requests

    When merging another SUBSCRIBE fix from 1.4, some braces were put in
    the wrong place. This patch fixes that.

    (closes issue #18597)
    Reported by: thsgmbh
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301684 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301595 via svnmerge from
Matthew Nicholson [Wed, 12 Jan 2011 18:52:30 +0000 (18:52 +0000)]
Merged revisions 301595 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301595 | mnicholson | 2011-01-12 12:51:37 -0600 (Wed, 12 Jan 2011) | 22 lines

  Merged revisions 301594 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r301594 | mnicholson | 2011-01-12 12:50:31 -0600 (Wed, 12 Jan 2011) | 15 lines

    Removed a usleep(1) that shouldn't be necessary in session_do, and removed the
    ms_t member from the mansession_session structure.

    Merged revisions 301591 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r301591 | mnicholson | 2011-01-12 12:39:03 -0600 (Wed, 12 Jan 2011) | 5 lines

      Don't store the thread id for the manager session in the structure we pass to
      the thread for the manager session.

      ABE-2543
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301596 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301504 via svnmerge from
Jeff Peeler [Wed, 12 Jan 2011 18:12:31 +0000 (18:12 +0000)]
Merged revisions 301504 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301504 | jpeeler | 2011-01-12 12:12:08 -0600 (Wed, 12 Jan 2011) | 26 lines

  Merged revisions 301503 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r301503 | jpeeler | 2011-01-12 12:11:49 -0600 (Wed, 12 Jan 2011) | 19 lines

    Merged revisions 301502 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r301502 | jpeeler | 2011-01-12 12:10:42 -0600 (Wed, 12 Jan 2011) | 12 lines

      Fix CPU spike when pressing DTMF after agent login.

      The problem here is that DTMF was being continuously deferred and requeued
      since ast_safe_sleep is called in a loop. There are serveral other places in the
      code that sleeps and then loops in a similar fashion. Because of this fact I
      opted to not defer DTMF any more, which will not affect the original fix:

      https://reviewboard.asterisk.org/r/674

      (closes issue #18130)
      Reported by: rgj
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301505 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301446 via svnmerge from
David Vossel [Wed, 12 Jan 2011 16:05:58 +0000 (16:05 +0000)]
Merged revisions 301446 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301446 | dvossel | 2011-01-12 10:05:12 -0600 (Wed, 12 Jan 2011) | 2 lines

  Removal of unused variables so Asterisk will compile.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301447 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agofix wrong text of rerun menuselect after user interface warning
Stefan Schmidt [Wed, 12 Jan 2011 15:59:28 +0000 (15:59 +0000)]
fix wrong text of rerun menuselect after user interface warning

the warning, if no user interface for menuselect warning was found is not right.
you have to rerun configure before make menuselect after installing a proper user interface.

(closes issue 0018594)
Reported by: Dovid

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301445 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301402 via svnmerge from
Tilghman Lesher [Wed, 12 Jan 2011 00:27:30 +0000 (00:27 +0000)]
Merged revisions 301402 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301402 | tilghman | 2011-01-11 18:26:39 -0600 (Tue, 11 Jan 2011) | 7 lines

  Call execl() directly for a better solution for paths with spaces.

  (closes issue #18600)
  Reported by: ebroad
  Patches:
        20110111__issue18600__2.diff.txt uploaded by tilghman (license 14)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301403 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301311 via svnmerge from
Paul Belanger [Tue, 11 Jan 2011 19:19:01 +0000 (19:19 +0000)]
Merged revisions 301311 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301311 | pabelanger | 2011-01-11 14:16:06 -0500 (Tue, 11 Jan 2011) | 9 lines

  Merged revisions 301310 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301310 | pabelanger | 2011-01-11 14:14:31 -0500 (Tue, 11 Jan 2011) | 2 lines

    Fix a logic issue when passing context ARG
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301319 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301308 via svnmerge from
Matthew Nicholson [Tue, 11 Jan 2011 18:55:16 +0000 (18:55 +0000)]
Merged revisions 301308 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301308 | mnicholson | 2011-01-11 12:51:40 -0600 (Tue, 11 Jan 2011) | 18 lines

  Merged revisions 301307 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r301307 | mnicholson | 2011-01-11 12:42:05 -0600 (Tue, 11 Jan 2011) | 11 lines

    Merged revisions 301305 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r301305 | mnicholson | 2011-01-11 12:34:40 -0600 (Tue, 11 Jan 2011) | 4 lines

      Prevent buffer overflows in ast_uri_encode()

      ABE-2705
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301309 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301263 via svnmerge from
Tilghman Lesher [Mon, 10 Jan 2011 22:40:23 +0000 (22:40 +0000)]
Merged revisions 301263 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301263 | tilghman | 2011-01-10 16:39:31 -0600 (Mon, 10 Jan 2011) | 8 lines

  Little endian machines were not converted properly.

  (closes issue #18583)
  Reported by: jcovert
  Patches:
        20110110__issue18583.diff.txt uploaded by tilghman (license 14)
  Tested by: jcovert
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301264 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301221 via svnmerge from
Paul Belanger [Sun, 9 Jan 2011 21:42:47 +0000 (21:42 +0000)]
Merged revisions 301221 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301221 | pabelanger | 2011-01-09 16:40:34 -0500 (Sun, 09 Jan 2011) | 21 lines

  Merged revisions 301220 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301220 | pabelanger | 2011-01-09 16:38:24 -0500 (Sun, 09 Jan 2011) | 14 lines

    SOUND_CACHE_DIR now defaults to empty

    Sounds files included in the Asterisk tarball were being ignored and
    re-downloaded.  Users wanting to cache the files can still override the setting
    using the --with-sounds-cache option.

    (closes issue #18589)
    Reported by: pabelanger
    Patches:
          issue18589.patch uploaded by pabelanger (license 224)
          Tested by: pabelanger

    Review: https://reviewboard.asterisk.org/r/1074/
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301222 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301177 via svnmerge from
Paul Belanger [Sat, 8 Jan 2011 22:02:39 +0000 (22:02 +0000)]
Merged revisions 301177 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301177 | pabelanger | 2011-01-08 17:00:12 -0500 (Sat, 08 Jan 2011) | 14 lines

  Merged revisions 301176 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301176 | pabelanger | 2011-01-08 16:58:24 -0500 (Sat, 08 Jan 2011) | 7 lines

    Indicate log level argument for Log() is not optional

    (closes issue #18586)
    Reported by: kshumard
    Patches:
          app_verbose.c.patch uploaded by kshumard (license 92)
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301178 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301134 via svnmerge from
Richard Mudgett [Sat, 8 Jan 2011 01:13:58 +0000 (01:13 +0000)]
Merged revisions 301134 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r301134 | rmudgett | 2011-01-07 19:11:31 -0600 (Fri, 07 Jan 2011) | 7 lines

  The DTMF attended transfer feature cannot callback a chan_dahdi BRI phone.

  The DAHDI ISDN channel name is not dialable.

  Make a channel name like DAHDI/i3/400-12 dialable when the sequence number
  is stripped off of the name.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301135 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301090 via svnmerge from
Jason Parker [Fri, 7 Jan 2011 20:53:45 +0000 (20:53 +0000)]
Merged revisions 301090 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301090 | qwell | 2011-01-07 14:53:02 -0600 (Fri, 07 Jan 2011) | 15 lines

  Merged revisions 301089 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301089 | qwell | 2011-01-07 14:52:00 -0600 (Fri, 07 Jan 2011) | 8 lines

    Initialize useropts/adminopts in case there is no column in the realtime DB.

    (closes issue #18182)
    Reported by: dimas
    Patches:
          v1-18182.patch uploaded by dimas (license 88)
    Tested by: dimas
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301091 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 301047 via svnmerge from
Jeff Peeler [Fri, 7 Jan 2011 19:58:52 +0000 (19:58 +0000)]
Merged revisions 301047 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r301047 | jpeeler | 2011-01-07 13:58:30 -0600 (Fri, 07 Jan 2011) | 15 lines

  Merged revisions 301046 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ........
    r301046 | jpeeler | 2011-01-07 13:57:42 -0600 (Fri, 07 Jan 2011) | 8 lines

    Fix regression causing forwarding voicemails to not work with file storage.

    I had actually already fixed this in 295200 in 1.4 and thought it wasn't
    missing in the other branches for some reason.

    (closes issue #18358)
    Reported by: cabal95
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301048 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoOops, missed the actual decoding part.
Tilghman Lesher [Fri, 7 Jan 2011 18:23:52 +0000 (18:23 +0000)]
Oops, missed the actual decoding part.

(closes issue #18046)
 Reported by: wdoekes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@301008 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoMerged revisions 300955 via svnmerge from
Jeff Peeler [Fri, 7 Jan 2011 17:24:52 +0000 (17:24 +0000)]
Merged revisions 300955 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

................
  r300955 | jpeeler | 2011-01-07 11:24:14 -0600 (Fri, 07 Jan 2011) | 21 lines

  Merged revisions 300951 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.6.2

  ................
    r300951 | jpeeler | 2011-01-07 11:23:37 -0600 (Fri, 07 Jan 2011) | 14 lines

    Merged revisions 300918 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r300918 | jpeeler | 2011-01-07 11:13:21 -0600 (Fri, 07 Jan 2011) | 7 lines

      Ensure good bye prompt in voicemail is played at the correct time.

      Specifically in the case of timing out but not leaving voicemail nothing
      should be heard. And when leaving voicemail it should be heard.

      ABE-2647
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@300959 65c4cc65-6c06-0410-ace0-fbb531ad65f3

8 years agoAdded support for postgres database retry query on disconnection to res_config_pgsql
Mark Murawki [Fri, 7 Jan 2011 07:47:36 +0000 (07:47 +0000)]
Added support for postgres database retry query on disconnection to res_config_pgsql

If your postgres connection died suddenly in between res_config_pgsql
queries, the next query will fail because the query is executed on a
disconnected/disconnecting handle.  The query is abandoned and is
returned from in error.

Now we will reconnect and try again if a query was run on a
disconnected connection.

(closes issue #18071)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@300882 65c4cc65-6c06-0410-ace0-fbb531ad65f3