asterisk/asterisk.git
12 years agosip option flags handled incorrectly
David Vossel [Fri, 17 Jul 2009 17:51:44 +0000 (17:51 +0000)]
sip option flags handled incorrectly

(closes issue #15376)
Reported by: Takehiko Ooshima
Tested by: dvossel, Takehiko_Ooshima

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@207029 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix segfault in sig_analog when using callwaiting, respect callwaiting options
Jeff Peeler [Fri, 17 Jul 2009 17:02:44 +0000 (17:02 +0000)]
Fix segfault in sig_analog when using callwaiting, respect callwaiting options

Sig_analog handles allocating the sub channel for callwaiting, so no longer try
to do it in chan_dahdi. Modified analog_alloc_sub to only mark the sub as
allocated upon success of the alloc_sub callback, which was responsible for the
segfault. Also, the callwaiting and callwaitingcallerid options were being
unconditionally set to true. Now, the options are properly set from
chan_dahdi.conf.

(closes issue #15508)
Reported by: elguero
Tested by: elguero

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206998 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206938 via svnmerge from
David Vossel [Fri, 17 Jul 2009 16:13:22 +0000 (16:13 +0000)]
Merged revisions 206938 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206938 | dvossel | 2009-07-17 11:05:06 -0500 (Fri, 17 Jul 2009) | 14 lines

  SIP incorrect From: header information when callpres is prohib

  Some ITSP make use of the "Anonymous" display name to detect a
  requirement to withhold caller id across the PSTN. This does
  not work if the display name is "Unknown".

  (closes issue #14465)
  Reported by: Nick_Lewis
  Patches:
        chan_sip.c-callerpres.patch uploaded by Nick (license 657)
        chan_sip.c-callerpres_trunk.patch uploaded by dvossel (license 671)
  Tested by: Nick_Lewis, dvossel
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206939 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoTIMEOUT(absolute) returned negative value.
David Vossel [Thu, 16 Jul 2009 21:45:14 +0000 (21:45 +0000)]
TIMEOUT(absolute) returned negative value.

(closes issue #15513)
Reported by: ys

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206877 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206872 via svnmerge from
David Vossel [Thu, 16 Jul 2009 21:33:51 +0000 (21:33 +0000)]
Merged revisions 206872 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206872 | dvossel | 2009-07-16 16:33:19 -0500 (Thu, 16 Jul 2009) | 6 lines

  error in iax.conf related IP-based access control

  (closes issue #15518)
  Reported by: pkempgen
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206873 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206867 via svnmerge from
David Vossel [Thu, 16 Jul 2009 21:25:22 +0000 (21:25 +0000)]
Merged revisions 206867 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206867 | dvossel | 2009-07-16 16:24:16 -0500 (Thu, 16 Jul 2009) | 8 lines

  avoid segfault caused by user error

  If the CALLERPRES() dialplan function is set to nothing,
  a segfault occurs.  This is user error to begin with, but
  I'd rather see a cli warning message than have Asterisk
  crash on me.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206868 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206807 via svnmerge from
Tilghman Lesher [Thu, 16 Jul 2009 16:51:05 +0000 (16:51 +0000)]
Merged revisions 206807 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206807 | tilghman | 2009-07-16 11:27:35 -0500 (Thu, 16 Jul 2009) | 6 lines

  Fix a memory leak.
  (closes issue #15517)
   Reported by: adomjan
   Patches:
         func_realtime.c-ast_variable_destroy.diff uploaded by adomjan (license 487)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206808 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoSession timer were not activated if Supported header field in INVITE had both "timer...
David Vossel [Wed, 15 Jul 2009 22:04:13 +0000 (22:04 +0000)]
Session timer were not activated if Supported header field in INVITE had both "timer" and other options.

(closes issue #15403)
Reported by: makoto
Patches:
      sip-session-timer.patch uploaded by makoto (license 38)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206768 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoThe dialing flag was mistakingly removed from sig_pri.
Jeff Peeler [Wed, 15 Jul 2009 22:02:55 +0000 (22:02 +0000)]
The dialing flag was mistakingly removed from sig_pri.

This readds the proper setting of the flag and is really a continuation of
r205731. The flag was being set properly in sig_analog, but use of the
newly added set_dialing callback allowed for some simplification in
chan_dahdi.

(closes issue #15486)
Reported by: rmudgett

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206767 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206706 via svnmerge from
Richard Mudgett [Wed, 15 Jul 2009 21:14:41 +0000 (21:14 +0000)]
Merged revisions 206706 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

................
  r206706 | rmudgett | 2009-07-15 15:44:55 -0500 (Wed, 15 Jul 2009) | 26 lines

  Merged revision 206700 from
  https://origsvn.digium.com/svn/asterisk/be/branches/C.2-...

  ..........
    Fixed chan_misdn crash because mISDNuser library is not thread safe.

    With Asterisk the mISDNuser library is driven by two threads concurrently:
    1. channels/misdn/isdn_lib.c::manager_event_handler()
    2. channels/misdn/isdn_lib.c::misdn_lib_isdn_event_catcher()

    Calls into the library are done concurrently and recursively from
    isdn_lib.c.

    Both threads can fiddle with the master/child layer3_proc_t lists.  One
    thread may traverse the list when the other interrupts it and then removes
    the list element which the first thread was currently handling.  This is
    exactly what caused the crash.  About 60 calls were needed to a Gigaset
    CX475 before it occurred once.

    This patch adds locking when calling into the mISDNuser library.
    This also fixes some cb_log calls with wrong port parameter.

    JIRA ABE-1913
        Patches: misdn-locking.patch (Modified with mostly cosmetic changes)
  ..........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206707 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agocallerid(num) is wrong when username is missing
David Vossel [Wed, 15 Jul 2009 20:20:01 +0000 (20:20 +0000)]
callerid(num) is wrong when username is missing

A domain only sip uri <sip:123.123.123.123> would return
123.123.123.123 as callid num.  Now, if the username is
missing from a uri, the callerid num field is left empty.

(closes issue #15476)
Reported by: viraptor

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206702 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206635 via svnmerge from
Sean Bright [Wed, 15 Jul 2009 16:00:24 +0000 (16:00 +0000)]
Merged revisions 206635 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206635 | seanbright | 2009-07-15 11:57:51 -0400 (Wed, 15 Jul 2009) | 1 line

  Only print debug info in codec_dahdi if we are asking for it.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206636 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agofix a typo in sample config file for option change
Jeff Peeler [Tue, 14 Jul 2009 20:38:56 +0000 (20:38 +0000)]
fix a typo in sample config file for option change

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206603 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoDocument all meetme realtime fields, and in the process, make some field lengths...
Tilghman Lesher [Tue, 14 Jul 2009 20:14:45 +0000 (20:14 +0000)]
Document all meetme realtime fields, and in the process, make some field lengths more consistent.
(closes issue #15493)
 Reported by: lasko
 Patches:
       meetme.diff uploaded by lasko (license 833)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206567 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRestore some missing functionality to sig_analog.
Jeff Peeler [Tue, 14 Jul 2009 20:01:10 +0000 (20:01 +0000)]
Restore some missing functionality to sig_analog.

The main purpose of this commit is to restore missing functionality present in
the ss_thread before all the sig related work was done. Two of the biggest
missing things were distinctive ring detection and cid handling for V23.
fxsoffhookstate and associated mwi variables have been moved inside sig_analog
as they were not being set properly as well.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206566 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoI AM A TERRIBLE PERSON
Mark Michelson [Tue, 14 Jul 2009 17:03:58 +0000 (17:03 +0000)]
I AM A TERRIBLE PERSON

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206490 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206487 via svnmerge from
Richard Mudgett [Tue, 14 Jul 2009 17:01:48 +0000 (17:01 +0000)]
Merged revisions 206487 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206487 | rmudgett | 2009-07-14 11:44:47 -0500 (Tue, 14 Jul 2009) | 28 lines

  Fixes several call transfer issues with chan_misdn.

  *  issue #14355 - Crash if attempt to transfer a call to an application.
  Masquerade the other pair of the four asterisk channels involved in the
  two calls.  The held call already must be a bridged call (not an
  applicaton) or it would have been rejected.

  *  issue #14692 - Held calls are not automatically cleared after transfer.
  Allow the core to initate disconnect of held calls to the ISDN port.  This
  also fixes a similar case where the party on hold hangs up before being
  transferred or taken off hold.

  *  JIRA ABE-1903 - Orphaned held calls left in music-on-hold.
  Do not simply block passing the hangup event on held calls to asterisk
  core.

  *  Fixed to allow held calls to be transferred to ringing calls.
  Previously, held calls could only be transferred to connected calls.
  *  Eliminated unused call states to simplify hangup code.
  *  Eliminated most uses of "holded" because it is not a word.

  (closes issue #14355)
  (closes issue #14692)
  Reported by: sodom
  Patches:
        misdn_xfer_v14_r205839.patch uploaded by rmudgett (license 664)
  Tested by: rmudgett
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206489 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoReset the sentringing indication when redirects occur.
Mark Michelson [Tue, 14 Jul 2009 16:09:38 +0000 (16:09 +0000)]
Reset the sentringing indication when redirects occur.

If a redirecting control frame is processed or a call forward occurs,
we need to reset the sentringing flag so that we can send another ringing
indication to the phone that may contain a connected line update.

AST-164

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206455 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206385 via svnmerge from
Russell Bryant [Tue, 14 Jul 2009 14:51:44 +0000 (14:51 +0000)]
Merged revisions 206385 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

................
  r206385 | russell | 2009-07-14 09:48:00 -0500 (Tue, 14 Jul 2009) | 13 lines

  Merged revisions 206384 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.2

  ........
    r206384 | russell | 2009-07-14 09:45:47 -0500 (Tue, 14 Jul 2009) | 6 lines

    Ensure apathetic replies are sent out on the proper socket.

    chan_iax2 supports multiple address bindings.  The send_apathetic_reply()
    function did not attempt to send its response on the same socket that the
    incoming message came in on.
  ........
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206386 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 206284 via svnmerge from
Richard Mudgett [Tue, 14 Jul 2009 00:48:59 +0000 (00:48 +0000)]
Merged revisions 206284 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r206284 | rmudgett | 2009-07-13 19:17:28 -0500 (Mon, 13 Jul 2009) | 4 lines

  Fix some memory leaks in chan_misdn.

  JIRA ABE-1911
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206341 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agodns lookup of peername rather than peer's host in transmit_register()
David Vossel [Mon, 13 Jul 2009 23:26:51 +0000 (23:26 +0000)]
dns lookup of peername rather than peer's host in transmit_register()

(closes issue #15052)
Reported by: fsantulli
Patches:
      chan_sip_bug_15052_[20090626204511].patch uploaded by fsantulli (license 818)
Tested by: fsantulli

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206280 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMake sure that since we are passing -c to asterisk that we have a console.
Sean Bright [Mon, 13 Jul 2009 18:46:47 +0000 (18:46 +0000)]
Make sure that since we are passing -c to asterisk that we have a console.

Without this line, Asterisk will busy-loop trying to read and write to
/dev/null (woops... my bad).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206225 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRemove reference to non-existent help file
Tilghman Lesher [Mon, 13 Jul 2009 16:23:07 +0000 (16:23 +0000)]
Remove reference to non-existent help file
(closes issue #15427)
 Reported by: brushtyler
 Patches:
       app_voicemail.c.diff uploaded by brushtyler (license 821)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206185 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoBlocked revisions 206126 via svnmerge
Russell Bryant [Mon, 13 Jul 2009 15:12:31 +0000 (15:12 +0000)]
Blocked revisions 206126 via svnmerge

........
  r206126 | russell | 2009-07-13 10:12:08 -0500 (Mon, 13 Jul 2009) | 7 lines

  Print CID match in "show dialplan".

  (closes issue #14702)
  Reported by: klaus3000
  Patches:
        patch_asterisk_1.4.23_CID_matching.txt uploaded by klaus3000 (license 65)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206127 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoBump up cleancount so that existing checkouts will update themselves properly for...
Kevin P. Fleming [Mon, 13 Jul 2009 14:06:37 +0000 (14:06 +0000)]
Bump up cleancount so that existing checkouts will update themselves properly for the 'Addons' -> 'ADDONS' change.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206094 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMake the menuselect category for Add-Ons consistent with the other directories (upper...
Kevin P. Fleming [Mon, 13 Jul 2009 13:29:23 +0000 (13:29 +0000)]
Make the menuselect category for Add-Ons consistent with the other directories (uppercase).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206092 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agonote the security events API in CHANGES
Russell Bryant [Sat, 11 Jul 2009 19:30:19 +0000 (19:30 +0000)]
note the security events API in CHANGES

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206049 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd an API for reporting security events, and a security event logging module.
Russell Bryant [Sat, 11 Jul 2009 19:15:03 +0000 (19:15 +0000)]
Add an API for reporting security events, and a security event logging module.

This commit introduces the security events API.  This API is to be used by
Asterisk components to report events that have security implications.
A simple example is when a connection is made but fails authentication.  These
events can be used by external tools manipulate firewall rules or something
similar after detecting unusual activity based on security events.

Inside of Asterisk, the events go through the ast_event API.  This means that
they have a binary encoding, and it is easy to write code to subscribe to these
events and do something with them.

One module is provided that is a subscriber to these events - res_security_log.
This module turns security events into a parseable text format and sends them
to the "security" logger level.  Using logger.conf, these log entries may be
sent to a file, or to syslog.

One service, AMI, has been fully updated for reporting security events.
AMI was chosen as it was a fairly straight forward service to convert.
The next target will be chan_sip.  That will be more complicated and will
be done as its own project as the next phase of security events work.

For more information on the security events framework, see the documentation
generated from doc/tex/.  "make asterisk.pdf"

Review: https://reviewboard.asterisk.org/r/273/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@206021 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoSIP register not using peer's outbound proxy
David Vossel [Fri, 10 Jul 2009 21:42:10 +0000 (21:42 +0000)]
SIP register not using peer's outbound proxy

If callbackextension is defined for a peer it successfully causes
a registration to occur, but the registration ignores the
outboundproxy settings for the peer.  This patch allows the
peer to be passed to obproxy_get() in transmit_register().

(closes issue #14344)
Reported by: Nick_Lewis
Patches:
      callbackextension_peer_trunk.diff uploaded by dvossel (license 671)
Tested by: dvossel

Review: https://reviewboard.asterisk.org/r/294/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205985 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoUpdate comments about the level of T.38 support in Asterisk.
Kevin P. Fleming [Fri, 10 Jul 2009 18:44:09 +0000 (18:44 +0000)]
Update comments about the level of T.38 support in Asterisk.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205939 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205877 via svnmerge from
Mark Michelson [Fri, 10 Jul 2009 17:39:57 +0000 (17:39 +0000)]
Merged revisions 205877 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

................
  r205877 | mmichelson | 2009-07-10 12:39:13 -0500 (Fri, 10 Jul 2009) | 23 lines

  Merged revisions 205776 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/trunk

  ................
    r205776 | mmichelson | 2009-07-10 10:56:45 -0500 (Fri, 10 Jul 2009) | 16 lines

    Merged revisions 205775 via svnmerge from
    https://origsvn.digium.com/svn/asterisk/branches/1.4

    ........
      r205775 | mmichelson | 2009-07-10 10:51:36 -0500 (Fri, 10 Jul 2009) | 10 lines

      Ensure that outbound NOTIFY requests are properly routed through stateful proxies.

      With this change, we make note of Record-Route headers present in any SUBSCRIBE
      request that we receive so that our outbound NOTIFY requests will have the proper
      Route headers in them.

      (closes issue #14725)
      Reported by: ibc
    ........
  ................
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205878 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205804 via svnmerge from
David Vossel [Fri, 10 Jul 2009 16:42:04 +0000 (16:42 +0000)]
Merged revisions 205804 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205804 | dvossel | 2009-07-10 11:23:59 -0500 (Fri, 10 Jul 2009) | 31 lines

  SIP registration auth loop caused by stale nonce

  If an endpoint sends two registration requests in a very short
  period of time with the same nonce, both receive 401 responses
  from Asterisk, each with a different nonce (the second 401
  containing the current nonce and the first one being stale).
  If the endpoint responds to the first 401, it does not match
  the current nonce so Asterisk sends a third 401 with a newly
  generated nonce (which updates the current nonce)... Now if
  the endpoint responds to the second 401, it does not match the
  current nonce either and Asterisk sends a fourth 401 with a
  newly generated nonce... This loop goes on and on.

  There appears to be a simple fix for this.  If the nonce from
  the request does not match our nonce, but is a good response
  to a previous nonce, instead of sending a 401 with a newly
  generated nonce, use the current one instead.  This breaks
  the loop as the nonce is not updated until a response is
  received. Additional logic has been added to make sure no
  nonce can be responded to twice though.

  (closes issue #15102)
  Reported by: Jamuel
  Patches:
        patch-bug_0015102 uploaded by Jamuel (license 809)
        nonce_sip.diff uploaded by dvossel (license 671)
  Tested by: Jamuel

  Review: https://reviewboard.asterisk.org/r/289/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205840 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoEliminate extraneous LOG_DEBUG messages generated by app_fax.
Kevin P. Fleming [Fri, 10 Jul 2009 16:00:44 +0000 (16:00 +0000)]
Eliminate extraneous LOG_DEBUG messages generated by app_fax.

The transmit_audio() and transmit_t38() functions in app_fax have processing
loops that are supposed to wait for frames to arrive on the channel and then
handle them, but they also have short timeouts so that the loops can have
watchdog timers and do other required processing. This commit changes the loops
to not actually call ast_read() and attempt to process the returned frame
unless a frame actually arrived, eliminating hundreds of LOG_DEBUG messages
and slightly improving performance.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205780 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205775 via svnmerge from
Mark Michelson [Fri, 10 Jul 2009 15:56:45 +0000 (15:56 +0000)]
Merged revisions 205775 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205775 | mmichelson | 2009-07-10 10:51:36 -0500 (Fri, 10 Jul 2009) | 10 lines

  Ensure that outbound NOTIFY requests are properly routed through stateful proxies.

  With this change, we make note of Record-Route headers present in any SUBSCRIBE
  request that we receive so that our outbound NOTIFY requests will have the proper
  Route headers in them.

  (closes issue #14725)
  Reported by: ibc
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205776 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix some remaining T.38 negotiation problems in app_fax.
Kevin P. Fleming [Fri, 10 Jul 2009 15:28:11 +0000 (15:28 +0000)]
Fix some remaining T.38 negotiation problems in app_fax.

Revision 205696 did not quite fix all the issues with the T.38 negotiation
changes and app_fax; this patch corrects them, along with a couple of other
minor issues.

(closes issue #15480)
Reported by: dimas
Patches:
      test2-15480.patch uploaded by dimas (license 88)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205770 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix mbl_fixup() in chan_mobile to update newchan->tech_pvt instead of oldchan.
Matthew Nicholson [Thu, 9 Jul 2009 21:32:31 +0000 (21:32 +0000)]
Fix mbl_fixup() in chan_mobile to update newchan->tech_pvt instead of oldchan.

(closes issue #15299)
Reported by: nikkk

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205700 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRepair ability of SendFAX/ReceiveFAX to respond to T.38 switchover.
Kevin P. Fleming [Thu, 9 Jul 2009 21:20:23 +0000 (21:20 +0000)]
Repair ability of SendFAX/ReceiveFAX to respond to T.38 switchover.

Recent changes in T.38 negotiation in Asterisk caused these applications to
not respond when the other endpoint initiated a switchover to T.38; this
resulted in the T.38 switchover failing, and the FAX attempt to be made
using an audio connection, instead of T.38 (which would usually cause the
FAX to fail completely).

This patch corrects this problem, and the applications will now correctly
respond to the T.38 switchover request. In addition, the response will include
the appopriate T.38 session parameters based on what the other end offered
and what our end is capable of.

(closes issue #14849)
Reported by: afosorio

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205696 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoConvert func_odbc to use ast_dummy_alloc_channel()
Matthew Nicholson [Thu, 9 Jul 2009 20:04:43 +0000 (20:04 +0000)]
Convert func_odbc to use ast_dummy_alloc_channel()

Review: https://reviewboard.asterisk.org/r/290/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205666 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205599 via svnmerge from
David Vossel [Thu, 9 Jul 2009 16:19:09 +0000 (16:19 +0000)]
Merged revisions 205599 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205599 | dvossel | 2009-07-09 11:18:09 -0500 (Thu, 09 Jul 2009) | 2 lines

  Changing ast_samp2tv to not use floating point.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205600 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agomake this compile again under devmode
Michiel van Baak [Thu, 9 Jul 2009 14:10:01 +0000 (14:10 +0000)]
make this compile again under devmode

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205562 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agopthread_self returns a pthread_t which is not an unsigned int on all
Michiel van Baak [Thu, 9 Jul 2009 08:31:24 +0000 (08:31 +0000)]
pthread_self returns a pthread_t which is not an unsigned int on all
pthread implementations. Casting it to an unsigned int fixes compiler warnings.

Tested on OpenBSD and Linux both 32 and 64 bit

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205532 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205471 via svnmerge from
David Vossel [Wed, 8 Jul 2009 23:19:09 +0000 (23:19 +0000)]
Merged revisions 205471 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205471 | dvossel | 2009-07-08 18:15:54 -0500 (Wed, 08 Jul 2009) | 10 lines

  Fixes 8khz assumptions

  Many calculations assume 8khz is the codec rate. This
  is not always the case.  This patch only addresses chan_iax.c
  and res_rtp_asterisk.c, but I am sure there are other areas
  that make this assumption as well.

  Review: https://reviewboard.asterisk.org/r/306/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205479 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix a CEL related regression with hints updating by subscribing to AST_DEVICE_STATE...
Matthew Nicholson [Wed, 8 Jul 2009 23:07:09 +0000 (23:07 +0000)]
Fix a CEL related regression with hints updating by subscribing to AST_DEVICE_STATE instead of AST_DEVICE_STATE_CHANGED.

(closes issue #15440)
Reported by: lmsteffan

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205469 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205409 via svnmerge from
David Vossel [Wed, 8 Jul 2009 22:15:06 +0000 (22:15 +0000)]
Merged revisions 205409 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205409 | dvossel | 2009-07-08 16:35:12 -0500 (Wed, 08 Jul 2009) | 6 lines

  moving ast_devstate_to_extenstate to pbx.c from devicestate.c

  ast_devstate_to_extenstate belongs in pbx.c.  This change
  fixes a compile time error with chan_vpb as well.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205412 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agomissing comma in devstatestring array
David Vossel [Wed, 8 Jul 2009 22:02:54 +0000 (22:02 +0000)]
missing comma in devstatestring array

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205410 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205349 via svnmerge from
Mark Michelson [Wed, 8 Jul 2009 19:26:55 +0000 (19:26 +0000)]
Merged revisions 205349 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205349 | mmichelson | 2009-07-08 14:26:13 -0500 (Wed, 08 Jul 2009) | 14 lines

  Prevent phantom calls to queue members.

  If a caller were to hang up while a periodic announcement or position
  were being said, the return value for those functions would incorrectly
  indicate that the caller was still in the queue. With these changes,
  the problem does not occur.

  (closes issue #14631)
  Reported by: latinsud
  Patches:
        queue_announce_ghost_call2.diff uploaded by latinsud (license 745)
     (with small modification from me)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205350 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205288 via svnmerge from
Jason Parker [Wed, 8 Jul 2009 18:19:46 +0000 (18:19 +0000)]
Merged revisions 205288 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205288 | qwell | 2009-07-08 13:19:03 -0500 (Wed, 08 Jul 2009) | 1 line

  Update config.guess and config.sub from the savannah.gnu.org git repo.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205291 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFixes Park() argument handling
David Brooks [Wed, 8 Jul 2009 17:26:26 +0000 (17:26 +0000)]
Fixes Park() argument handling

Park() was not respecting the arguments passed to it. Any extension/context/priority
given to it was being ignored. This patch remedies this.

(closes issue #15380)
Reported by: DLNoah

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205254 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoOops, fixing build
Tilghman Lesher [Wed, 8 Jul 2009 16:59:32 +0000 (16:59 +0000)]
Oops, fixing build

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205221 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205215 via svnmerge from
David Vossel [Wed, 8 Jul 2009 16:54:24 +0000 (16:54 +0000)]
Merged revisions 205215 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205215 | dvossel | 2009-07-08 11:53:40 -0500 (Wed, 08 Jul 2009) | 10 lines

  ast_samp2tv needs floating point for 16khz audio

  In ast_samp2tv(), (1000000 / _rate) = 62.5 when _rate is 16000.
  The .5 is currently stripped off because we don't calculate
  using floating points.  This causes madness with 16khz audio.

  (issue ABE-1899)

  Review: https://reviewboard.asterisk.org/r/305/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205216 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix a few compilation problems found when building Asterisk against uClibc.
Sean Bright [Wed, 8 Jul 2009 16:43:12 +0000 (16:43 +0000)]
Fix a few compilation problems found when building Asterisk against uClibc.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205214 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 205188 via svnmerge from
Tilghman Lesher [Wed, 8 Jul 2009 16:27:50 +0000 (16:27 +0000)]
Merged revisions 205188 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r205188 | tilghman | 2009-07-08 11:26:15 -0500 (Wed, 08 Jul 2009) | 2 lines

  Add redirection warnings for the invalid language codes previously removed.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205196 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoUse tabs instead of spaces for indentation.
Russell Bryant [Wed, 8 Jul 2009 15:56:28 +0000 (15:56 +0000)]
Use tabs instead of spaces for indentation.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205151 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoBlocked revisions 205149 via svnmerge
Russell Bryant [Wed, 8 Jul 2009 15:54:42 +0000 (15:54 +0000)]
Blocked revisions 205149 via svnmerge

........
  r205149 | russell | 2009-07-08 10:54:21 -0500 (Wed, 08 Jul 2009) | 8 lines

  Make OpenSSL usage thread-safe.

  OpenSSL is not thread-safe by default.  However, making it thread safe is
  very easy.  We just have to provide a couple of callbacks.  One callback
  returns a thread ID.  The other handles locking.  For more information,
  start with the "Is OpenSSL thread-safe?" question on the FAQ page of
  openssl.org.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205150 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMove OpenSSL initialization to a single place, make library usage thread-safe.
Russell Bryant [Wed, 8 Jul 2009 15:17:19 +0000 (15:17 +0000)]
Move OpenSSL initialization to a single place, make library usage thread-safe.

While doing some reading about OpenSSL, I noticed a couple of things that
needed to be improved with our usage of OpenSSL.

1) We had initialization of the library done in multiple modules.  This has now
   been moved to a core function that gets executed during Asterisk startup.
   We already link OpenSSL into the core for TCP/TLS functionality, so this
   was the most logical place to do it.

2) OpenSSL is not thread-safe by default.  However, making it thread safe is
   very easy.  We just have to provide a couple of callbacks.  One callback
   returns a thread ID.  The other handles locking.  For more information,
   start with the "Is OpenSSL thread-safe?" question on the FAQ page of
   openssl.org.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205120 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFreeBSD now has autoconf 2.62 in the ports, 2.61 has disappeared.
Luigi Rizzo [Wed, 8 Jul 2009 14:45:15 +0000 (14:45 +0000)]
FreeBSD now has autoconf 2.62 in the ports, 2.61 has disappeared.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205118 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoPermit setting custom headers from the peer definition.
Tilghman Lesher [Tue, 7 Jul 2009 21:10:14 +0000 (21:10 +0000)]
Permit setting custom headers from the peer definition.
(closes issue #14059)
 Reported by: fnordian

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205086 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoFix a deadlock in sig_analog
Matthew Nicholson [Tue, 7 Jul 2009 18:24:13 +0000 (18:24 +0000)]
Fix a deadlock in sig_analog

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205047 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd CEL transfer events to analog (chan_dahdi) transfers.
Matthew Nicholson [Mon, 6 Jul 2009 23:24:57 +0000 (23:24 +0000)]
Add CEL transfer events to analog (chan_dahdi) transfers.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@205014 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 981 via svnmerge from
Tilghman Lesher [Mon, 6 Jul 2009 21:37:39 +0000 (21:37 +0000)]
Merged revisions 981 via svnmerge from
https://origsvn.digium.com/svn/asterisk-addons/branches/1.4

........
  r981 | tilghman | 2009-07-06 16:30:13 -0500 (Mon, 06 Jul 2009) | 7 lines

  Don't reset reconnect time, unless a reconnect really occurred.
  (closes issue #15375)
   Reported by: kowalma
   Patches:
         20090628__issue15375.diff.txt uploaded by tilghman (license 14)
   Tested by: kowalma, jacco
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204986 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoImprove handling of AST_CONTROL_T38 and AST_CONTROL_T38_PARAMETERS for non-T.38-capab...
Kevin P. Fleming [Mon, 6 Jul 2009 13:38:29 +0000 (13:38 +0000)]
Improve handling of AST_CONTROL_T38 and AST_CONTROL_T38_PARAMETERS for non-T.38-capable channels.

This change allows applications that request T.38 negotiation on a channel that
does not support it to get the proper indication that it is not supported, rather
than thinking that negotiation was started when it was not.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204948 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd a configure check for Reverse Charging Indication support in LibPRI.
Sean Bright [Fri, 3 Jul 2009 15:44:01 +0000 (15:44 +0000)]
Add a configure check for Reverse Charging Indication support in LibPRI.

Also go back and wrap all of the places that use the specific reverse charge
APIs with preprocessor conditionals.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204919 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoWrap rtp_engine.h header comments to 80 characters.
Sean Bright [Fri, 3 Jul 2009 02:02:50 +0000 (02:02 +0000)]
Wrap rtp_engine.h header comments to 80 characters.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204893 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 204834 via svnmerge from
Richard Mudgett [Thu, 2 Jul 2009 22:01:28 +0000 (22:01 +0000)]
Merged revisions 204834 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204834 | rmudgett | 2009-07-02 16:59:43 -0500 (Thu, 02 Jul 2009) | 10 lines

  Removed confusing warning message "Got Busy in Connected State"

  If an incoming mISDN call is answered with the Answer application and a
  subsequent Dial gets a busy endpoint then it is valid for that already
  connected channel to get the busy indication.  Asterisk will play the busy
  tones until the dialplan plays something else or hangs up the call.

  (closes issue #11974)
  Reported by: fvdb
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204835 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMoved trigger for BRIDGE_END CEL event so that it is more accurate.
Matthew Nicholson [Thu, 2 Jul 2009 20:37:16 +0000 (20:37 +0000)]
Moved trigger for BRIDGE_END CEL event so that it is more accurate.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204807 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoSupport setting and receiving Reverse Charging Indication over ISDN PRI.
Sean Bright [Thu, 2 Jul 2009 17:46:14 +0000 (17:46 +0000)]
Support setting and receiving Reverse Charging Indication over ISDN PRI.

This is a continuation of revision 885 to LibPRI (Capture and expose the Reverse
Charging Indication IE on ISDN PRI) which added the ability to get/set Reverse
Charging Indication in LibPRI.  This patch adds the ability to specify RCI on
the outbound leg of a PRI call from within Asterisk, by prefixing the dialed
number with a capital 'C' like:

...,Dial(DAHDI/g1/C4445556666)

And to read it off an inbound channel:

exten => s,1,Set(RCI=${CHANNEL(reversecharge)})

Thanks again to rmudgett for the thorough review.

(closes issue #13760)
Reported by: mrgabu

Review: https://reviewboard.asterisk.org/r/303/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204749 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 204681 via svnmerge from
David Vossel [Thu, 2 Jul 2009 16:03:44 +0000 (16:03 +0000)]
Merged revisions 204681 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204681 | dvossel | 2009-07-02 10:05:57 -0500 (Thu, 02 Jul 2009) | 14 lines

  Improved mapping of extension states from combined device states.

  This fixes a few issues with incorrect extension states and adds
  a cli command, core show device2extenstate, to display all possible
  state mappings.

  (closes issue #15413)
  Reported by: legart
  Patches:
        exten_helper.diff uploaded by dvossel (license 671)
  Tested by: dvossel, legart, amilcar

  Review: https://reviewboard.asterisk.org/r/301/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204710 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years ago- cfgbasic.html has been replaced by index.html in the GUI for some time now
Ryan Brindley [Wed, 1 Jul 2009 19:47:38 +0000 (19:47 +0000)]
- cfgbasic.html has been replaced by index.html in the GUI for some time now

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204654 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoA bunch of CODING_GUIDELINES related fixes. Not even close to done.
Sean Bright [Wed, 1 Jul 2009 16:06:18 +0000 (16:06 +0000)]
A bunch of CODING_GUIDELINES related fixes.  Not even close to done.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204622 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 204556 via svnmerge from
Tilghman Lesher [Tue, 30 Jun 2009 20:41:04 +0000 (20:41 +0000)]
Merged revisions 204556 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204556 | tilghman | 2009-06-30 15:23:51 -0500 (Tue, 30 Jun 2009) | 6 lines

  More incorrect language codes, plus ensuring that regionalizations use the specified language, and not English for grammar.
  (closes issue #15022)
   Reported by: greenfieldtech
   Patches:
         20090519__issue15022.diff.txt uploaded by tilghman (license 14)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204563 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRemove an unnecessary #ifdef
Sean Bright [Tue, 30 Jun 2009 20:39:39 +0000 (20:39 +0000)]
Remove an unnecessary #ifdef

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204561 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMove the masquerade in local_attended_transfer to a point where we hold the channel...
Mark Michelson [Tue, 30 Jun 2009 19:59:20 +0000 (19:59 +0000)]
Move the masquerade in local_attended_transfer to a point where we hold the channel lock.

Masquerading without the channel's lock held is a *horrible* idea.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204532 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRemove some bogus deadlock avoidance code from local_attended_transfer.
Mark Michelson [Tue, 30 Jun 2009 19:55:59 +0000 (19:55 +0000)]
Remove some bogus deadlock avoidance code from local_attended_transfer.

First of all, the code was unnecessary. The goal was to lock a channel
which was already locked. Second, the assumption of the deadlock avoidance
loop was that the sip_pvt was already locked and we were trying to get the
channel lock. The problem is that the sip_pvt was unlocked a few lines above.

Basically, I'm removing 5 lines of no-op.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204530 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 204474 via svnmerge from
Jason Parker [Tue, 30 Jun 2009 18:48:35 +0000 (18:48 +0000)]
Merged revisions 204474 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204474 | qwell | 2009-06-30 13:47:06 -0500 (Tue, 30 Jun 2009) | 1 line

  Fix ast_say_counted_noun to correctly handle Polish.  Fix a comment typo in passing.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204475 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRecorded merge of revisions 204469 via svnmerge from
Tilghman Lesher [Tue, 30 Jun 2009 18:36:24 +0000 (18:36 +0000)]
Recorded merge of revisions 204469 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204469 | tilghman | 2009-06-30 13:23:35 -0500 (Tue, 30 Jun 2009) | 11 lines

  "tw" is the language specification for Twi (from Ghana) not Taiwanese.
  (closes issue #15346)
   Reported by: volivier
   Patches:
         20090617__issue15346__1.4.diff.txt uploaded by tilghman (license 14)
         20090617__issue15346__trunk.diff.txt uploaded by tilghman (license 14)
         20090617__issue15346__1.6.0.diff.txt uploaded by tilghman (license 14)
         20090617__issue15346__1.6.1.diff.txt uploaded by tilghman (license 14)
         20090617__issue15346__1.6.2.diff.txt uploaded by tilghman (license 14)
   Tested by: volivier
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204470 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename res_config_sqlite.conf to res_config_sqlite.conf.sample (missing .sample).
Russell Bryant [Tue, 30 Jun 2009 17:22:16 +0000 (17:22 +0000)]
Rename res_config_sqlite.conf to res_config_sqlite.conf.sample (missing .sample).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204440 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename ooh323.conf to chan_ooh323.conf, make module support both names
Russell Bryant [Tue, 30 Jun 2009 17:18:18 +0000 (17:18 +0000)]
Rename ooh323.conf to chan_ooh323.conf, make module support both names

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204428 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename mobile.conf to chan_mobile.conf, make module support old name, too
Russell Bryant [Tue, 30 Jun 2009 17:16:56 +0000 (17:16 +0000)]
Rename mobile.conf to chan_mobile.conf, make module support old name, too

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204423 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename res_mysql.conf to res_config_mysql.conf, make module support both
Russell Bryant [Tue, 30 Jun 2009 17:15:09 +0000 (17:15 +0000)]
Rename res_mysql.conf to res_config_mysql.conf, make module support both

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204422 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMake addons build last - this is for Qwell.
Russell Bryant [Tue, 30 Jun 2009 17:11:31 +0000 (17:11 +0000)]
Make addons build last - this is for Qwell.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204420 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename mysql.conf to app_mysql.conf, make module support both names
Russell Bryant [Tue, 30 Jun 2009 17:10:45 +0000 (17:10 +0000)]
Rename mysql.conf to app_mysql.conf, make module support both names

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204419 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename cdr_addon_mysql to cdr_mysql
Russell Bryant [Tue, 30 Jun 2009 17:09:04 +0000 (17:09 +0000)]
Rename cdr_addon_mysql to cdr_mysql

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204418 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRename app_addon_sql_mysql to app_mysql
Russell Bryant [Tue, 30 Jun 2009 17:08:14 +0000 (17:08 +0000)]
Rename app_addon_sql_mysql to app_mysql

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204417 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd-ons related build system improvements.
Kevin P. Fleming [Tue, 30 Jun 2009 17:04:35 +0000 (17:04 +0000)]
Add-ons related build system improvements.

Ensure that add-on modules can be embedded, fix up Makefile.moddir_rules
to allow module directory Makefiles to more easily specify the modules to
be built, and explicitly list the addons modules in its Makefile, since
the module names don't follow any pattern.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204415 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMove Asterisk-addons modules into the main Asterisk source tree.
Russell Bryant [Tue, 30 Jun 2009 16:40:38 +0000 (16:40 +0000)]
Move Asterisk-addons modules into the main Asterisk source tree.

Someone asked yesterday, "is there a good reason why we can't just put these
modules in Asterisk?".  After a brief discussion, as long as the modules are
clearly set aside in their own directory and not enabled by default, it is
perfectly fine.

For more information about why a module goes in addons, see README-addons.txt.

chan_ooh323 does not currently compile as it is behind some trunk API updates.
However, it will not build by default, so it should be okay for now.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204413 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoA few const changes in app_meetme.c that I noticed while browsing the source.
Sean Bright [Mon, 29 Jun 2009 23:50:46 +0000 (23:50 +0000)]
A few const changes in app_meetme.c that I noticed while browsing the source.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204355 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 204300 via svnmerge from
Mark Michelson [Mon, 29 Jun 2009 22:50:35 +0000 (22:50 +0000)]
Merged revisions 204300 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204300 | mmichelson | 2009-06-29 17:45:34 -0500 (Mon, 29 Jun 2009) | 9 lines

  Add error message so that it is clear why a SIP peer was not processed when
  a DNS lookup fails on a host or outboundproxy.

  (closes issue #13432)
  Reported by: p_lindheimer
  Patches:
        outboundproxy.patch uploaded by p (license 558)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204301 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 204243,204246 via svnmerge from
Mark Michelson [Mon, 29 Jun 2009 21:48:54 +0000 (21:48 +0000)]
Merged revisions 204243,204246 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r204243 | mmichelson | 2009-06-29 16:23:43 -0500 (Mon, 29 Jun 2009) | 22 lines

  Fix a problem where chan_sip would ignore "old" but valid responses.

  chan_sip has had a problem for quite a long time that would manifest when
  Asterisk would send multiple SIP responses on the same dialog before receiving
  a response. The problem occurred because chan_sip only kept track of the highest
  outgoing sequence number used on the dialog. If Asterisk sent two requests out,
  and a response arrived for the first request sent, then Asterisk would ignore
  the response. The result was that Asterisk would continue retransmitting the
  requests and ignoring the responses until the maximum number of retransmissions
  had been reached.

  The fix here is to rearrange the code a bit so that instead of simply comparing
  the sequence number of the response to our latest outgoing sequence number, we
  walk our list of outstanding packets and determine if there is a match. If there is,
  we continue. If not, then we ignore the response.

  In doing this, I found a few completely useless variables that I have now removed.

  (closes issue #11231)
  Reported by: flefoll

  Review: https://reviewboard.asterisk.org/r/298
........
  r204246 | mmichelson | 2009-06-29 16:37:05 -0500 (Mon, 29 Jun 2009) | 3 lines

  Fix build oops.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204247 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoReorganize this adaptive CEL config a bit.
Sean Bright [Mon, 29 Jun 2009 20:29:10 +0000 (20:29 +0000)]
Reorganize this adaptive CEL config a bit.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204217 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoBlocked revisions 204170 via svnmerge
Tilghman Lesher [Mon, 29 Jun 2009 19:36:57 +0000 (19:36 +0000)]
Blocked revisions 204170 via svnmerge

........
  r204170 | tilghman | 2009-06-29 14:36:01 -0500 (Mon, 29 Jun 2009) | 3 lines

  Revision 189537 was supposed to make 1.4 more correct.  Instead, it broke func_odbc.  Reverting.
  (closes issue #15317, issue #14614)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204171 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoGet app_rpt compiling again. I doubt seriously that it actually works.
Sean Bright [Mon, 29 Jun 2009 18:44:44 +0000 (18:44 +0000)]
Get app_rpt compiling again.  I doubt seriously that it actually works.

Also, the code in this module is horrendous and we should remove it from the
tree.  I'm not sure who is supposed to be maintaning this thing, but they
clearly are not.  I don't see the sense of leaving it in the main tree.  If it
lives *anywhere* it should be in addons.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204143 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAdd common headers to CEL related configs.
Sean Bright [Mon, 29 Jun 2009 18:05:27 +0000 (18:05 +0000)]
Add common headers to CEL related configs.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204119 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAllow trunk to once again compile under MALLOC_DEBUG
Tilghman Lesher [Mon, 29 Jun 2009 17:56:29 +0000 (17:56 +0000)]
Allow trunk to once again compile under MALLOC_DEBUG

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204118 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoRemove invalid entries in the config.
Tilghman Lesher [Mon, 29 Jun 2009 17:15:15 +0000 (17:15 +0000)]
Remove invalid entries in the config.
This might seem like a legitimate comment that merely needed semicolon
prefixes, but in reality, the adaptive layer is designed to allow arbitrary
CDR variables, without needing the use of a userfield to store multiple items.
It's therefore not only invalid syntax but also goes against the intent of the
adaptive method.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204069 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoBlocked revisions 204012 via svnmerge
Mark Michelson [Mon, 29 Jun 2009 15:04:39 +0000 (15:04 +0000)]
Blocked revisions 204012 via svnmerge

........
  r204012 | mmichelson | 2009-06-29 10:04:17 -0500 (Mon, 29 Jun 2009) | 6 lines

  Place unlock of mutex in an else block so that it does not get unlocked twice.

  (closes issue #15400)
  Reported by: aragon
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@204013 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoAnother CHANGES spelling fix.
Sean Bright [Sat, 27 Jun 2009 20:26:01 +0000 (20:26 +0000)]
Another CHANGES spelling fix.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@203985 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoOnly update total silence counter after a counter reset.
Russell Bryant [Sat, 27 Jun 2009 10:04:51 +0000 (10:04 +0000)]
Only update total silence counter after a counter reset.

(closes issue #2264)
Reported by: pfn
Patches:
      silent-vm-1.6.2-fix2.txt uploaded by pfn (license 810)
Tested by: pfn

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@203962 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMinor tweaks and spelling fixes for CHANGES and UPGRADE.txt.
Russell Bryant [Sat, 27 Jun 2009 09:51:45 +0000 (09:51 +0000)]
Minor tweaks and spelling fixes for CHANGES and UPGRADE.txt.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@203960 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 203908 via svnmerge from
Richard Mudgett [Sat, 27 Jun 2009 01:07:52 +0000 (01:07 +0000)]
Merged revisions 203908 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r203908 | rmudgett | 2009-06-26 19:55:12 -0500 (Fri, 26 Jun 2009) | 16 lines

  The ISDN CPE side should not exclusively pick B channels normally.

  Before this patch, Asterisk unconditionally picked B channels exclusively
  on the CPE side and normally allowed alternative B channels on the network
  side.  Now Asterisk does the opposite.

  Reasons for the CPE side to normally not pick B channels exclusively:
  *  For CPE point-to-multipoint mode (i.e. phone side), the CPE side does
  not have enough information to exclusively pick B channels.  (There may be
  other devices on the line.)
  *  Q.931 gives preference to the network side picking B channels.
  *  Some telcos require the CPE side to not pick B channels exclusively.

  (closes issue #14383)
  Reported by: mbrancaleoni
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@203909 65c4cc65-6c06-0410-ace0-fbb531ad65f3

12 years agoMerged revisions 203848 via svnmerge from
Jeff Peeler [Fri, 26 Jun 2009 22:11:31 +0000 (22:11 +0000)]
Merged revisions 203848 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r203848 | jpeeler | 2009-06-26 17:09:19 -0500 (Fri, 26 Jun 2009) | 5 lines

  Make sure to recreate the dahdi pseudo channel after dahdi restart

  (closes issue #14477)
  Reported by: timking
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@203853 65c4cc65-6c06-0410-ace0-fbb531ad65f3