irroot: Author Summary

mISDN Round Robin break when no channel is available

Prevent channels been parsed repetitively.
........

Merged revisions 344965 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........

Merged revisions 344966 from http://svn.asterisk.org/svn/asterisk/branches/10

mISDN Round Robin break when no channel is available

Prevent channels been parsed repetitively.
........

Merged revisions 344965 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........

Merged revisions 344966 from http://svn.asterisk.org/svn/asterisk/branches/10

mISDN Round Robin break when no channel is available

Prevent channels been parsed repetitively.
........

Merged revisions 344965 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........

Merged revisions 344966 from http://svn.asterisk.org/svn/asterisk/branches/10

mISDN Round Robin break when no channel is available

Prevent channels been parsed repetitively.
........

Merged revisions 344965 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........

Merged revisions 344966 from http://svn.asterisk.org/svn/asterisk/branches/10

mISDN Round Robin break when no channel is available

Prevent channels been parsed repetitively.
........

Merged revisions 344965 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........

Merged revisions 344966 from http://svn.asterisk.org/svn/asterisk/branches/10

mISDN Round Robin break when no channel is available

Prevent channels been parsed repetitively.
........

Merged revisions 344965 from http://svn.asterisk.org/svn/asterisk/branches/1.8

queues container needs locking when using the OBJ_NOLOCK flag
........

Merged revisions 342017 from http://svn.asterisk.org/svn/asterisk/branches/10

queues container needs locking when using the OBJ_NOLOCK flag
........

Merged revisions 342017 from http://svn.asterisk.org/svn/asterisk/branches/10

queues container needs locking when using the OBJ_NOLOCK flag
........

Merged revisions 342017 from http://svn.asterisk.org/svn/asterisk/branches/10

queues container needs locking when using the OBJ_NOLOCK flag
........

Merged revisions 342017 from http://svn.asterisk.org/svn/asterisk/branches/10

Remove some ref leaks and a return without unlock.

There some resource leaks introduced in asterisk 10
make sure that locks are not held on return and we
release ref's held.

Merged revisions 339463 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

........
  r339463 | irroot | 2011-10-05 08:28:46 +0200 (Wed, 05 Oct 2011) | 9 lines
  
  Only change the capabilities on the gateway when
  the session is been destroyed there is still
  a race condition that ends in a segfault.
  
  if the caps are changed the logic in res_fax_spandsp
  will run T30 code not gateway code to end the session.
  this has been experienced on a "slower" under spec system.
........

Merged revisions 339463 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

........
  r339463 | irroot | 2011-10-05 08:28:46 +0200 (Wed, 05 Oct 2011) | 9 lines
  
  Only change the capabilities on the gateway when
  the session is been destroyed there is still
  a race condition that ends in a segfault.
  
  if the caps are changed the logic in res_fax_spandsp
  will run T30 code not gateway code to end the session.
  this has been experienced on a "slower" under spec system.
........

Remove T38 Gateway capability when detaching framehook.

SET(FAXOPT(gateway)=no) does not remove the capability when
detaching the framehook.

small patch to fix this problem.

Merged revisions 338417 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r338417 | irroot | 2011-09-29 14:16:42 +0200 (Thu, 29 Sep 2011) | 19 lines
  
  Merged revisions 338416 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r338416 | irroot | 2011-09-29 14:13:05 +0200 (Thu, 29 Sep 2011) | 12 lines
    
    The rtptimeout setting is ignored on a per peer basis.
    
    Not only is the rtptimeout ignored in some cases but
    rtpkeepalive and rtpholdtimeout is affected.
    
    this commit also removes rtptimeout/rtpholdtimeout on
    text rtp.
    
    (closes issue ASTERISK-18559)
    
    Review: https://reviewboard.asterisk.org/r/1452
  ........
................

Merged revisions 337542 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r337542 | irroot | 2011-09-22 13:44:22 +0200 (Thu, 22 Sep 2011) | 14 lines
  
  Merged revisions 337541 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r337541 | irroot | 2011-09-22 13:39:49 +0200 (Thu, 22 Sep 2011) | 8 lines
    
    Add warned to ast_srtp to prevent errors on each frame from libsrtp
    
    The first 9 frames are not reported as some devices dont use srtp
    from first frame these are suppresed.
    
    the warning is then output only once every 100 frames.
  ........
................

Merged revisions 337487 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r337487 | irroot | 2011-09-22 11:26:26 +0200 (Thu, 22 Sep 2011) | 16 lines
  
  Merged revisions 337486 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r337486 | irroot | 2011-09-22 11:22:26 +0200 (Thu, 22 Sep 2011) | 10 lines
    
    If IP address is used in chan_h323 host parameter of peer configuration.
    module tries to resolve IP address to IP address and fails.
    
    Simple fix to set family of socket this is a hangover from ipv6 changes.
    
    (closes issue ASTERISK-18237)
    (issue ASTERISK-17278)
    (issue ASTERISK-17500)
  ........
................

Merged revisions 337541 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r337541 | irroot | 2011-09-22 13:39:49 +0200 (Thu, 22 Sep 2011) | 8 lines
  
  Add warned to ast_srtp to prevent errors on each frame from libsrtp
  
  The first 9 frames are not reported as some devices dont use srtp
  from first frame these are suppresed.
  
  the warning is then output only once every 100 frames.
........

Revert commit r337261

This commit is for trunk not version 10

-----
Adds a timeout argument to app_originate

the default is 30s this will be used if the timout supplied is invalid or
no timeout is supplied.
-----

Merged revisions 337486 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r337486 | irroot | 2011-09-22 11:22:26 +0200 (Thu, 22 Sep 2011) | 10 lines
  
  If IP address is used in chan_h323 host parameter of peer configuration.
  module tries to resolve IP address to IP address and fails.
  
  Simple fix to set family of socket this is a hangover from ipv6 changes.
  
  (closes issue ASTERISK-18237)
  (issue ASTERISK-17278)
  (issue ASTERISK-17500)
........

Merged revisions 335991 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r335991 | irroot | 2011-09-15 10:29:12 +0200 (Thu, 15 Sep 2011) | 17 lines
  
  Merged revisions 335978 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r335978 | irroot | 2011-09-15 10:15:22 +0200 (Thu, 15 Sep 2011) | 11 lines
    
    lock the channel before calling ast_bridged_channel() to prevent a seg fault.
    
    AMI agents list called on shutdown causes a segfault, introducing proper locking
    will prevent this.
    
    (closes issue ASTERISK-18092)
    
    Reported by: agustina
    Patches: chan_agent.patch (License #5041) patch uploaded by irroot
  ........
................

Merged revisions 335014 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

........
  r335014 | irroot | 2011-09-09 09:23:53 +0200 (Fri, 09 Sep 2011) | 9 lines
  
  
  Move code for VALID_EXTEN from app_readexten to func_dialplan
  
  Mark VALID_EXTEN deprecated.
  
  Review: https://reviewboard.asterisk.org/r/1396/
........

Move code for VALID_EXTEN from app_readexten to func_dialplan

Mark VALID_EXTEN deprecated.

Review: https://reviewboard.asterisk.org/r/1396/

Whitespace Fixups / Add Braces

This janitorial patch is related to work on RB1538
........

Merged revisions 341906 from http://svn.asterisk.org/svn/asterisk/branches/1.8

Only change the capabilities on the gateway when
the session is been destroyed there is still
a race condition that ends in a segfault.

if the caps are changed the logic in res_fax_spandsp
will run T30 code not gateway code to end the session.
this has been experienced on a "slower" under spec system.

Merged revisions 338950 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

........
  r338950 | irroot | 2011-10-03 11:37:59 +0200 (Mon, 03 Oct 2011) | 14 lines
  
  Fixup a race condition in res_fax.c where FAXOPT(gateway)=no will
  turn off the gateway but the framehook is not destroyed.
  
  this problem happens when a gateway is attempted in the dialplan and
  the device is not available i may want to do fax to mail in the server
  it will not be allowed.
  
  instead of checking only AST_FAX_TECH_GATEWAY also check gateway_id
  
  Reverts 338904
  
  Fix some white space.
........

Fixup a race condition in res_fax.c where FAXOPT(gateway)=no will
turn off the gateway but the framehook is not destroyed.

this problem happens when a gateway is attempted in the dialplan and
the device is not available i may want to do fax to mail in the server
it will not be allowed.

instead of checking only AST_FAX_TECH_GATEWAY also check gateway_id

Reverts 338904

Fix some white space.

Merged revisions 337839 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r337839 | irroot | 2011-09-23 10:34:03 +0200 (Fri, 23 Sep 2011) | 11 lines
  
  Make sure a CDR is on the stack for call in the Queue.
  Only let update_cdr act on the last CDR in the stack.
  
  In some circumstances [Attended transfer to queue] a
  CDR record is not inserted for this call where it should.
  
  (closes issue ASTERISK-18567)
  
  Review: https://reviewboard.asterisk.org/r/1266
........

Blocked revisions 337433 via svnmerge

........
  r337433 | irroot | 2011-09-22 08:42:42 +0200 (Thu, 22 Sep 2011) | 12 lines
  
  Revert commit r337261
  
  This commit is for trunk not version 10
  
  -----
  Adds a timeout argument to app_originate
  
  the default is 30s this will be used if the timout supplied is invalid or
  no timeout is supplied.
  -----
........

Merged revisions 337431 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r337431 | irroot | 2011-09-22 08:29:09 +0200 (Thu, 22 Sep 2011) | 25 lines
  
  Merged revisions 337430 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r337430 | irroot | 2011-09-22 08:18:33 +0200 (Thu, 22 Sep 2011) | 19 lines
    
    Its possible to loose audio on ast_write when the channel is not transcoded correctly.
    in the case of DAHDI the channel is hungup.
    
    This patch tries to "fix" the problem and make the channel compatiable and warn the user of
    this problem.
    
    Please note there is a underlying problem with codec negotion this does not fix the problem
    it does try to rectify it and prevent loss of service.
    
    Review: https://reviewboard.asterisk.org/r/1442/
    
    (closes issue ASTERISK-17541)
    (closes issue ASTERISK-18063)
    (issue ASTERISK-14384)
    (issue ASTERISK-17502)
    (issue ASTERISK-18325)
    (issue ASTERISK-18422)
  ........
................

Merged revisions 337430 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.8

........
  r337430 | irroot | 2011-09-22 08:18:33 +0200 (Thu, 22 Sep 2011) | 19 lines
  
  Its possible to loose audio on ast_write when the channel is not transcoded correctly.
  in the case of DAHDI the channel is hungup.
  
  This patch tries to "fix" the problem and make the channel compatiable and warn the user of
  this problem.
  
  Please note there is a underlying problem with codec negotion this does not fix the problem
  it does try to rectify it and prevent loss of service.
  
  Review: https://reviewboard.asterisk.org/r/1442/
  
  (closes issue ASTERISK-17541)
  (closes issue ASTERISK-18063)
  (issue ASTERISK-14384)
  (issue ASTERISK-17502)
  (issue ASTERISK-18325)
  (issue ASTERISK-18422)
........

Merged revisions 336094 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r336094 | irroot | 2011-09-15 17:54:46 +0200 (Thu, 15 Sep 2011) | 26 lines
  
  Merged revisions 336093 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r336093 | irroot | 2011-09-15 17:46:21 +0200 (Thu, 15 Sep 2011) | 20 lines
    
    
    Locking order in app_queue.c causes deadlocks.
    
    a channel lock must never be held with the queues container lock held.
    
    the deadlock occured on masquerade.
    
    the queues container lock is a relic of the past the old queue module lock.
    with ao2 there is no need to hold this lock when dealing with members this
    patch removes unneeded locks.
    
    (closes issue ASTERISK-18101)
    (closes issue ASTERISK-18487)
    Reported by: Paul Rolfe, Jason Legault
    Tested by: irroot, Jason Legault, Paul Rolfe
    Reviewed by: Matthew Nicholson
    
    Review: https://reviewboard.asterisk.org/r/1402/
  ........
................

Merged revisions 334455 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/10

................
  r334455 | irroot | 2011-09-06 15:58:56 +0200 (Tue, 06 Sep 2011) | 18 lines
  
  Merged revisions 334453 via svnmerge from
  https://origsvn.digium.com/svn/asterisk/branches/1.8
  
  ........
    r334453 | irroot | 2011-09-06 15:48:03 +0200 (Tue, 06 Sep 2011) | 13 lines
    
    
    Make SQL query in app_voicemail.c portable LIMIT is not portable.
    
    Regression from r312212
    
    (closes issue ASTERISK-18255)
    Reported by: Leif Madsen
    Tested by: Leif Madsen
    
    Review: https://reviewboard.asterisk.org/r/1415/
  ........
................