[asterisk-bugs] [Asterisk 0013747]: Indications are not passed from old peer to new peer during masquerade

[Asterisk Bug Tracker]

A NOTE has been added to this issue. 
====================================================================== 
http://bugs.digium.com/view.php?id=13747 
====================================================================== 
Reported By:                davidw
Assigned To:                putnopvut
====================================================================== 
Project:                    Asterisk
Issue ID:                   13747
Category:                   Core/Channels
Reproducibility:            always
Severity:                   major
Priority:                   normal
Status:                     acknowledged
Asterisk Version:           1.6.0 
SVN Branch (only for SVN checkouts, not tarball releases): N/A 
SVN Revision (number only!):  
Disclaimer on File?:        N/A 
Request Review:              
====================================================================== 
Date Submitted:             2008-10-20 13:04 CDT
Last Modified:              2008-10-22 09:23 CDT
====================================================================== 
Summary:                    Indications are not passed from old peer to new peer
during masquerade
Description: 
We want to transfer a channel that is already indicating (typically
ringing) to a parked channel, which we do with AMI Originate.  When we do
this, the person on the parked channel has MOH changed to silence until the
ringing channel answers, or otherwise changes indication.

It looks to me as though there may have been some attempt to support this,
but it looks like it copies the indications at the wrong end of the
masquerade.
======================================================================
Relationships       ID      Summary
----------------------------------------------------------------------
related to          0012548 SIP channel protocol illegally reverses...
====================================================================== 

---------------------------------------------------------------------- 
 (0094130) davidw (reporter) - 2008-10-22 09:23
 http://bugs.digium.com/view.php?id=13747#c94130 
---------------------------------------------------------------------- 
This is messier than I thought; I got false hopes from the channel.c code.

By the same token that there isn't a useful indication for the channel
being masqueraded, because it will be re-starting in the PBX (not sure if
the new bridging operations differ here), there won't actually be a new
peer yet.  What one actually needs to do is to send the indication state of
the old peer to the new peer when it exists, and is in a fit state to
receive it. One consequence of this, I think, is that one needs to store
visible indication states for both directions (or just the other one!)
against a channel. 

Issue History 
Date Modified    Username       Field                    Change               
====================================================================== 
2008-10-22 09:23 davidw         Note Added: 0094130                          
======================================================================