[asterisk-bugs] [Asterisk 0016664]: [patch] Random DTMF duplicate emulation on bridged OOH323 channel on outgoing calls

[Asterisk Bug Tracker]

A NOTE has been added to this issue. 
====================================================================== 
https://issues.asterisk.org/view.php?id=16664 
====================================================================== 
Reported By:                vmikhelson
Assigned To:                may213
====================================================================== 
Project:                    Asterisk
Issue ID:                   16664
Category:                   Addons/chan_ooh323
Reproducibility:            random
Severity:                   major
Priority:                   normal
Status:                     ready for testing
Asterisk Version:           SVN 
JIRA:                        
Regression:                 No 
Reviewboard Link:            
SVN Branch (only for SVN checkouts, not tarball releases): N/A 
SVN Revision (number only!):  
Request Review:              
====================================================================== 
Date Submitted:             2010-01-21 00:44 CST
Last Modified:              2010-03-03 17:27 CST
====================================================================== 
Summary:                    [patch] Random DTMF duplicate emulation on bridged
OOH323 channel on outgoing calls
Description: 
In majority of cases DTMF tones are duplicated by Asterisk when bridging a
SIP or DAHDI channel with OOH323 channel on outgoing calls.

In case of a normal DTMF processing the following sequence is observed:

[Jan 20 19:59:56] DTMF[21084]: channel.c:2855 __ast_read: DTMF begin '9'
received on SIP/464-00000012
[Jan 20 19:59:56] DTMF[21084]: channel.c:2865 __ast_read: DTMF begin
passthrough '9' on SIP/464-00000012
[Jan 20 19:59:56] DTMF[21084]: channel.c:2783 __ast_read: DTMF end '9'
received on SIP/464-00000012, duration 120 ms
[Jan 20 19:59:56] DTMF[21084]: channel.c:2823 __ast_read: DTMF end
accepted with begin '9' on SIP/464-00000012
[Jan 20 19:59:56] DTMF[21084]: channel.c:2839 __ast_read: DTMF end
passthrough '9' on SIP/464-00000012

In case of a problematic DTMF processing it looks like that:

[Jan 20 19:18:54] DTMF[20916]: channel.c:2855 __ast_read: DTMF begin '9'
received on SIP/464-00000011
[Jan 20 19:18:54] DTMF[20916]: channel.c:2865 __ast_read: DTMF begin
passthrough '9' on SIP/464-00000011
[Jan 20 19:18:54] DTMF[20916]: channel.c:2783 __ast_read: DTMF end '9'
received on OOH323/172.17.135.2:1720-9915, duration 0 ms
[Jan 20 19:18:54] DTMF[20916]: channel.c:2809 __ast_read: DTMF begin
emulation of '9' with duration 100 queued on OOH323/172.17.135.2:1720-9915
[Jan 20 19:18:54] WARNING[20916]: chan_ooh323.c:1054 ooh323_indicate:
Don't know how to indicate condition 20 on ooh323c_o_21
[Jan 20 19:18:54] WARNING[20916]: chan_ooh323.c:1054 ooh323_indicate:
Don't know how to indicate condition 20 on ooh323c_o_21
[Jan 20 19:18:54] DTMF[20916]: channel.c:2932 __ast_read: DTMF end
emulation of '9' queued on OOH323/172.17.135.2:1720-9915
[Jan 20 19:18:54] WARNING[20916]: chan_ooh323.c:1054 ooh323_indicate:
Don't know how to indicate condition 20 on ooh323c_o_21
[Jan 20 19:18:54] DTMF[20916]: channel.c:2783 __ast_read: DTMF end '9'
received on SIP/464-00000011, duration 120 ms
[Jan 20 19:18:54] DTMF[20916]: channel.c:2823 __ast_read: DTMF end
accepted with begin '9' on SIP/464-00000011
[Jan 20 19:18:54] DTMF[20916]: channel.c:2839 __ast_read: DTMF end
passthrough '9' on SIP/464-00000011
[Jan 20 19:18:55] WARNING[20916]: chan_ooh323.c:1054 ooh323_indicate:
Don't know how to indicate condition 20 on ooh323c_o_21

====================================================================== 

---------------------------------------------------------------------- 
 (0118916) may213 (manager) - 2010-03-03 17:27
 https://issues.asterisk.org/view.php?id=16664#c118916 
---------------------------------------------------------------------- 
Vladimir, 

i did not look sources of ooh323 version 0.9 while but i think it still
not usable in asterisk due to single thread model of ooh323c stack and this
model was general reason for making rework channel with modifications of
stack code more than asterisk channel code. It's possible that Objective
Systems changed threading model of their stack then implementation channel
driver based on 0.9 code will new and big work and will be finished with
new version of chan_ooh323 ;)
Also i don't see what serious enhancement is needed for signalling part of
code.

IMHO, chan_ooh323 versions before current trunk aren't usable for
production due to 1thread and synchronous stack code. Any trouble with
network proccesing cause delay in ooh323 stack processing. These delay can
be very big for real-time. I don't speak about many memory leaks and other
bugs in ooh323 code. 

Issue History 
Date Modified    Username       Field                    Change               
====================================================================== 
2010-03-03 17:27 may213         Note Added: 0118916                          
======================================================================