[asterisk-bugs] [Asterisk 0017692]: [patch] subchannel remains half-open after call transfer

Asterisk Bug Tracker noreply at bugs.digium.com
Thu Jul 29 16:57:33 CDT 2010


A NOTE has been added to this issue. 
====================================================================== 
https://issues.asterisk.org/view.php?id=17692 
====================================================================== 
Reported By:                jmhunter
Assigned To:                
====================================================================== 
Project:                    Asterisk
Issue ID:                   17692
Category:                   Channels/chan_skinny
Reproducibility:            always
Severity:                   minor
Priority:                   normal
Status:                     ready for testing
Asterisk Version:           SVN 
JIRA:                       SWP-1960 
Regression:                 No 
Reviewboard Link:            
SVN Branch (only for SVN checkouts, not tarball releases):  trunk 
SVN Revision (number only!): 278578 
Request Review:              
====================================================================== 
Date Submitted:             2010-07-23 18:02 CDT
Last Modified:              2010-07-29 16:57 CDT
====================================================================== 
Summary:                    [patch] subchannel remains half-open after call
transfer
Description: 
Call transfers don't seem to work properly using chan_skinny.

To reproduce:

 [ jmhtest1 is a Skinny phone (Cisco 7905G)   ]
 [ 5022 is a DAHDI-connected analogue handset ]

1. jmhtest1 picks up handset and calls 6001 (music on hold)
2. jmhtest1 presses 'Transfer'
3. jmhtest1 dials 5022
4. 5022 picks up
5. jmhtest1 presses 'Transfer'
6. jmhtest1 replaces handset in cradle

At this point, jmhtest1 still shows "Connected" on its display, and now
won't give out a dialtone when handset is picked up. Pressing the 'EndCall'
softkey leads to messages such as:
[Jul 23 23:26:21] WARNING[19173]: chan_skinny.c:1673
find_subchannel_by_instance_reference: Could not find subchannel with
reference '1' on 'jmhtest1'
 Received Softkey Event: End Call(1/1)
[Jul 23 23:26:22] WARNING[19173]: chan_skinny.c:1673
find_subchannel_by_instance_reference: Could not find subchannel with
reference '1' on 'jmhtest1'
 Received Softkey Event: End Call(1/1)

Hanging up 5022 (the recipient of the call transfer) does not make any
difference, jmhtest1 still does not work and I have to reboot the phone.

Trying a blind transfer (missing out step 5 above) seems to be a bit hit
and miss - this either works fine, or places the call on hold. It's done
both for me at various times, I think it's to do with whether Asterisk has
crashed recently (see https://issues.asterisk.org/view.php?id=17680).

In addition, at step 3 above, I can hear the audio from the original call
to 6001, as well as the audio from 5022 (where I am transferring to). That
surely shouldn't happen - jmhtest1 should talk to 5022 and only 5022 at
that point, right?
====================================================================== 

---------------------------------------------------------------------- 
 (0125340) jmhunter (reporter) - 2010-07-29 16:57
 https://issues.asterisk.org/view.php?id=17692#c125340 
---------------------------------------------------------------------- 
Dan - you have made me very happy :-)

Applied v4 patch, and my first transfer was entirely successful! No
Asterisk crash, and the phone knew the subchannel had gone!

That's excellent - thank you very much.

Is the patch OK for regular use? I'll be setting up the production server
with these phones next week, so will apply this patch, and the
https://issues.asterisk.org/view.php?id=17680
patch, and report back after a few days of actual usage..

Thank you again! 

Issue History 
Date Modified    Username       Field                    Change               
====================================================================== 
2010-07-29 16:57 jmhunter       Note Added: 0125340                          
======================================================================




More information about the asterisk-bugs mailing list