[asterisk-dev] [Code Review] Fix Dial m and r options generating warnings for voice frames.

rmudgett reviewboard at asterisk.org
Thu Mar 8 12:24:17 CST 2012


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/1805/
-----------------------------------------------------------

(Updated March 8, 2012, 12:24 p.m.)


Review request for Asterisk Developers and Mark Michelson.


Changes
-------

Address Mark and Matt's comments.
Also reexamined the control hold/unhold handling.


Summary
-------

Incoming calls that do not use ulaw will generate a bunch of warning messages and possibly fail when a DAHDI channel is then called if the Dial m or r options are used.

You get:
[Mar 6 20:39:05] WARNING[2994]: chan_dahdi.c:9061 dahdi_write: Cannot handle frames in g722 format
[Mar 6 20:39:05] WARNING[2994]: app_dial.c:1410 wait_for_answer: Unable to forward voice or dtmf

A bandaid patch was applied to silence the above warnings for each frame by setting up a translation path for future frames.  The bandaid patch also emits this warning:
Codec mismatch on channel %s setting write format to %s from %s native formats %s

This problem also occurs if the dial is forked to multiple destinations: Dial(DAHDI/1&DAHDI/2)


This addresses bugs ASTERISK-16901 and ASTERISK-17541.
    https://issues.asterisk.org/jira/browse/ASTERISK-16901
    https://issues.asterisk.org/jira/browse/ASTERISK-17541


Diffs (updated)
-----

  /branches/1.8/apps/app_dial.c 358691 
  /branches/1.8/main/channel.c 358691 

Diff: https://reviewboard.asterisk.org/r/1805/diff


Testing
-------

Without patch, the codecs are not setup when Dial options m or r are used and the SIP codec is not ulaw.

With the patch, the voice frames are not passed and the diagnostic messages are not generated.


Thanks,

rmudgett

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20120308/39b53fca/attachment.htm>


More information about the asterisk-dev mailing list