[asterisk-bugs] [Asterisk 0018583]: Codec negotiation fails on IAX calls from 1.8.1.1 to 1.8.1.1
Asterisk Bug Tracker
noreply at bugs.digium.com
Sat Jan 8 02:07:39 UTC 2011
A NOTE has been added to this issue.
======================================================================
https://issues.asterisk.org/view.php?id=18583
======================================================================
Reported By: jcovert
Assigned To:
======================================================================
Project: Asterisk
Issue ID: 18583
Category: Channels/chan_iax2
Reproducibility: always
Severity: major
Priority: normal
Status: new
Asterisk Version: 1.8.1.1
JIRA:
Regression: No
Reviewboard Link:
SVN Branch (only for SVN checkouts, not tarball releases): N/A
SVN Revision (number only!):
Request Review:
======================================================================
Date Submitted: 2011-01-06 08:41 CST
Last Modified: 2011-01-07 20:07 CST
======================================================================
Summary: Codec negotiation fails on IAX calls from 1.8.1.1 to
1.8.1.1
Description:
In the example below, an originating 1.8.1.1 system with the ONLY change in
iax.conf being changing one line from bandwidth=low to bandwidth=high is
not able to connect to a terminating 1.8.1.1 system with the only CODEC
change relevant to a guest user being (a) the change of bandwidth=low to
bandwidth=high and (b) codecpriority=caller inserted.
An originating 1.6.1.6 system with a configuration identical to the
failing originating 1.8.1.1 system is able to connect to the same
terminating 1.8.1.1 system, no problem.
======================================================================
----------------------------------------------------------------------
(0130324) jcovert (reporter) - 2011-01-07 20:07
https://issues.asterisk.org/view.php?id=18583#c130324
----------------------------------------------------------------------
Endianness is definitely the problem.
On BIG_ENDIAN machines, the ntohll and htonll functions do nothing, since
network byte order is big endian.
The fix is to still do the byte swap, even on BIG_ENDIAN machines (or at
least on Darwin PPC). The implication is that ntohll and htonll should not
have ever been called.
The patch takes care of this for Darwin. I do not have any other
BIG_ENDIAN machines to test against.
/john
Issue History
Date Modified Username Field Change
======================================================================
2011-01-07 20:07 jcovert Note Added: 0130324
======================================================================
More information about the asterisk-bugs
mailing list