[asterisk-bugs] [Asterisk 0014041]: IAXy runs ok for a while, then goes bonkers

[Asterisk Bug Tracker]

A NOTE has been added to this issue. 
====================================================================== 
http://bugs.digium.com/view.php?id=14041 
====================================================================== 
Reported By:                jcovert
Assigned To:                russell
====================================================================== 
Project:                    Asterisk
Issue ID:                   14041
Category:                   Channels/chan_iax2
Reproducibility:            random
Severity:                   major
Priority:                   normal
Status:                     assigned
Asterisk Version:           1.4.21.2 
Regression:                 No 
SVN Branch (only for SVN checkouts, not tarball releases): N/A 
SVN Revision (number only!):  
Request Review:              
====================================================================== 
Date Submitted:             2008-12-09 09:18 CST
Last Modified:              2009-01-18 23:35 CST
====================================================================== 
Summary:                    IAXy runs ok for a while, then goes bonkers
Description: 
After up to a month, or as little as a day or two, both the IAXy I
originally purchased, and a replacement given to me by Stephen Burcham in
October (to try to eliminate a specific piece of hardware) become
unreliable.  In some cases it just stops processing calls, in the more
likely case, documented in the attached log, it goes into a
reachable/unreachable loop like this:

[Dec  9 10:01:45] NOTICE[26712]: chan_iax2.c:8832 __iax2_poke_noanswer:
Peer 'x29' is now UNREACHABLE! Time: 5
[Dec  9 10:01:45] NOTICE[26712]: chan_iax2.c:7912 socket_process: Peer
'x29' is now REACHABLE! Time: 10
[Dec  9 10:02:29] NOTICE[26712]: chan_iax2.c:8832 __iax2_poke_noanswer:
Peer 'x29' is now UNREACHABLE! Time: 10
[Dec  9 10:02:29] NOTICE[26712]: chan_iax2.c:7912 socket_process: Peer
'x29' is now REACHABLE! Time: 13

Since it provides service to the phone next to the main place at which
"she who expects phones to work" generally sits, this causes me a major
headache.
====================================================================== 

---------------------------------------------------------------------- 
 (0098106) jcovert (reporter) - 2009-01-18 23:35
 http://bugs.digium.com/view.php?id=14041#c98106 
---------------------------------------------------------------------- 
I'm now running with the test patch (against 1.6.0.3) that I've uploaded. 
This is the point in routine socket_process where packets with scallno==0
are about to be thrown away (the "return 1").

/john 

Issue History 
Date Modified    Username       Field                    Change               
====================================================================== 
2009-01-18 23:35 jcovert        Note Added: 0098106                          
======================================================================