Summary:	ASTERISK-16201: [patch] valid IP address in RTP offer when Asterisk is attached to several networks
Reporter:	under (under)	Labels:
Date Opened:	2010-06-03 09:38:30	Date Closed:	2011-07-26 15:28:54
Priority:	Major	Regression?	No
Status:	Closed/Complete	Components:	Channels/chan_h323
Versions:		Frequency of Occurrence
Related Issues:
Environment:		Attachments:	( 0) multinet2.diff
Description:	Currently Asterisk sends host IP address (the one that is get from /etc/hosts or via DNS, function ast_find_ourip()) in RTP media offer of H225. This is not acceptable when Asterisk is attached to several networks (for example, via several NICs, or IP aliases), because this leads to no voice in worst cases (depends on network setup). For this case Asterisk should be bind to 0.0.0.0 and should place correct media IP address in media offer that corresponds to the IP address of NIC/network the call is sent from. SIP channel already handles this case correctly. What we need is the same handling in H323 channel. ****** ADDITIONAL INFORMATION ****** The patch attached has been tested in all combinations of fastStart on/off for both calling and called party.
Comments:	By: Leif Madsen (lmadsen) 2010-06-03 09:45:58 Updated all instances of * to actually say Asterisk as it is hard to ready short form like that. By: under (under) 2010-06-03 09:54:23 uploaded multinet2.diff as the first patch contents redundant code By: ddkprog (ddkprog) 2010-06-07 14:39:25 >under check my patch to fixx your problem https://issues.asterisk.org/view.php?id=12685 By: under (under) 2010-06-11 16:26:38 I don't see invocation of ast_ouraddrfor() in your patch. Without it there is no logical way to fix this problem without extra config options in file. By: ddkprog (ddkprog) 2010-06-11 17:11:02 you just do not understand the H323 protocol and openh323 library we can check this very simply you collect chan_h323 with my patch put bindaddr = 0.0.0.0 in the config h323.conf that and TCS  would correctly transfered RTP address your problem will be properly corrected By: under (under) 2010-06-14 02:23:49 ddkprog, with your patch RTP port in asterisk will never be bind to 0.0.0.0 (see ast_rtp_new_with_bindaddr() in chan_h323.c), would it? By: ddkprog (ddkprog) 2010-06-14 04:51:55 under, sure? check it if you have no knowledge for a proper patch, then do not offer them it makes no sense to discuss something if you're not an expert in this By: under (under) 2010-06-14 05:32:22 ddkprog, I don't understand what is the reason that you post comments that way. 1) you don't seem to be willing to help to understand your patch and to find possible issues that it could cause and whether it's better than mine patch or not. 2) You have the same status as me (reporter) and therefore you are not the man who judges either to accept my patch or not. 3) I find your messages to be offensive, therefore I will not continue any discussion with you. Good luck. By: Leif Madsen (lmadsen) 2010-06-14 13:55:32 Dropping this from Ready for Testing to Confirmed as there is a patch here, but there is discussion about the approach. Unfortunately ddkprog has chased off the other interested party. By: iye (iye) 2010-08-02 03:11:49 I have problem with RTP on 1.6.2.9, when I use more than one interface. That patch fixes that problem. By: Russell Bryant (russell) 2011-07-26 15:28:49.217-0500 Per the Asterisk maintenance timeline page at http://www.asterisk.org/asterisk-versions maintenance (bug) support for the 1.4 and 1.6.x branches has ended. For continued maintenance support please move to the 1.8 branch which is a long term support (LTS) branch. For more information about branch support, please see https://wiki.asterisk.org/wiki/display/AST/Asterisk+Versions If this is still an issue, please open a new issue so it can be re-triaged appropriately. Thanks!