| Summary: | ASTERISK-24467: STUN Request is not completing from ice_reset_session | ||
| Reporter: | Ajay Choudary (AjayG) | Labels: | STUN WebRTC |
| Date Opened: | 2014-10-29 02:23:26 | Date Closed: | 2014-12-03 09:32:58.000-0600 |
| Priority: | Major | Regression? | |
| Status: | Closed/Complete | Components: | Resources/res_rtp_asterisk |
| Versions: | 11.12.0 11.13.0 | Frequency of Occurrence | Frequent |
| Related Issues: | |||
| Environment: | CentOS-6.5 (Both Asterisk & stund) | Attachments: | |
| Description: | {color:blue} *Issue is related to WebRTC Clients like SIPML5*.{color}
After accepting the call in client (sipML5 + Chrome), again we are calling _ast_stun_request_ {{(ast_rtp_ice_start --> ice_reset_session -->ice_create --> rtp_add_candidates_to_ice --> ast_stun_request )}}. 1) From _ast_stun_request_ function we are sending STUN Request to stun server, and waiting for a Response. 2) By this time Client is already sending huge number of STUN Binding Requests/Responses to Same RTP/RTCP port (_Which should be handled by pjnath_) 3) STUN Server is sending Response immediately along with Binding requests from client. But Service is busy in processing STUN Binding Requests from client 4) So here _ast_stun_request_ is waiting for Stun Server Response for about 20 Seconds in a loop, by this time call is getting ring timedout. {color: brown} *Note:* If we comment _ice_reset_session_ function from _ast_rtp_ice_start_,Then no delay is observed in call connection. But this will break hold functionality {color} | ||
| Comments: | By: Matt Jordan (mjordan) 2014-10-29 09:58:40.822-0500 We require a complete debug log to help triage the issue. This document will provide instructions on how to collect debugging logs from an Asterisk machine for the purpose of helping bug marshals troubleshoot an issue: https://wiki.asterisk.org/wiki/display/AST/Collecting+Debug+Information By: Rusty Newton (rnewton) 2014-11-14 16:54:08.410-0600 [~AjayG] we haven't heard from you in a couple weeks. Can you provide the requested debug that we need to continue investigation of the issue? By: Rusty Newton (rnewton) 2014-12-03 09:32:58.647-0600 [~jcolp] looked at this one. He indicates that this is expected behavior with the current code and that to fix this would require implementation of Trickle ICE. Therefore this currently falls under a feature request with no patch. I'm going to close this out for now as Not a Bug. | ||