Summary: | ASTERISK-18201: Asterisk should fall back to AVP when SRTP module is not loaded and both SAVP and AVP have been offered | ||||||
Reporter: | Fabio Torchetti (mrwho) | Labels: | |||||
Date Opened: | 2011-07-28 06:26:37 | Date Closed: | 2015-03-14 23:59:15 | ||||
Priority: | Major | Regression? | |||||
Status: | Closed/Complete | Components: | Channels/chan_sip/Interoperability | ||||
Versions: | 1.8.5.0 | Frequency of Occurrence | Constant | ||||
Related Issues: |
| ||||||
Environment: | Asterisk 1.8.5, Nortel NSR | Attachments: | ( 0) fallback_srtp.patch | ||||
Description: | When both SRTP and RTP (RTP/AVP and RTP/SAVP) are offered Asterisk should fall back to the RTP protocol if it fails to load the SRTP module. Up until 1.6 Asterisk ignored the SRTP requests and - if available - would fall back to RTP; this is a regression test failure in an environment where SRTP is not to be used, even if it's offered. This is a sample INVITE SDP content of such type: v=0 o=- 189845755 1 IN IP4 XXX.XXX.XXX.XXX s=- t=0 0 m=audio 5262 RTP/AVP 0 8 101 111 c=IN IP4 XXX.XXX.XXX.XXX a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-15 a=rtpmap:111 X-nt-inforeq/8000 a=ptime:20 a=maxptime:20 a=sendrecv m=audio 5262 RTP/SAVP 8 0 101 111 c=IN IP4 XXX.XXX.XXX.XXX a=crypto:1 AES_CM_128_HMAC_SHA1_80 inline:dnUo0FOvQmZF+QJqoT/JlsrcyjFyiDZe5IDM/V 6V|2^031|003007014426:004 a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-15 a=rtpmap:111 X-nt-inforeq/8000 a=ptime:20 a=maxptime:20 a=sendrecv | ||||||
Comments: | By: Fabio Torchetti (mrwho) 2011-07-28 06:28:03.806-0500 First approach to solve the issue. Keep track of offered AVP and use that if SAVP fails. By: Gregory Hinton Nietsky (irroot) 2011-07-29 03:25:33.656-0500 Please see ASTERISK-17895 id like to work this into the patch im busy with there https://reviewboard.asterisk.org/r/1173/ Thank You By: Fabio Torchetti (mrwho) 2011-08-08 05:44:55.938-0500 I'm looking into it, it would be good to get your patch resolve both issues at the same time. By: Tomo Takebe (tomo1657) 2013-04-09 14:59:03.019-0500 By the way, I noticed that the patch checks for if ( clear_video_offered ) during the audio part. This should be: if ( clear_audio_offered ) By: Matt Jordan (mjordan) 2015-03-14 23:59:15.193-0500 Closing this out as a duplicate of ASTERISK-23989. The patch on this issue should be compared with Olle's when it is put up for peer review. |