[SR-Users] SDP ending with CRLF and a=nortpproxy:yes
Narsay, Deep
deep.narsay at mezocliq.com
Tue Aug 12 17:49:02 CEST 2014
Hello,
Please let me know if there is any work around for this.
Details of the issue:
kamailio.cfg has this sequence in kamailio.cfg,
......
modparam("rtpproxy", "ice_candidate_priority_avp",
"$avp(ice_priority)" )
........
sdp_remove_line_by_prefix("a=candidate");
set_iattr("ice_priority", "1");
rtpproxy_manage("frwie");
I'm seeing an issue towards the bottom of the SDP message.
....
a=ssrc:738345272 mslabel:VPELrlXvdfGmtevlYLEeVm6zq4JHvI4g
a=ssrc:738345272 label:VPELrlXvdfGmtevlYLEeVm6zq4JHvI4ga0
a=ice-ufrag:YJuJWr9clIJHhENw
a=ice-pwd:3v1D1ShCg8xIlDZ9
a=candidate:1297581420 1 udp 659136 10.1.2.3 17534 typ host
generation 0
a=candidate:1297581420 2 udp 659136 10.1.2.3 17535 typ host
generation 0
a=ptime:20
\r\n
a=nortpproxy:yes\r\n
The call gets rejected by UA due to an invalid (empty) line in SDP.
Very old but somewhat similar issue:
http://lists.sip-router.org/pipermail/users/2008-September/019605.html
(Module authors, I'll be happy to code/test or assist in any way possible
if needed.)
Just to test it, I commented the following code snippet and the call
worked.
But that can't be the permanent fix.
Thank you,
- Deep N
Code snippet:rtp_proxy.c::force_rtp_proxy starting from line 2850
* if (proxied == 0 && nortpproxy_str.len) { cp =
pkg_malloc((nortpproxy_str.len + CRLF_LEN) * sizeof(char));
if (cp == NULL) { LM_ERR("out of pkg memory\n");
return -1; } anchor =
anchor_lump(msg, body.s + body.len - msg->buf, 0, 0); if
(anchor == NULL) { LM_ERR("anchor_lump failed\n");
pkg_free(cp); return -1;
} memcpy(cp, CRLF, CRLF_LEN);
memcpy(cp + CRLF_LEN, nortpproxy_str.s, nortpproxy_str.len);
if (insert_new_lump_after(anchor, cp, nortpproxy_str.len + CRLF_LEN, 0) ==
NULL) { LM_ERR("insert_new_lump_after failed\n");
pkg_free(cp); return -1;
} }*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-users/attachments/20140812/6a85ba52/attachment.html>
More information about the sr-users
mailing list