No RTP after server change

I have a OLD CentOS server with asterisk 18.5.0 (chan_sip) running works fine.
I try to migrate to a new UBuntu 22.04 asterisk 18.18.0 (chan_sip) that does not pass RTP

SIP device for paging - is 10.11.1.30 (exten 601)
the server is 10.11.1.15
the SIP PBX (non-asterisk) is 10.11.1.1

simple application - they pic ip SIP phone dial a number - comes into the server and we dial the extension and connect the call.

on the OLD server I “rtp set debug on” and I see RTP from 10.11.1.30
on the new server I do the same and NO RTP data from 10.11.1.30
I see RTP data in both cases from 10.11.1.1 and 10.11.1.15

What do I look for? Its all simple setup (typically)

[601]
type=friend
defaultname=601
defaultuser=601
secret=XXXXX
dtmfmode=RFC2833
host=dynamic
description=SH-PA-01
context=smvoice-sip
rtptimeout=60
rtpholdtimeout=60
rtpkeepalive=60
callerid=601
qualify=no
canreinvite=yes
timezone=0
nat=no
disallow=all
allow=ulaw
allow=alaw
allow=gsm

I COPIED all the old config file from teh old server to the new for asterisk. It should be the same.

Thanks,

Jerry

  • Double-check firewall rules on the new server. Ensure UDP ports 5060 (SIP) and the RTP port range (default 16384-32767) are open for bi-directional traffic between the paging device and Asterisk.

  • Use iptables -L or ufw status to list firewalls and iptables -D INPUT/OUTPUT -p udp --source 10.11.1.30 --destination 10.11.1.15 --dport 5060/16384-32767 -j ACCEPT to add temporary rules if needed.

  • Compare rtp.conf and sip.conf files on both servers meticulously, line by line. Look for differences in settings like localnet , nat , canreinvite , rtpstart , rtpend , icesupport , and media-related options.

  • Adjust settings on the new server to match the old server based on your network setup. Test changes incrementally.

    • If the above steps don’t help, capture network packets using tools like tcpdump or Wireshark on both servers and the paging device. Analyze the captured packets to identify where RTP packets are dropped or misinterpreted.

Thanks for the thoughts. I compared everything - all was good - tcpdump showed no traffic with calling the one device 601. However if I called multiple devices RTP ran just fine. even for 601.

So we left it calling multiple endpoints (6 in all) and it works- calling just the one had no RTP still.
Crazy strange.

jerry

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.