Brief audio loss ~5s into each call

Asterisk Asterisk Support
1

Wildcard TE-133
Asterisk-13.7.2
FreePBX 13.0.95
Libpri-1.4.15
Dahdi-2.10.2
~25 Polycom 430s
Using chan_sip

Each call whether coming in over the PRI or ext <> ext loses audio (on the far side (callee)) for about 5s. This loss of audio happens almost exclusively about 3-5s from the beginning of the call.

Once audio is re-established the call continues with no further issues.

The issue occurs whether coming from a queue, ring group, or ext <> ext.
The phones are all on a local network, gigabit POE switch, and a centOS linux router acting as the gateway. I have fiddled with the NAT settings thinking this may have been the issue to no avail.

From what I am reading it looks like chan_sip uses 5061 now? I have adjusted the phone configs to use 5061 as well as the extension’s Advanced --> Port config. I have fiddled with the NAT settings thinking this may have been the issue to no avail.

I am still seeing the following:
`<— SIP read from UDP:10.10.10.101:5060 —>
SIP/2.0 200 OK
Via: SIP/2.0/UDP 10.10.10.7:5061;branch=z9hG4bK667b80b5
From: “Unknown” sip:Unknown@10.10.10.7:5061;tag=as42c9cc2e
To: sip:200@10.10.10.101;tag=93A48426-D2E23C45
CSeq: 102 NOTIFY
Call-ID: 333e113a428637e3739f78305fc378e2@10.10.10.7:5061
Contact: sip:200@10.10.10.101
Event: message-summary
User-Agent: PolycomSoundPointIP-SPIP_601-UA/3.1.1.0137
Accept-Language: en
Content-Length: 0

<------------->`

My suspicion is that I am losing audio when the RTP stream is setup/switched over?

[200] deny=0.0.0.0/0.0.0.0 secret=XXXXXXXXXXXX dtmfmode=rfc2833 canreinvite=no context=from-internal host=dynamic trustrpid=yes sendrpid=no type=friend nat=no port=5061 qualify=yes qualifyfreq=60 transport=udp,tcp,tls avpf=no force_avp=no icesupport=no encryption=no namedcallgroup= namedpickupgroup= dial=SIP/200 mailbox=200@default permit=0.0.0.0/0.0.0.0 callerid=Front Desk <200> callcounter=yes faxdetect=no

I can send/pb any logging output you need. I don’t see anything that sticks out in the debug log.
Any help would be appreciated.

2

My guess is some NAT device or firewall is relearning the allowable data flows.

3

I would tend to agree but the issue occurs between phones on the same LAN (no weird IP scheme, everything is on a local /24). I actually have no phones registering ‘through’ the firewall as everything is local and calls are pushed out over the PRI.

Iptables is off on the phone server and the FreePBX firewall thing is disabled as well.

4

Is the hostname of your machine present in the hosts file of your asterisk box?

5

It was not, I have added it. Would that matter? I will not be able to test till later but does asterisk do a lookup based on fqdn or hostname?

[root@ucs ~]# ping ucs
PING localhost (127.0.0.1) 56(84) bytes of data.
64 bytes from localhost (127.0.0.1): icmp_seq=1 ttl=64 time=0.029 ms
^C
— localhost ping statistics —
1 packets transmitted, 1 received, 0% packet loss, time 501ms
rtt min/avg/max/mdev = 0.029/0.029/0.029/0.000 ms
[root@ucs ~]#