NO JOY 2/4 phones getting "congestion" for incoming calls - configurations nearly identical

[Ben Mannino]

Hi Everyone,

I have learned a lot over the past couple of months through extensive trial and error, but I'm stuck at this point and would appreciate any guidance to help get these phones ringing.

So here is my config:
- Four identical Linksys SPA941 phones
- Four stores, each using the same Internet Provider (1 static IP) and the same model Watchguard firewall
- FreePBX 2.11 Asterisk 11.2.1 running on an Amazon EC2 instance. This was setup using the guidelines here (http://nerdvittles.com/?p=5060)

Each phone was given a static IP on each store's LAN (172.168.x.222) and all four stores can communicate with each other through the Branch Office VPN setup on the Watchguard firewalls.

I configured extensions 201 - 204 for each store through the FreePBX GUI and chose the default options for the most part. Here are a few of the key config values:

dtmfmode: RFC 2833
canreinvite: no
host: dynamic
trustrpid: yes
sendrpid: no
type: friend
nat: yes
port: 5060
qualify: yes
qualifyfreq: 30
transport: UDP only

I may have toggled some of these values while troubleshooting, but these are the current values for all four extensions.

I ordered 4 DIDs from Vitelity and setup the inbound/outbound trunks as indicated on their help pages, and setup an inbound route for each number to the corresponding extension.

I upgraded the firmware on all four phones to the latest version, and basically just used the web config page to set the proxy address for ext 1 to the public IP of the EC2 instance, the username and password to match the extension. I left everything else with their default values.

After each phone rebooted, it registered on the server successfully. I do see four phone registrations on the FreePBX system status page. It also shows 2 IP trunks online and 1 IP trunk registration. While digging around, I went to Reports -> Asterisk Info and then chose the Peers link on the right side of the page.

It shows all four extensions, 201-204 with a status of OK, and the Host columns shows the public IP address of each store. What's interesting is that two of the peers (202,203) are using Port 5060 and two are using 1024 (201,204). Also, after each OK status message, there is a number in parenthesis. Ext 201 has a very high number (1054 ms) and the remaining three are 43 - 48 ms.

Outbound calling worked fine on all phones. When I try calling each DID number, the two locations using port 5060 rang through fine, and the two with 1024 gave me a busy signal followed by the following emails from Vitelity:

A call to your DID xxxxxxxxxx from yyyyyyyyyy has failed at 5:56pm on 02/03/2015 MDT. We received 'CONGESTION' when attempting to route the call to your server or device.

I tried rebooting the phones and the server, but still get the same results. I called Vitelity and they basically told me it's something on my end, which I'm sure it is, but I can't seem to narrow it down. The only real differences in each store configuration is the public static IP address, and possibly the firmware versions of the Firewalls. I haven't touched any of the firewalls when setting all four phones up.

Your assistance would be greatly appreciated,
-Ben

 

[Ben Mannino]

Update on the current config...

So every once in a while, I try calling each of the Vitelity DID numbers to see if they do through. I have had a few times where all of phones rang, and others where one or two will give me a busy signal.

Checking the peer info just now, I noticed that one of the two phones that previously showed up as connecting via port 1024 is now on port 5060. It still produced a busy signal however, and the email I received from Vitelity now has a different message:

"A call to your DID xxxxxxxxxx from yyyyyyyyyy has failed at 8:13am on 02/09/2015 MDT. We received 'CHANUNAVAIL' when attempting to route the call to your server or device."

The call quality is not that hot either over two of the calls that were actually picked up in the stores. So, do you think it's safe to assume that this is purely a network issue? We have Comcast Business in all four stores and the speed is decent - really didn't think it would be a problem.

Is anyone using Amazon EC2 to host their server? Vitelity for their trunks? Maybe one of these (or both) is the problem. I was thinking of two other things to try:

1) Bring the server in-house at one of the stores for all four phones to connect locally over the VPN

2) Set up a virtual server at each store, running four separate instances of PIAF.

Call reliability and quality are obviously of the utmost importance, and I'm willing to setup whatever configuration would yield the best results.

Any guidance would be greatly appreciated,

-Ben