Vonage and Comcast troubleshooting notes
I had a client recently get Vonage Business. We’re using a PFSense which is in my honest opinion, “Enterprise Class”. It scales like no other for the price!
I’m going to document my troubleshooting, more to come. My client was experiencing voice jitter, voice fading and call dropping on occasion.
The first thing I did was a trace route. Linux that command is: traceroute hostname, Windows it’s: tracert hostname. In my example I trace routed to the SIP provisioning IP / host name of Vonage which at the time of writing is: prov.vocalocity.com.
I did the trace route on CTS Telecom (Fiber), Charter and Comcast. The interesting thing is that on CTS and Charter it worked from start to finish!
I’m removing the first 5 hops so you don’t know where I live :P Stalkers!
Charter HOME connection trace:
traceroute to prov.vocalocity.com (184.108.40.206), 30 hops max, 60 byte packets
6 lag-200.ear3.Chicago2.Level3.net (220.127.116.11) 50.797 ms 42.233 ms *
7 ae-1-3501.edge4.Chicago3.Level3.net (18.104.22.168) 41.505 ms ae-14-3604.edge4.Chicago3.Level3.net (22.214.171.124) 20.111 ms 20.375 ms
8 126.96.36.199 (188.8.131.52) 21.629 ms 21.607 ms 21.618 ms
9 cr2-tengig-0-5-3-0.chd.savvis.net (184.108.40.206) 19.501 ms 50.857 ms 57.366 ms
10 220.127.116.11 (18.104.22.168) 76.101 ms 74.702 ms 75.813 ms
11 hr1-te-1-0-0.atlantaat1.savvis.net (22.214.171.124) 74.757 ms 72.764 ms 74.133 ms
12 das1-v3006.at1.savvis.net (126.96.36.199) 64.993 ms 64.200 ms 64.097 ms
13 188.8.131.52 (184.108.40.206) 62.473 ms 52.542 ms 51.231 ms
14 * * *
15 * * *
16 * * *
17 * * *^C
…it never finishes!
Notice I preface this with a HOME connection…why? Because the business Charter is DIFFERENT!
Charter BUSINESS trace:
Tracing route to prov.vocalocity.com [220.127.116.11]
over a maximum of 30 hops:
6 18 ms 15 ms 15 ms bbr01aldlmi-bue-1.aldl.mi.charter.com [18.104.22.168]
7 * * * Request timed out.
8 30 ms 22 ms 22 ms ae-2-3601.edge4.Chicago3.Level3.net [22.214.171.124]
9 20 ms 22 ms 32 ms 126.96.36.199
10 21 ms 22 ms 21 ms cr2-tengig-0-5-3-0.chd.savvis.net [188.8.131.52]
11 38 ms 38 ms 38 ms 184.108.40.206
12 37 ms 38 ms 38 ms hr1-te-1-0-0.atlantaat1.savvis.net [220.127.116.11]
13 50 ms 45 ms 49 ms das1-v3005.at1.savvis.net [18.104.22.168]
14 40 ms 44 ms 55 ms 22.214.171.124
15 40 ms 38 ms 38 ms 126.96.36.199
16 37 ms 41 ms 38 ms 188.8.131.52
It finishes and the ROUTES are different!
Charter must route their traffic differently based on home user or business.
Comcast was much like the Charter failures for home users:
10 184.108.40.206 (220.127.116.11) 52.204 ms 55.201 ms 48.346 ms
12 das1-v3006.at1.savvis.net (18.104.22.168) 49.561 ms 49.876 ms 56.578 ms
13 22.214.171.124 (126.96.36.199) 53.821 ms 58.421 ms 46.651 ms
14 * * *
15 * * *
16 * * *
17 * * *
I had to actually call Comcast tier 1 tech support – yes, the people who just tell you to reboot your router. Then, after discussing it with the tier 1 tech he logged into the modem I was on and did his own trace route. He said, it works for me. It finishes. I said, really – what IP address did you resolve to and get to? He told me and it was some IP address in the Netherlands! I then insisted he get this escalated because who in their right mind would do VOIP from the midwest to servers across the Atlantic ocean to the Netherlands? That makes no sense.
You have to push these guys to THINK!
After submitting my ticket, the trace routes FINALLY work like they should.
I can’t say this fixed my call quality issues but I’m still troubleshooting. These things take time BUT don’t EVER rule out the ISP’s network as possibly causing the problem. This is the 2nd time and ISP has had the problem with routing on their backbone that has affected service. The other time I dealt with Frontier about their backbone in Muskegon, MI – they were great about it and I actually talked to the guy who fixed it out of New York. I did an IP Who Is lookup which lists the maintainers email (that never works) and the telephone of the maintainer – which told me to call another number, then after a brief phone tree battle I got to the guy who fixed it!
I also updated / forced the phones on the network to use Google DNS servers, 188.8.131.52 and 184.108.40.206. I am trying that rather than the local DNS server and in favor of using Comcast’s which is 220.127.116.11 and 18.104.22.168. Comcast (believe it or not) DNS servers have 150 ms + ping times OMG! Google DNS servers; something like 20 ms or less, maybe a touch over BUT they’re fast. I don’t know if that will make a difference to the phone quality (probably not) but it’s another thing I’m touching and changing.
QoS might be next BUT once your traffic hits the public InternetS all bets for QoS are off. QoS is really for your local network and routing but my client is barely touching their WAN – they’re too busy working and nobody steams or uses the Internet for more than work.
I’ll update this as I move forward, I thought about SIP proxy with the PFSense and QoS on that SIP Proxy per some tutorials on the PFSense forum but haven’t gone there yet.