Child Item

Site Totals

61432 Questions Answered 270 Categories

Verizon FIOS Latency Issue and the Syracuse Call Center

My Verizon FIOS latency recently jumped from 6-8 ms to 24-26 ms as measured by speedtest.net. I called in for assistance because traceroute shows the hop in the Verizon network with the delay. Rather than address the issue Mike a Network Tech from the Syracuse Call Center told me that FIOS only supports between 25 - 50 ms latency. To add to that he (falsely) claims 25-50 ms latency is the "best in the business". I've looked but I can't find any document that discusses FIOS's agreement with customers. Can someone point me in the right direction?

Is it just me or does everyone else have issues when they work with the Syracuse call center? It seems they spend more time trying to get Customers to "go away" rather than investigate an issue.

Jim

Comments

smith6612

Which speedtest.net server are you testing to? Speedtest.net isn't exactly the most reliable way to check your latency, simply due to the fact that the Flash application has a tendency to bog down the test more than it needs to be, and also since the servers can change routing/hosting providers at any given time. The true way to test your latency, at least in a basic sense is to perform a trace route. Try this:

Go to Start > Run > type cmd and press OK, or go to Start > Search for "cmd" and open cmd. Run the following command once Command Prompt comes up: tracert www.google.com

Once the Traceroute completes, Copy and Paste the results to this thread. You can do so by clicking on the Command Prompt Icon located in the Top Left corner of the cmd window, going down to Edit, and then pressing Select All. From there, press Enter, and then paste the traceroute into a post here.

Also, the best in the industry for latency is a fiber-based connection, but most FiOS connections won't really go below 6ms. This is due to the latency MoCa networking adds to the connection, and MoCa generally lands you in the 6-10ms latency range to Verizon's network; similar to Cable modem or DSL service. Under best case scenarios with FiOS and an Ethernet connection to the ONT, you can achieve 1ms latency after excluding NAT to Verizon. That's as good as it's ever going to get. Realistically you'll see 2-4ms latency while on Ethernet to Verizon's network, but that's how such things work out to be. Ethernet is a far superior medium to deliver a WAN connection to a router, but Verizon is using Coax simply due to convenience of wiring.

jimryan

I've had 6-8 ms latency until last week. Now I'm around 24-26 ms. All speedtest.net measurements which Verizon Customer Support asked me to use. I have a Ethernet not a MoCa connection. I see the problem but I'm not sure how to get Verizon to fix it ...

traceroute to 8.8.8.8 (8.8.8.8), 30 hops max, 60 byte packets

1 gw (192.168.1.1) 0.808 ms 1.153 ms 1.228 ms

2 L100.PGHKNY-VFTTP-16.verizon-gni.net (173.62.56.1) 3.281 ms 3.675 ms 4.774 ms

3 G3-0-8-216.PGHKNY-LCR-02.verizon-gni.net (130.81.180.76) 4.816 ms 5.506 ms 5.629 ms

4 P14-0-0.PGHKNY-LCR-01.verizon-gni.net (130.81.151.16) 22.631 ms 23.009 ms 23.411 ms

5 130.81.199.0 (130.81.199.0) 130.859 ms 130.989 ms 131.674 ms

6 0.xe-4-0-2.XL3.NYC4.ALTER.NET (152.63.10.61) 26.264 ms 25.191 ms 25.799 ms

7 TenGigE0-4-2-0.GW8.NYC5.ALTER.NET (152.63.16.73) 26.434 ms TenGigE0-6-2-0.GW8.NYC4.ALTER.NET (152.63.21.117) 20.003 ms TenGigE0-4-0-0.GW8.NYC4.ALTER.NET (152.63.21.61) 20.080 ms

8 Internet-gw.customer.alter.net (152.179.72.66) 20.300 ms 20.653 ms 21.254 ms

9 216.239.43.114 (216.239.43.114) 31.045 ms 209.85.252.215 (209.85.252.215) 22.525 ms 22.863 ms

10 72.14.239.93 (72.14.239.93) 32.400 ms 33.598 ms 209.85.249.11 (209.85.249.11) 51.359 ms

11 64.233.175.109 (64.233.175.109) 33.327 ms 72.14.236.200 (72.14.236.200) 34.341 ms 64.233.175.109 (64.233.175.109) 34.484 ms

12 72.14.232.21 (72.14.232.21) 34.632 ms 34.548 ms 216.239.49.145 (216.239.49.145) 30.010 ms

thegeek

If you call the FSC and report a 'latency' problem most of the Customer Support Agents won't have a clue what you are saying.

They are trained to troubleshoot specific problems, such as 'can't connect' or 'can't login to my verizon account'.

If you give them a description of the problem you are having, such as 'slow downloads from a web site' or something like that, then tell them about the latency issue, you'll stand a better chance of getting an technically sound answer.

Do you get the latency if you tracerout to more than one site? It may be a routing issue and the results will identify the bad hop.

idburns

@thegeek wrote:
Do you get the latency if you tracerout to more than one site? It may be a routing issue and the results will identify the bad hop.

FYI - 8.8.8.8 is one of Google's public DNS servers.

jimryan

Thanks for the responses. The request must have found its way to someone who did something. It appears that the route in hop 3 was changed to take another path which is cutting my latency measured by speedtest.net from 24-26 ms down to 12 ms. Still not the 6-8 I had but web response time has improved greatly and streaming music no longer times out.

However Sean at the FSC RI office is telling me that I should expect latencies to go up as Verizon takes on more customers and that latencies of 24-26 ms should be expected in the future.

Hmmm ... if that's truly the case I'll likely not be a customer for long.

traceroute to 8.8.8.8 (8.8.8.8), 30 hops max, 60 byte packets

1 gw (192.168.1.1) 0.810 ms 1.678 ms 1.884 ms

2 L100.PGHKNY-VFTTP-16.verizon-gni.net (173.62.56.1) 3.412 ms 3.810 ms 4.442 ms

3 G3-0-8-216.PGHKNY-LCR-02.verizon-gni.net (130.81.180.76) 4.549 ms 5.316 ms 5.488 ms

4 so-7-3-0-0.NY325-BB-RTR1.verizon-gni.net (130.81.29.100) 7.468 ms 8.159 ms 8.477 ms

5 0.so-0-0-0.XL3.NYC4.ALTER.NET (152.63.1.41) 8.918 ms 10.289 ms 10.390 ms

6 TenGigE0-4-1-0.GW8.NYC4.ALTER.NET (152.63.21.69) 11.047 ms 10.329 ms 10.357 ms

7 Internet-gw.customer.alter.net (152.179.72.66) 10.429 ms 4.578 ms 3.969 ms

8 209.85.252.215 (209.85.252.215) 4.615 ms 5.214 ms 216.239.43.114 (216.239.43.114) 5.842 ms

9 72.14.239.93 (72.14.239.93) 22.776 ms 209.85.249.11 (209.85.249.11) 14.935 ms 17.336 ms

10 64.233.175.109 (64.233.175.109) 23.923 ms 24.253 ms 24.821 ms

11 72.14.232.21 (72.14.232.21) 25.623 ms 25.878 ms 26.420 ms

12 google-public-dns-a.google.com (8.8.8.8) 20.128 ms 20.296 ms 20.673 ms

Comparing it to the original traceroute when delays were noticed it is a substanical improvement ....

1 gw (192.168.1.1) 0.808 ms 1.153 ms 1.228 ms

2 L100.PGHKNY-VFTTP-16.verizon-gni.net (173.62.56.1) 3.281 ms 3.675 ms 4.774 ms

3 G3-0-8-216.PGHKNY-LCR-02.verizon-gni.net (130.81.180.76) 4.816 ms 5.506 ms 5.629 ms

4 P14-0-0.PGHKNY-LCR-01.verizon-gni.net (130.81.151.16) 22.631 ms 23.009 ms 23.411 ms

5 130.81.199.0 (130.81.199.0) 130.859 ms 130.989 ms 131.674 ms

fibertechguru

I've heard that there are latency issues coming from google sites (and all their other channels like GoogleDocs and YouTube .. etc.) to the VZ network. I suspect that it's possible ERX issues from the central office so it should be fixed. Fios has the best data plan for the money you can get so I think you should be patient in this matter, I'm sure it will get resolved asap.

smith6612

@jimryan wrote:
However Sean at the FSC RI office is telling me that I should expect latencies to go up as Verizon takes on more customers and that latencies of 24-26 ms should be expected in the future.

Hmmm ... if that's truly the case I'll likely not be a customer for long.

For the short term (eg: 1-2 days) I would expect those kinds of latencies if they get a sudden surge of people consuming connections. However, a properly maintained network should not suffer such issues, and would be upgraded before such loads even begin to exist. Running backhaul near capacity is never a good thing for anyone. Both sides (ISP and Customer) wind up getting backlash as a result.

But if that is in fact true, it would devaluate what FiOS has going for it. It'd be essentially like an Interleaved DSL line but with more Jitter than DSL. Essentially you'd be buying a pipe based solely on throughput and not latency bundled into it, which makes no sense.

Child Item