@hitesh wrote:
Thanks again. Do you do this to the Input Rule Sets or Output Rule Sets?
Input rule.
@memelbert wrote:No, there is a problem in Verizon's network. I have been running tracroutes for several days now and finally got a Verizon Tech supervisor, Jason, to admit there is a problem. Below are the results from "MTR" which is a continous traceroute.
HOST: new-host-11.home Loss% Snt Last Avg Best Wrst StDev
1.|-- wireless_broadband_router 0.0% 100 1.5 1.9 0.8 25.5 3.0
2.|-- l1100.pitbpa-vfttp-51.ver 0.0% 100 4.6 7.9 4.3 42.2 5.8
3.|-- g12-0-9-451.pitbpa-lcr-04 0.0% 100 7.6 8.6 4.8 30.4 3.1
4.|-- p5-0-0.pitbpa-lcr-03.veri 0.0% 100 13.4 27.7 11.6 208.7 39.5
5.|-- so-15-0-3-0.res-bb-rtr1.v 0.0% 100 13.7 18.1 11.7 102.4 14.6
6.|-- 0.xe-7-0-0.xl3.iad8.alter 0.0% 100 11.8 17.0 11.5 46.9 7.1
7.|-- tengige0-6-1-0.gw7.iad8.a 0.0% 100 17.2 18.3 14.2 48.5 6.0
| `|-- 152.63.37.82
| |-- 152.63.37.78
| |-- 152.63.32.190
| |-- 152.63.37.154
| |-- 152.63.37.150
8.|-- google-gw.customer.alter. 10.0% 100 83.3 86.6 80.2 125.4 8.3
9.|-- 216.239.46.250 0.0% 100 16.4 18.8 12.8 96.9 13.2
10.|-- 72.14.238.175 0.0% 100 14.5 16.1 14.0 33.2 2.8
11.|-- iad23s06-in-f4.1e100.net 0.0% 100 15.1 16.5 14.0 68.8 5.8You can see at hop 8, google-gw.customer.alter.net, there is 10% packet loss. I always see packet loss at this hop in some form. Earlier tonight it was at 25% packet loss. google-gw.customer.alter.net is part of Verizon's backbone and IS part of their network.
Jason said not enough customers had called yet in the past week to justify a outage ticket to be opened. This is completely ludacris. As a sidenote, the Teir 1 tech support tried to upsell me service telling me that would help, not when you see packet loss like above. The only thing that will help that is to fix the problem, or route around it. <sigh>
That supervisor is not knowledgeable if he said that.
google-gw.customer.alter.net
This is a Google Peering point, that they set up with Verizon and paid for. So they called verizon, said they need access to the network, this is how much bandwidth they will need, and verizon set it up according to googles requirement. IOW they are Verizon Customers too.
If google is getting hammered, or experiencing difficulty on their network, you will see it represented in higher ping times, and latency and jitter. They need to increase their bandwidth.
Google is acting just like you in this scenario. They are calling up verizon as a customer and saying sign me up for xyz. Verizon sets it up.
Now if you start downloading and pushing more traffic than your network can handle with the price plan you purchased, what do you do? you call and ask for more, or you make adjustments to your network to compensate. Same thing here.
Also, here is a permalink to a similar discussion that outlines what you are seeing in more detail. Also the filter rule described allows you to take a different route, that's why most users have had such great success with it.
@Hubrisnxs wrote:That supervisor is not knowledgeable if he said that.
google-gw.customer.alter.net
This is a Google Peering point, that they set up with Verizon and paid for. So they called verizon, said they need access to the network, this is how much bandwidth they will need, and verizon set it up according to googles requirement. IOW they are Verizon Customers too.
If google is getting hammered, or experiencing difficulty on their network, you will see it represented in higher ping times, and latency and jitter. They need to increase their bandwidth.
Google is acting just like you in this scenario. They are calling up verizon as a customer and saying sign me up for xyz. Verizon sets it up.
Now if you start downloading and pushing more traffic than your network can handle with the price plan you purchased, what do you do? you call and ask for more, or you make adjustments to your network to compensate. Same thing here.
Also, here is a permalink to a similar discussion that outlines what you are seeing in more detail. Also the filter rule described allows you to take a different route, that's why most users have had such great success with it.
This was the second night in a row that I called. The first night I spoke with Terry, a Teir-1 tech, and after he spoke with his network engineer said there was no problem with this Verizon owned router, but did say that this a Verizon owned peice of equipment. Last night when I called in, Tony, the Teir-1 tech told me that upgrading my service would solve this issue and that Verizon did not own alter.net. After much consternation I was finally able to put through to Jason, Tony's supervisor. Jason admitted that there is a problem with google-gw.customer.alter.net, and they have been seeing problems for some time now (at least a week), and Verizon techs have been "tinkering" with that system, although he did not have any more specifics since a outage ticket was not opened.
Do you work for Verizon? It would seem so based on your comment. Otherwise, can you tell us how you know the details of this specific peering point?
I wonder why Verizon techs would be "tinkering" with the router if all it requires is some extra bandwidth.......
No, not needed, it's just clear that he isn't abreast of the situation. Also peering points are common knowledge for IT professionals, at the links provided you'll find a plethora of such individuals including Tier III VZ technical support and others.
It made the news at BBR and REDDIT
If you choose to ignore it, that's you're decision, but at the thread posted it clearly goes into the high level technical problem google and it's caching servers are having.
a quote from the news article reads:
Spend a few minutes in any of our forums and you'll find this is a universal problem with many carriers, including AT&T U-Verse, Verizon FiOS and Time Warner Cable.
I've just done the homework is all.
I just wanted to add that I used the firewall trick posted before and it worked instantly. I will say that I had to add it to BOTH my "Ethernet/Coax Rules" and my "Broadband Connection (Ethernet/Coax) Rules" for it to work. I wasn't sure which to use and I tried one and then the other and it didn't work, but when I added it to both it started working right away. I was actually mid load on a 25min 1080p video that was loading slow as heck, but as soon as I clicked "Apply" the rest of the video loaded in about 3-5 secs. Thanks for the fix. I just switched back to Fios (previous user that moved and moved back) and never saw anything like this before so I was really disappointed at first, but if this fix holds out I have nothing to complain about since gaming and other streaming content works great!
Wow. I did this on my Rev. I router (red router with two antennas). It actually improved performance on Youtube. I put these rules on the Input section for both Ethernet/Coax Rules and Broadband Connection (Ethernet/Coax) Rules. Thank you very much!
When you create your rule, did you use REJECT instead of DROP?
Drop waits for a timeout and then drops, REJECT flat out terminates it, and there is no lag with that one.
@gbfraz wrote:
Im a little behind since all the replies but I just tried the custom router settings posted earlier by Hibrusnxs. im still having the same probs when watching youtube. He mentions you can still have probs and when you do, get the ip of the video your watching and add it the same way. So how do you find out what those new numbers are?
Start a new topic or ask questions in the open forum.
