A message from Treyarch?!!

[iAmMoDBoX]

 

[ColonicBoom]

lol, you've scared them!

[CleanSlate]

keep us updated MoD

[iAmMoDBoX]

So he's basically saying that ICMP isn't reliable.

[Zennon]

Which can be true, as a sever starts to get more traffic it will make PING request's the lest priority.

[fuzzy clam]

That's good stuff,when devs start asking question about servers

[iAmMoDBoX]

Which can true, as a sever stars to get more traffic it will make PING request's the lest priority.

 

Comcast told me the same story, it ended up being a water leak into the node in my neighborhood.

[iAmMoDBoX]

Not to bump an old thread or anything, but this problem is STILL  going on and I've been fighting with Comcast that it is on their end. They tried to fix everything except for admitting that their network is the problem. I got fed up and went a different route which lead me to getting a phone call from a "senior level network tech" who opens with "Hello this is Mark from Comcast calling in regards to your slow internet speeds." So right off the bat I knew this was going to be fun.

 

I proceed to answer him with "No my internet SPEEDS are fine, so if that's why you called then I'm afraid this problem is going to be much too complicated for you to understand or fix because it's way above your pay-grade."

 

Him: "Um okay, sir, well tell me what seems to be going on and maybe I can help"

 

Me: "Sure, no problem" I go into detail explaining the whole story from the problems I was having last year, to my pingplot results as well as DSL reports smoke ping tests which all conclude that the last hop in my router to my local game servers, which is Comcast owned, is having massive ping spikes at the exact same time every day as well as huge amounts of jitter"

 

Him: "Wow, I never heard of this before"

 

Me: "I figured, which is why I said this is above your pay-grade"

 

Him: "Um okay, let me see what I can do here.... Can you email me everything you have so I can forward it to someone else who can better diagnose this issue for you?"

 

So I send him an email stating this:

I've been running the program called PingPlotter for the past 2 years since I got my Netduma R1 router which was showing huge ping spikes while gaming. The problem turned out to be a amp in the area that was leaking water into it and causing signal issues for the area. That problem was fixed last summer after about 4 months of fighting and arguing. I ended up having 7 modems at my house at once because for about 2 weeks straight I had a tech at my house who would say "I don't know why they sent me, but do you want me to change the modem?" so I would keep it just to show the next tech in hopes someone would wake up and realize the problem wasn't in my house.... Anyways, on to my new problem.

 

For the past 4 months, possibly longer, I have noticed that during certain times my game traffic is having a higher ping and jitter. This happens at peak times, so I figured it was bad servers the game was being hosted on. I was actually contacted by one of the developers of the game personally to try to figure out where the problem was because on their end the servers were perfect and it seemed to be an issue only on my end. After they looked over all my tests we both came to the conclusion that hop 8 in my traces to the servers IP addresses was bad and that was causing my pings to go crazy. Those bad Comcast owned hops are 50.242.151.70 and 50.242.150.38.

 

The game puts me in the closest server to me geographically which is in Northern New Jersey. ALL of my game traffic passes through one of these Comcast hops and causes my game to suffer because of it. I've been going in circles trying to talk to someone who can fix this problem. I talked to the person on twitter support who fixed the issue last year with the bad amp and even they are giving me the run around on this issue. I have tons and tons of ping plots and trace routes that show the problems with these IPs as well as running a smokeping on here to them shows it.

 

https://www.dslreports.com/smokeping?target=cf764c54ded169ead94cbaf1c2709c34

 

https://www.dslreports.com/smokeping?target=d21b487e33ad501dcf678fa6f292de22.Michigan

 

and I sent him about 50 screenshots of all my pingplotter results to different NJ PS4 BO3 servers that all have a different IP address but all have the same exact graph because they all pass through the bad comcast hop.

 

His reply to my email is this:

This is the response from our engineer’s.

 

The screenshots and description of the issue that the customer is providing indicates that this issue is not originating due to Comcast equipment. In all provided examples the customer response time is less than 200  ms (.2 of a second) which is perfectly acceptable when attempting to use any connected internet service. The reported hop failure of 50.242.151.70 is not an actual indication of a routing failure as this is a DMZ firewall server handing traffic off to the external network associated with choopa.net, all of these DMZ servers rate limit ICMP (Internet Control Message Protocol) traffic such as ping tests and will not always return detailed information during ping tests due to the extreme amount of traffic they handle (especially during peak hours) as this would add unnecessary overhead to the packets. As the customers packet successfully travels out of the Comcast network to the choopa.net servers there is nothing which we can address from our end. Furthermore, the customer tests do not start to indicate routing or communication issues until the leave the Comcast network. As the customer states this issue is only occurring during peak hours the likely cause here is the obvious one  that this is being caused by elevated network traffic during those times. Once again this will not be something we can address and is likely due to the external backbone networks having elevated traffic during those hours rather than an issue with either our network, or with the game servers the customer is connecting to. Additionally, the ping tests this customer is conducting appear to last for upwards of 12 hours at a time and as these are unnecessary traffic the networks involved may be rate limiting the traffic as these are non-essential or junk packets.

 

My reply back:

That's all well and good, but that's just a typical copy/paste reply I got last year when I was having ping spikes to 9999 in ping plotter (saying it's not a problem on our end, it's just ICMP limiting), ended up being a problem in the area that was fixed after months of complaining... The problem with their theory of ICMP limiting is that I'm using UDP or TCP packets to ping, which is the same protocol game traffic uses and I'm seeing these increased ping times. Plus, when in game I can correlate the higher ping on the plots to the lag in my game.

 

Also, if they can say 200ms ping is acceptable for gaming, then someone else needs to look at this problem. Preferably someone who understands how gaming works.

 

Just as proof that it's not ICMP rate limiting I have attached a picture that shows PingPlotter on the top part (using TCP) using port 3074 and the bottom part is www.dslreports.com smoke ping which is using ICMP and both graphs are IDENTICAL over a 24 hour period to 50.242.151.70

 

Also attached is a zoomed in 5 minute time period of when the ping is at it's highest. This shows how much jitter there is which is being carried down the trace to my game sever making online play almost impossible because if you know anything about gaming you know that you need a low and stable ping.

 

 

I have yet to get a reply back lolol

[fuzzy clam]

That sucks MoD,of course nothings ever the ISP's issue  

 

Wish you the best of luck getting that shit sorted out.

[ColonicBoom]

"if that's why you called then I'm afraid this problem is going to be much too complicated for you to understand or fix"

 

Crushed him.

[abc123]

just so you know you can run pings with TCP or UDP: https://wiki.itadmins.net/network/tcp_udp_ping

 

This will eliminate their bullshit about IMCP being lower priority...i hate morons.

 

200ms isn't acceptable for even VOIP...so WTF are they talking about

[iAmMoDBoX]

just so you know you can run pings with TCP or UDP: https://wiki.itadmins.net/network/tcp_udp_ping

 

This will eliminate their bullshit about IMCP being lower priority...i hate morons.

 

200ms isn't acceptable for even VOIP...so WTF are they talking about

 

 

Just as proof that it's not ICMP rate limiting I have attached a picture that shows PingPlotter on the top part (using TCP) using port 3074 and the bottom part is www.dslreports.com smoke ping which is using ICMP and both graphs are IDENTICAL over a 24 hour period to 50.242.151.70

 

Also attached is a zoomed in 5 minute time period of when the ping is at it's highest. This shows how much jitter there is which is being carried down the trace to my game sever making online play almost impossible because if you know anything about gaming you know that you need a low and stable ping.

 

The last two pictures

[abc123]

I hate when ISPs fight their customers so hard when they are just trying to help

[iAmMoDBoX]

I hate when ISPs fight their customers so hard when they are just trying to help

 

They are just in denial or lazy. You can ping the two IPs yourself from anywhere in the world and see the same exact ping patterns

[DumaRocks]

How did you obtain the IP's for their servers. I want to ping the server I am constantly getting connected too but the IP in the upper right hand of the screen says in ping plotter it can't be resolved. 

[abc123]

How did you obtain the IP's for their servers. I want to ping the server I am constantly getting connected too but the IP in the upper right hand of the screen says in ping plotter it can't be resolved. 

 

he likely did a trace route, what this does is sets the time to live on a packet to 1 hop, then 2 hops, then 3 hops, then 4 hops, then X hops, then hits target (doing three samples on each hop).

 

Example

tracert google.com

Would give me:

1    <1 ms    <1 ms    <1 ms  192.168.1.1
  2     8 ms     8 ms     7 ms  142.254.147.49
  3    10 ms     8 ms     8 ms  ae63.lsvqkydb02h.midwest.rr.com [74.128.7.165]
  4    14 ms    15 ms    14 ms  be22.lsvmkyzo01r.midwest.rr.com [65.29.31.38]
  5    19 ms    19 ms    18 ms  be24.clmkohpe01r.midwest.rr.com [65.189.140.162]
  6    29 ms    30 ms    27 ms  bu-ether25.chctilwc00w-bcr00.tbone.rr.com [107.14.19.16]
  7    27 ms    32 ms    30 ms  bu-ether11.chcgildt87w-bcr00.tbone.rr.com [66.109.6.20]
  8    35 ms    27 ms    29 ms  216.3.52.41
  9    28 ms    27 ms    28 ms  216.3.52.62
 10    28 ms    27 ms    27 ms  209.85.243.29
 11    27 ms    26 ms    27 ms  209.85.246.195
 12    36 ms    37 ms    35 ms  209.85.143.214
 13    44 ms    49 ms    41 ms  216.239.48.160
 14    42 ms    42 ms    42 ms  216.239.49.77
 15     *        *        *     Request timed out.
 16    42 ms    53 ms    41 ms  qh-in-f102.1e100.net [74.125.22.102]

This shows me that the total ping time is an avg of 49ms ( 42 + 53 + 41 / 3 ) = 48.66666666666667

 

I can use this to see if one of the hops is causing a ton of latency by monitoring the ping time.

[DumaRocks]

he likely did a trace route, what this does is sets the time to live on a packet to 1 hop, then 2 hops, then 3 hops, then 4 hops, then X hops, then hits target (doing three samples on each hop).

 

Example

tracert google.com

Would give me:

1    <1 ms    <1 ms    <1 ms  192.168.1.1
  2     8 ms     8 ms     7 ms  142.254.147.49
  3    10 ms     8 ms     8 ms  ae63.lsvqkydb02h.midwest.rr.com [74.128.7.165]
  4    14 ms    15 ms    14 ms  be22.lsvmkyzo01r.midwest.rr.com [65.29.31.38]
  5    19 ms    19 ms    18 ms  be24.clmkohpe01r.midwest.rr.com [65.189.140.162]
  6    29 ms    30 ms    27 ms  bu-ether25.chctilwc00w-bcr00.tbone.rr.com [107.14.19.16]
  7    27 ms    32 ms    30 ms  bu-ether11.chcgildt87w-bcr00.tbone.rr.com [66.109.6.20]
  8    35 ms    27 ms    29 ms  216.3.52.41
  9    28 ms    27 ms    28 ms  216.3.52.62
 10    28 ms    27 ms    27 ms  209.85.243.29
 11    27 ms    26 ms    27 ms  209.85.246.195
 12    36 ms    37 ms    35 ms  209.85.143.214
 13    44 ms    49 ms    41 ms  216.239.48.160
 14    42 ms    42 ms    42 ms  216.239.49.77
 15     *        *        *     Request timed out.
 16    42 ms    53 ms    41 ms  qh-in-f102.1e100.net [74.125.22.102]

This shows me that the total ping time is an avg of 49ms ( 42 + 53 + 41 / 3 ) = 48.66666666666667

 

I can use this to see if one of the hops is causing a ton of latency by monitoring the ping time.

 

Forgive my ignorance but how can you do a trace route resulting in that displaying their IP addresses from your console. Netduma lets you do a trace route when your connected to their servers  but just a graphical one. Part of the reason I asked before is I wanted to do a trace route from computer using the ping plotter but I have no idea what IP address to attempt to do that too to access the COD servers in IA. Now I am more curious than anything but is there a way to figure out IA servers IP so I can do a ping plotter from my computer without displaying them on a forum. Not a big deal if not but just curious. 

 

[abc123]

Forgive my ignorance but how can you do a trace route resulting in that displaying their IP addresses from your console. Netduma lets you do a trace route when your connected to their servers  but just a graphical one. Part of the reason I asked before is I wanted to do a trace route from computer using the ping plotter but I have no idea what IP address to attempt to do that too to access the COD servers in IA. Now I am more curious than anything but is there a way to figure out IA servers IP so I can do a ping plotter from my computer without displaying them on a forum. Not a big deal if not but just curious. 

 

open command line (windows) or terminal (mac os x)

Then just run the above command