Tracert connection issues only on certain servers

Technical Support
1

I’ve done everything on my end: refreshed IP, flushed dns, checked for driver and graphics card updates… I even went so far as to do a complete restore on my PC and redownload everything. None of this, even the over the top stuff, has resolved the issue. After spending hours combing through the forums it seems you guys (support staff) always want winMTR reports. Yet I can see something very clear with just my tracert findings:

Opening up a bot game and checking the IP connection of my current game (24.105.45.19) gives me a tracert that times out over 10 hops and does not come back within the given 30 hops. It fails at as57976

Using the IP given for US central (24.105.62.129) the hops fail after 9 hops again. Interestingly this also fails at as57976 which if you google (try it) it only comes up with other blizzard games experiencing this very issue on this very domain.

However, for some reason, when I tracert to the US West server, the tracert is completed in 13 hops. There is a high ping in this instance, yet at least it eventually connects, not only to as57976 but it actually finishes in, like I said, 13 hops.

I see you doing a few things to “solve” this issue mostly. Your main fix is to request people use an ethernet cord and not be on wireless, sometimes you tell them to try to flushdns and/or reset IP. Usually the go to though is foregoing any blame from Blizzard itself and telling us to contact our ISPs. Rarely do you state using a vpn would help, which I suppose it can given the inconsistencies from US Central versus US West.

However, AS57976, is you. It is owned by Blizzard Entertainment Inc. All other hops are fine until it gets to you, the Blizzard domain. Is there anything being done about this on your end Blizzard?

Frequent forum agents and Tech support, Nicole and Kershew can you please address what this consistent, longstanding problem happening on your, Blizzards, end? What is this weird proxy server Blizzard owns and uses and why does it encounter so many frequent problems?

Edit: 3 Years ago Nicole stated “The spikes on that hop are due to packet mitigation to protect against DDoS.” Given that they aren’t simply spiking on that hop, rather not being continued further, leads me to believe that Blizzards DDoS protection wrongly identifies regular people (myself included) trying to connect to the game they want to play as an attack. Is this correct?

2

I don’t work at Blizzard. I do recall saying what you quoted, and it still stands today. High ping spikes and no response can be the same thing if the server is refusing some pings. This will be more obvious in a tracert since it only tests the connection in one direction. Not sending data roundtrip isn’t very useful in terms of diagnosing a problem, and that’s why a WinMTR is the go-to.

As a fellow player, I can confidently say that most connection problem posts show the issue occurring with the ISP or the peering companies they contract to route data. When Blizzard has a connection problem, there is a sudden flood on the forum and ticket system. Add on to that how Blizzard’s network team monitors the connections 24/7, and it’s hard to understand why 1–3 posts per day blame the gaming server.

What’s the actual problem you’re running into, other than the test results being vague? Blizzard doesn’t want you to directly ping that server node, but it doesn’t mean their own software can’t connect through it.

1 Like
3

Alright I’m running a bunch of WinMTRs each for about 200 pings each.

  1. In game listed ISP (24.105.45.19) had an immediate packet loss of 20% on the second hop (10.125.64.1) which was considerably lower throughout until reaching as57976 where it then again returned to 20%. Decent ping overall, that only topped out at 82. https://pastebin.com/k0wQ4kVD

  2. OW Default Central ISP (24.105.62.129) had a little hiccup of 3% in the first hop and then 23% on the second(10.125.64.1) which fluctuated up and down a a bit until reaching as57976 ending at 24%. The ping was decent overall and only once spiked to 77. The average excluding as57976 was in the 20s. (Forgot to save this one, but can rerun if needed) https://pastebin.com/5xjjvhWt

  3. OW Default Western ISP (24.105.30.129) had a massive loss for the first packets sent starting in the high 58%s. After 200 pings the loss settles to an average of 28% (10.125.64.1) which subsided down to 12% at desired ISP that is listed from OWs Default Western page.https://pastebin.com/32z5eNUx

Overall, I do notice a considerably high packet loss throughout that I was unaware of prior to just tracert. Notably the packet loss is always on the 2nd hop and is consistent throughout. I tried to do a whois of (10.125.64.1) but I couldn’t find anything relevant other than that it is private. Pinging (10.125.64.1) directly leads to a 19% packet loss out of 200.

I sincerely apologize for saying that as57976 was the issue itself when I noticeably have lots of packet loss when using WinMTR. I’m still not sure why as57976 doesn’t process to the intended domain on US Central, yet it does in US West to the requested OW IP. Is there an explanation for this?

Other noticeably weird pings:

I pinged my Network Interface Card (NIC) aka IPv4 address and it gives an initial spike of 16% packet loss at the same (10.125.64.1) but winds up at an odd IP (204.98.170.138). Doing a WHOIS of that hostname states that it is owned by CenturyLink, yet I use Cox. Is this normal?

Sending 100 pings on my loopback address (127.0.0.1) show 0% loss and 0 ping.

Pinging 1.1.1.1 and 1.0.0.1 is a ~80% packet loss for both. +/- 5

I tried to establish a baseline by pinging www.google.com and it is also abysmal. At 275 pings it encounters a 48% loss with an absurd amount of 100% loss/no response until it is resolved at a 34%. https://pastebin.com/EsuD4a7g

Once again, I apologize. The issue has gotten considerably worse during the daytime. The disconnects prior seemed to be only kicking me off overwatch. I didn’t notice any problems streaming twitch, youtube, or anything else then. Today, I am constantly disconnecting on all platforms and for minutes at a time. I haven’t slept for a couple of days and also considerably irritable due to my disorder. Further investigation only proves that there is severe packet loss on my end for some reason.

Is any of this indicative of hardware issue or network clogging and what shows you either?
I’ve called my ISP and informed them about the severe packet loss, but I would like to know how to interpret this information better to know exactly what is wrong. My ISP is sending someone out on Monday and no matter what they’re going to try to swindle me into something superfluous and not needed. They already backed me into a corner of having to pay for advanced tech support for 90 days over the phone or have a tech come out for free. They refused to elaborate further on my packet loss issues over the phone for free, even when they also pinged my router to confirm it happening. Any and all information regarding WinMTR would help. Interpretations of what I’ve already tracked, or further things to look for.

I’ll edit this post with pastebin links of the relevant pings shortly when I’m on my desktop.

4

All of your WinMTRs (including the one to Google) are showing the same thing, data loss starting practically at the start of your end of the connection. I am going to say you need to check your own local computer, router, and modem first and make sure they are working correctly. Make sure no one else (or any other device or program) is using the same internet connection as you. If all else fails, you should reach out to your ISP for assistance.

5

I think there is some confusion about the data you’re looking at. These nodes on the 2nd and 3rd hops are in your local and ISP network, not the server’s network, and not even in close proximity to it. All of those IPs along the way are different networks communicating with each other.

It’s the line coming in or local hub in your neighborhood serving internet to your home. It doesn’t have a named address, only a number.

Correct, that’s the same thing shown in the test. The connection between your home and your internet source in the neighborhood is very poor and needs to be addressed by the ISP.

Short answer: Blizzard configured those hops differently. Nothing you can do to change how it responds to your pings.

That is not your NIC, that is the ISP line I explained above. It might be on your block, your city, we can’t tell you exactly where because we don’t work at the ISP. Your NIC is 127.0.0.1 (this is not listed in a WinMTR for anyone).

They are acknowledging there is an issue at the modem, so until that is addressed don’t bother yourself with running and analyzing tests to servers across the country. You need a stable home network before you do so.

Remedying this packet loss on your ISP line will likely improve everything across the board.

For education: Check the number of packets you sent from the PC to the modem in the first test (210) and compare it to the number of packets the server received (150). Then you look at the first hop showing data loss in the % column, and you’ll see it’s the ISP line you’re trying to ping.

6

Thank you so much! I’ve been having fun and playing around with this winMTR thing a bit because I can’t play overwatch until this is resolved and I’ve encountered something weird happening occasionally. Sometimes at the start of the test I encounter huge packet loss that goes down over time. But what I’m noticing is that (10.125.64.1) is sending more packets than it is given, which is weird because it’s the first hop after me. I believe you said it was the first internet source in my neighborhood after me.

Here’s what happens at first while pinging google: https://pastebin.com/PdEwganc

Eventually it subsides a bit overt time seen here while pinging an OW server: https://pastebin.com/gTxceapE

Here’s some more of it happening here pinging blizzard.com https://pastebin.com/Q1fqvVSM

Now here’s a really weird one. When I ping facebook it sends over 700 packets immediately: https://pastebin.com/gULg38N7

When pinging the same place twice in a row though, this never happens: https://pastebin.com/UHERhfH4

What exactly is going on here?!

They are acknowledging there is an issue at the modem , so until that is addressed don’t bother yourself with running and analyzing tests to servers across the country. You need a stable home network before you do so.

I know you said don’t bother. But it’s become fun for me.

Edit: I’m trying to trigger the HUGE responses again.

Instagram/reddit had a normal amount of 20-30 extra initial packets sent.

Tripadvisor freaked it out just like facebook and sent over 700 packets: https://pastebin.com/0EpzRt3V

So then I tried deltas airlines own website and got over 150: https://pastebin.com/457Atbpj

Am I crazy for thinking (10.125.64.1) is something not good?

7

I don’t know how all these other random servers and websites are configured, so I won’t be able to comment on those questions. Plus, this ventures way outside the scope of this forum. Connection troubleshooting is offered as a courtesy by Blizzard in the ticket system, and on a volunteer basis in the forum from community members/fellow players like me.

What matters here is that your ISP is noticing data loss when pinging your modem, so we know the source of the issues, and they need to address it for you.

8

I don’t know how all these other random servers and websites are configured, so I won’t be able to comment on those questions. Plus, this ventures way outside the scope of this forum.

Fair enough. Take care and thanks for the help

9

You might find this site about the MTR function useful: Diagnosing Network Issues with MTR | Linode.

10

Thanks I’ve read it over and after analyzing reports there is significant and immediate latency from my router while the extra packets from residential one send, which can be seen in this prior report: https://pastebin.com/0EpzRt3V

What I’ve gathered from this article is what I’m experience is not normal. This is not what it looks like when providers to rate limit. However, residential gateways can sometimes cause misleading reports, which possibly could be whats happening?

I have bipolar which sometimes gives me psychotic features, so. I can understand if this is all in my head. I already know I’m crazy in general. But please just at least tell me, is this normal?

Edit: I’ve been informed hop two is my ISP’s cable modem termination system or CMTS

11

Again, the data in the tests is not normal. There is data loss the ISP needs to address.

12

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.