Welcome to the Spectrum Community

Get answers, share solutions, help your community.

Start here

Unstable connection to Warframe servers

ZhainZhain Posts: 14
edited August 21 in Internet 2018 Archive Sep 06, 2018

Hello all,

I am in the process of trying to diagnose a connection problem, and my initial investigations seem to indicate that there is a problem somewhere around the Syracuse area for traffic routed between me (Upstate NY) and Warframe's servers (content.warframe.com, I believe in London, Toronto, Canada).   


I have gone through the steps of trying a direct wired connection, powercycling, ensuring all programs are on exceptions lists for AV, changing my DNS, etc.  I am at the point where I am checking WinMTR traces and wondering if I'll need to use a VPN to work around this.  Here is the winMTR trace:


WinMTR Statistics (destination content.warframe.com (

No response from host1003600000
No response from host1003600000


The no response from host continues from that point.  66.109 nodes seem to be TWC owned.  Any further suggestions would be greatly appreciated.  Thank you all for your time!

Best Answer

  • ZhainZhain Posts: 14
    Sep 12, 2018 Accepted Answer

    Well, as of a recent hotfix, I am no longer experiencing the issue, so the routing was either fixed or went somewhere else I would assume.


    Thank you all for your responses and suggestions, I appreciate your time!  I'll keep it all in mind if this happens again, I've learned a bit about diagnosing this sort of thing now. 


  • RAIST5150RAIST5150 Posts: 777 ✭✭✭✭
    Sep 06, 2018
    When you see hops consistently zero out it doesn't necessarily mean there is a problem... could just mean they are set to not respond to the request. It is actually fairly common practice.

    What is more telling is when a hop intermittently times out or has the odd really big lag spikes... often an indication utilization is likely spiking too high, or there may be a signalling issue somewhere along those segments if it is causing a lot of dropped packets. To that latter point, may want to look a bit closer to home for potential issues... like where you showed a peak latency of 472ms just a couple hops from home.

    May also want to check out the game's online support as well... looks like there were some issues they were trying to address from their end:

  • ZhainZhain Posts: 14
    Sep 06, 2018

    Hi RAIST5150, thank you for the reply!


    I have a ticket in with their support which has been sitting for about a week with no response yet, so I was attempting to diagnose the problem by myself. 


    A few interesting things to note, this problem is isolated entirely to Warframe, no issues browsing, other games, streaming, etc.  I also attempted to boot up a VPN, which when activated and placing me in Wash. D.C, can connect to the game and play with no issues.  This would (seemingly at least) back up the idea that a node is bad somewhere. 


    I have also participated in a thread in Warframe's forums where people are posting similar issues, all of us seem to have TWC as our ISP as the common factor.


    With all that in mind, and taking your response into account (the latency spike, etc.), what would you recommend doing next?  Is it possible to get a TWC tech to look at the network routing? 


    EDIT:  I should mention that there are times where I can get a complete trace to the destination without the zero-ing out.  The issue is intermittent but frequent.  I'll try to get one of those traces to post when I get home, but it won't be for several hours. 

  • RAIST5150RAIST5150 Posts: 777 ✭✭✭✭
    Sep 06, 2018
    It gets a bit screwy because of how support is tiered... following the normal path can send you up and down trough several layers of teams before it finally gets to the more administrative level that can look more closely at routing/peering issues. An unfortunate side effect of the sort of "hybrid" nature of a cable internet system.

    They will need to send someone to verify your local environment isn't contributing. Even though you may appear to be experiencing things further upstream, there can still be some troublesome factors closer to home that are making a normally smaller problem have a more profound impact. You may be able to bypass some of that legwork if the signal levels and logs look good enough though. There is a sticky about what I formation to provide, as well as another sticky about signal levels. Getting a good look at this data both during good and troubled times can go a long way towards demonstrating you may not be dealing with CPE issues, but rather upstream (or at least on the other side of the distribution point on the premises)--effectively knocking out a couple cycles of technician visits and escalations between the first two tiers.

    It isn't just about trying to skip straight to tier3 though. A lot of times there are in fact signs of things like bad coax/connectors/amps and such that allow ingress (or even egress in more extreme cases) that NEED to be addressed not only for yourself, but the betterment of the community at large. Unwanted noise in the system can have a ripple effect throughout the neighborhood depending on where it is generated and how bad it gets. So, regardless... if they are sending a tech to inspect thinvs, humor them as they may find something that eventually leads to improving things on a larger scale. Rare though it may be... It can and does happen once in a while.

    What you will want to do is document all you can that may be relevant. As best you can, save it as text based data so it can easily be copy/pasted into online communications. Sometimes images and pdf files just aren't practical... what they really need are the data points. So long as I can be provided in an easily readable format of sorts, they can more easily just forward it via social media and emails for quick review (plain text is sometimes just the fastest way to get the data in their hands). Running traces/pings in a CMD prompt and just copy/pasting the data into notepad and saving as text files to copy/paste into a message later can make good things happened when forwarded to the right person. Some aren't aware of the editing functions of that powerful tool... get to the menu by clicking the black icon in the corner. MARK lets you click/drag the mouse to highlight specific text like in a word processor... pressing the ENTER key copies highlighted text to clipboard. Running the CLS command before starting a test clears the screen if you want to use the SELECT ALL option instead.

    If in the course of tracking your packets, you are able to document a pattern where a hop with someone like level3 or XO.net is exhibiting delayed/dropped packets during high usage times or something... you now have porltentially actionable info that needs to get to tier3 for review. And then it is time to push through to them... either by requesting it directly over phone or through the community managers/social media paths online. Key things to make sure to mention are the issue of delayed/dropped packets that you have tracked and documented (paste in the data if you can), as well as the timeframe for when it happens and the region's involved (ie, you are having trouble getting through Syracuse from Watertown on the way to Austin, or something to that effect... art least identify the endpoints as best you can).

    I know... It sounds like a lot. And I know it can be a burden to those not comfortable with the process. But taking these measures goes a wealth of information that goes a LONG way towards getting the ball rolling. So much of the complaints give virtually nothing to go on these days. In some cases it is quite literally as bad as dropping your keys off at the mechanic and just saying "it's broke, fix it" and then walking away... expecting to get your car back like new 3 hours later. They NEED good, relevant info... all else can just be noise for the most part, as there is nothing giving them direction, much less so.ething potentially actionable.
  • ZhainZhain Posts: 14
    Sep 06, 2018

    That was a lot to take in, yes, but I appreciate the detailed response!


    What I will plan to do when I get home then is post my modem/signal info, requested post info, and see if I can get another trace to WF's servers both successful and not.  Is that a decent place to start?  I can also grab a trace from when I use the VPN, since that seems to work. 

  • RAIST5150RAIST5150 Posts: 777 ✭✭✭✭
    Sep 06, 2018
    Some may poo-poo the idea of VPN tracerts/pings... but I find them a great way to demonstrate potential route based issues.

    From here in SC's coastal regions, Atlanta's Level3 exchange gets wonky from time to time. Being able to flip between Atlanta and Charlotte or other NC exit points was useful in prodding changes to NC's Cogent or TATA exchanges to clean up some lag in the past. YMMV of course, but it doesn't hurt when it shows an obvious benefit when you switch up to a different "local" exit point. Going from the NYC area to somewhere like Chicago wouldn't be a good contrast, but if going to something closer like Trenton was a dramatic change, it may put forth a better argument.
  • karlbeckmankarlbeckman Posts: 2,222 ✭✭✭✭
    Sep 06, 2018

    Keep in mind that both your 'clear' and  VPN connections should normally be routing exactly the same within the Spectrum transport network.  As far as we can see, Spectrum does not perform any load-balancing or shortest distance routing to their IXP handoff locations.  The routing will change AFTER you exit Spectrum through one of their IXP transfer points to other third-party bulk carriers such as XO or Level3 to your VPN host site and then overseas to your foreign destination gaming server. 

    Yes, Canada is considered a foreign country for data transport and all such data traffic is subject to monitoring by various federal agencies.  You can compare the routings and hop delays inside Spectrum to look for delay patterns along their segments of the route.  One thing you need to do first is fix the 3% packet loss on hop 1 between your computer and the router.  Then you have to analyze and potentially fix each glitchy hop in sequential order looking outwards from your home toward the destination. 

  • Julia_RJulia_R Posts: 4,284 Lead Mod
    Sep 07, 2018

    A thought, when using a VPN you have a different public IP address. You can reach

    the server via this but not your Spectrum IP address. 


    You might want to see if your IP is blacklisted or check with the folks who 

    manage the server you are trying to reach to see if the Spectrum IP has

    been blocked. 


    Julia R.
    Spectrum-Social Media Customer Care
    Lead Moderator-Community Forums

This discussion has been closed.