Welcome to the Spectrum Community

Get answers, share solutions, help your community.

Start here

Ping instability to the first gateway - 4 truck rolls so far

WantsStabilityWantsStability Posts: 16
in Connectivity Feb 18, 2018

4 truck rolls in and no solution. 

My first hop (I assume Charter Spectrum CMTS): 75.177.64.1

30 pings to the server from my modem: 

Minimum = 10ms, Maximum = 122ms, Average = 21ms

30 pings to the same Charter server from my friends house that is on Northstate:

Minimum = 28ms, Maximum = 46ms, Average = 29ms

 

Today is a very good day for this problem, usually the ping spikes are much worse.  These ping tests were run simultainiously and are repeatable any time, any day. The packets sent from Northstate went through at least 11 servers, from NC to GA and back again to SPECTRUM'S SERVER. This in contrast to my ping through Spectrum to their own gateway that is probably within a mile of me. Basically, a Northstate connection to a Spectrum server is FAR cleaner than a Spectrum connection to ANY server. Most days I'm getting 5-25% packet loss. Today is 1% but the ping spikes are still appearant. 

 

Full tracert from Spectrum: 

1 <1 ms * <1 ms unknown [192.168.1.1]
2 12 ms 24 ms 14 ms cpe-75-177-64-1.triad.res.rr.com [75.177.64.1]

 

Abbreviated tracert from the Northstate connection:

2 72-11-54-1-dynamic.northstate.net (72.11.54.1) 2.808 ms 2.987 ms 2.996 ms
3 97-75-159-21-static.northstate.net (97.75.159.21) 4.006 ms 3.538 ms 5.474 ms
4 * * lag-119.ear1.Atlanta2.Level3.net (4.35.198.5) 11.288 ms
5 0.ae11.pr0.dca20.tbone.rr.com (66.109.9.73) 17.327 ms 0.ae12.pr0.atl20.tbone.rr.com (66.109.9.69) 11.195 ms 35.951 ms
6 bu-ether25.vinnva0510w-bcr00.tbone.rr.com (107.14.19.161) 20.809 ms bu-ether23.atlngamq46w-bcr00.tbone.rr.com (107.14.19.10) 10.094 ms bu-ether15.vinnva0510w-bcr00.tbone.rr.com (66.109.6.166) 17.914 ms
7 bu-ether11.asbnva1611w-bcr00.tbone.rr.com (66.109.6.30) 23.429 ms 66.109.6.83 (66.109.6.83) 15.527 ms 20.858 ms
8 66.109.6.81 (66.109.6.81) 29.995 ms 25.111 ms 28.472 ms
9 24.28.255.35 (24.28.255.35) 37.640 ms 37.577 ms be1.gnbpncnd01r.southeast.rr.com (24.93.64.197) 29.941 ms

 

Comments

  • WantsStabilityWantsStability Posts: 16
    Feb 18, 2018

    This was a short ping test, for the record. I am regularly getting pings of over 200ms and drops.

  • WantsStabilityWantsStability Posts: 16
    Feb 18, 2018

    Just ran a test tethered to my phone. Sprint LTE, all pings  to the CHARTER SPECTRUM SERVER at 80-90ms.

    So there is a distinct advatage to using a cellphone's LTE instead of Spectrum, for anything where latency is a factor.

    I'm going to post my test results to some gaming forums, they care. For me, the dropped packets are making it hard to work at home and even cause VPN issues when the problem is bad.

  • karlbeckmankarlbeckman Posts: 2,205 ✭✭✭✭
    Feb 18, 2018

    The forums are peer-to-peer support by other customers, not from Spectrum employees.  You can help by NOT posting edited reports from traceroute or ping.  Every character is put into the report for a reason.  Editing out sections hints that someone is trying to hide something, rather than helping to resolve your open issue.

    What you measured was delays between Sprint's network and the Spectrum IXP gateway over Sprint's internet transport.  And your business IT department knows that VPNs require errorfree data transfer end-to-end.  Every uncorrected packet error will cause a VPN to drop, something that you can view in your modem's performance statistics.

  • MsRayeMsRaye Posts: 25,606
    Feb 18, 2018

    Need the signal level and error log printouts to go any further.

    DO NOT RESET THE MODEM, need 4 hours of run time

    You are on a wired connectioon, not wireless, correct?

     

  • WantsStabilityWantsStability Posts: 16
    Feb 18, 2018

    After going though 4 truck rolls and exchanging the modem twice, I have solved this issue.

    All my local spectrum store has ar two Arris models of modems. One has WiFi and one doesn't. i have learned they are both powered by an Intel SOC called Puma 6 which has a defect that causes erratic pings. There are real world perceptable consequences for this in activities that rely on consistent latency. It seems nobody at Spectrum support is aware of the public issue though so I had to arrive at it on my own. That's a lot of waste.

     

    I bought a Netgear CM600, and while there is still more jitter than the Northstate connect to Spectrum's server (passing hundereds of miles), my ping to the first hop now peaks at 60ms instead of over 200ms.

     

    I abbreviated the ping tests for the sake of posting. Four different Spectrum techs have agreed that the levels are fine, and I confirmed myself. So I guess the conclusion is that if gamers want to use Specrtum, they cant use the provided modems or any that are based on the Intel Puma 6 SOC.

    Thanks goes to the fine people at DSLREPORTS forum.

    https://www.dslreports.com/forum/r31122204-SB6190-Puma6-TCP-UDP-Network-Latency-Issue-Discussion

  • WantsStabilityWantsStability Posts: 16
    Feb 18, 2018

    In reply to  karlbeckman:

    What I measured was a high but consistent ping through Sprint's network while entering by LTE. This in contrast to a directly wired ethernet connection to the provided modem. Over cable modem I was getting spikes to over 200ms, often. Over sprint it stayed between 80ms and 110ms. Much more predictable. I have nothing to hide, please lookup the term "jitter" for more info.

    I had no "uncorrectables" in the log, I kept track for the past two weeks. 

     

    I still find a spike of 60ms to my first hop to be sad, dial-up was more stable. Getting off of the Intel modems has made a world of difference. I'd also add that it's possible that network upgrades are exposing the problem with these modems. I don't know, it's hard to imagine how this could be so bad and the techs have no idea what the issue is.

  • MsRayeMsRaye Posts: 25,606
    Feb 19, 2018

    Still need to see the signal page and tracert  off of either modem, but prefferably the first one..... It's not a modem issue but changing modems temporarily fixes it.  need to compare the tracert's between the 2

     written in that threadd is:"Careful about pinging at a fast rate to things like google or dslreports. It causes thier servers to be unhappy and they might even lock you out as they might think of you as a attack. You can see the issue at .1. Even then .1 might look like a attack. Use your gateway at your ISP or DNS server or CMTS. Even then dont just leave it running for a long time, they might get angry too. Test and then stop. To look up your Gateway, DNS servers use google for your OS and see how to do that. If your using a router then the router will have that info and you need to google how to find that with your router make and model. To fidn your CMTS, look up how to do a trace route on your OS. Then its the first hop out that returns a ping. Usally the 3rd one down the list, maybe 2nd. Traceroute google.com..

    But caution with high ping rates they may look like a attack. ISP have rules on that."

  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018

    " It's not a modem issue but changing modems temporarily fixes it."

    When changing back to the Puma 6 modems, the problem is instantly measurable and no waiting is needed. All 4 service techs that came concured but were not aware previously of this issue. Intel, Charter, and Comcast are aware though, and have been for at least 18 months now.  It's possible that this modem flaw exacerbated a local system problem. My ping spikes to the CMTS are not 60ms instead of being over 200ms. I checked at a friends house who is also on Charter and his ping is more stable than mine. His modem is owned by him and is not powered by the defective chipset (Puma 6).  So, I'm about to bypass my router again and check the logs on the new modem to report back here, as there may be additional issues that are compounding to cause extra high spikes of latency.

     

    In addition to the link I already provided, just Google search "Puma 6 latency". I'm afraid now that Spectrum is distributing these modems themselves, you will see more support cases caused by it. Also, I do not play World of Warcraft, but I did find in my research that they are troubled by it. WoW has an "expansion" coming soon that will draw back probably millions of players who now may be on Puma 6 modems that still have no resolution. This is going to be a mess.  

    Reference: https://us.battle.net/forums/en/wow/topic/20752441321

     

     

  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018

    I was going to post a link to a Google search of this issue on this very forum. Then I stumbled across this thread where MsRaye suggested that a person dump a Puma 6 modem. 

    https://forums.timewarnercable.com/t5/Connectivity/When-will-Charter-get-around-to-pushing-the-firmware-update/td-p/129431

     

    Why are you acting like this isn't an issue now? I think I was accused in this thread of having something to hide because I cleaned part of a tracert that had no response, but now I'm perplexed. Please explain MsRaye.

  • MsRayeMsRaye Posts: 25,606
    Feb 20, 2018

    Still need to see the signal page and tracert  off of either modem, but prefferably the first one..... It's not a modem issue but changing modems temporarily fixes it.  need to compare the tracert's between the 2

  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018
    I am preparing this with screenshots. What server would you like me to traceroute to? Have any specific servers to ping other than the CMTS?
    I have been using SSH to a server I have at a friends house locally to get a baseline. It is on Northstate residential line. That location is physically a mile away from me but the route between us seems to go all the way to Miami, we are in NC.
  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018

    I don't understand why you would expect the signal levels or traceroute to be different after a modem change, but here we go. cm600levels.jpgTM1602levels.jpgCM600tracertToNS.jpgTM1602tracertToNS.jpg

  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018

    CM600 ping to CMTS :

    cm600pingCMTS.jpg

    TM1602 ping to CMTS:

    TM1602pingCMTS.jpg

    CM600 ping to Northstate server I selected for this test:

    CM600pingToNS.jpg

    TM1602 to that same Northstate server:

    TM1602pingToNS.jpg

  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018

    For comparison, I ran a test from a server I have at a friend's house a mile away from me with a residential Northstate connection. I have SSH access to a server there. I ran a traceroute from there to me and chose a Level3 server right in the middle. I am in the northern part of North Carolina and his ping test to this server in Atlanta is more stable than my ping is to my own CMTS, even with the CM600 modem.

    "lag-119.ear1.Atlanta2.Level3.net (4.35.198.5)"

    NorthstatePingToL3.jpg

     

    Thanks for the assistance. I'm so very frustrated with Spectrum. All these tests have been run wired directly to the cable modem, for testing.

     

    There seems to be a problem with my CMTS that is excacerbated by the Puma 6 flaw. Do you agree?

  • WantsStabilityWantsStability Posts: 16
    Feb 20, 2018

    After having the TM1602 back up for a few hours, the count for "corrected" has gone up. I'm not sure if this is relevant, because DOCSIS is not my forte. I can tell you that the CM600 was running for 2 days without this issue, and the TM1602 has the issue anytime I've checked.

     

    TM1602levels-new.jpg

  • WantsStabilityWantsStability Posts: 16
    Feb 21, 2018

    Left the TM1602 connected overnight to reverse this test. Also, I was waiting to see if there were any suggestions. Before I disconnected it, there were still no "Uncorrected". 

    This is the TM1602 ping to CMTS today (it wont let me upload the screenshots today):

    Ping statistics for 75.177.64.1:
    Packets: Sent = 121, Received = 121, Lost = 0 (0% loss),
    Approximate round trip times in milli-seconds:
    Minimum = 9ms, Maximum = 171ms, Average = 22ms

     

    CM600 today:

    Ping statistics for 75.177.64.1:
    Packets: Sent = 121, Received = 121, Lost = 0 (0% loss),
    Approximate round trip times in milli-seconds:
    Minimum = 7ms, Maximum = 43ms, Average = 12ms

     

  • MsRayeMsRaye Posts: 25,606
    Feb 21, 2018

    You are having coaxial issues....

    Good luck if they really replaced everything.. Post a photo of the outside grounding block and the street tap off.

     

  • WantsStabilityWantsStability Posts: 16
    Feb 22, 2018

    I'm not totally sure they have it grounded at my house anymore. They took out the splitter. 

    I will take pics. I don't doubt you and I appreciate your help. Why would it be so much worse with the TM1602? 

  • WantsStabilityWantsStability Posts: 16
    Feb 22, 2018

    The ground was very sloppy after 4 tech visits. There is one copper wire from my power meter to a grounding post. The block that terminates it to the post was VERY loose. I cleaned and tightened it. The cable is jumped to that line with a block, both wires were oxidized so i cleaned them and reconnected. It didn't improve my issue but it does express how sloppy the 4 techs were. 

    Previous techs have literally left the cover off my tap before, I had to walk out and put it on. The most recent guy locked it up tight so I can't give you any photos. It's a green post the size of a fire hydrant.

     

    Where do I go from here? Just keep getting truck rolls? I'll just show the next technician this thread. Is there anything the tech could could be told about this situation that would convice them it's a coax problem?

    Thanks.

  • MsRayeMsRaye Posts: 25,606
    Feb 22, 2018

    Is this a new grounding block and connectors?

    If the connections were loose, corroded, that will generate all sorts of noise and create issues.

    Maybe you fixed it, Need to keep an eye on the corrected/ uncorrecteds, see if they are still increasinf=g, but I'll bet there's numerous bad connections on the street if yours was in such sad shape.

     

  • WantsStabilityWantsStability Posts: 16
    Feb 26, 2018

    Just an update. Had the 5th truck roll today. The problem seems to be getting a little worse, the latency spikes hit 210ms now with the Spectrum provided modem. I found the program "Pingplotter" and used that to make some graphs that I printed. I used the same laptop and Cat5e cable to test at 3 locations. A different Spectrum home, an AT&T residential line, and my line. This tech promised he would put it on his supervisors desk. He also said they will not fix it and that I should go to AT&T DSL. 

     

    The point where someone decides that there is no problem and closes my ticket, I need that .......person..... to call and say it directly to me. The techs are getting annoyed and that's getting me annoyed. I'm the one paying for this junk, they are getting paid to tell me it won't get fixed.

  • MsRayeMsRaye Posts: 25,606
    Feb 26, 2018

    Does the electric meter have a ground wire coming off it as well as a main breaker there?

    Are all the houses wired like that? If so, there may be a lot ofnoise getting into the system

    from multiple homes.

     The first address beyond your modem if bad, affects everything else.

    Copy and paste the signal level page, let's see if anything changed. Don't reset it

     

This discussion has been closed.