agg62.srspny0101h.northeast.rr.com 24.58.201.65 switch latency - Time Warner Internet

#1
This question is about "agg62.srspny0101h.northeast.rr.com 24.58.201.65 switch latency", with Time Warner Cable internet and apps. Being that this is my first post ever here, I'm a little worried about how it will be received and I don't want to offend anyone at Spectrum tech / admin staff. I'm not trying to be critical, I just know what I know after over 20 years as a tech myself. If I were to call Spectrum about this (based on over decades of using Time Warner, now Spectrum, and my experience with the process) - if I were to call the support phone line I'll first be connected to someone who doesn't really understand DNS and waste a lot of time with the usual (unplug all devices, reboot routers / modems / devices, clear cache, only connect one device via direct ethernet to modem, and so on... ad-nauseam) until ultimately they'll file a ticket with an actual admin, and then they'll want to send a truck to my house over something that has nothing to do with my house and everything to do with Spectrum's routing of my connection. Lots of time wasted on something that is easily / obviously determined, so I'm rolling the dice and posting here first.



In recent days (especially today / yesterday) one of the switches / hops they've got me routed through is experiencing high latency and killing my load times for pages while surfing.



Bottleneck - agg62.srspny0101h.northeast.rr.com 24.58.201.65



That's the third hop in all of my tracert resuslts and the ping times are bad, and it's noticeable. And as anyone knows - your connection speed is only as good as your worst hop / ping on a tracert.



Perfect example, here's a tracert just now from my residence to google.com :



Tracing route to google.com [172.217.6.206]
over a maximum of 30 hops:

1. 1 ms 2 ms 1 ms 192.168.0.1
2. 41 ms 12 ms 11 ms 142.254.216.177 (indicates it's a Charter IP via Dallas Texas)
3. 409 ms 22 ms 189 ms agg62.srspny0101h.northeast.rr.com [24.58.201.65]
4. 14 ms 12 ms 13 ms 24.58.34.216
5. 15 ms 21 ms 13 ms be26.albynyyf01r.northeast.rr.com [24.58.32.56]
6. 24 ms 21 ms 23 ms bu-ether16.nycmny837aw-bcr00.tbone.rr.com [66.109.6.74]
7. 20 ms 20 ms 20 ms 0.ae4.pr0.nyc20.tbone.rr.com [66.109.1.35]
8. 20 ms 28 ms 20 ms ix-ae-6-0.tcore1.n75-new-york.as6453.net [66.110.96.53]
9. 21 ms 20 ms 23 ms 72.14.195.232
10. * * * Request timed out.
11. 41 ms 23 ms 24 ms 108.170.227.208
12. 22 ms 22 ms 23 ms 108.170.248.52
13. 44 ms 25 ms 20 ms 108.170.227.148
14. 21 ms 22 ms 22 ms 216.239.58.120
15. 20 ms 20 ms 21 ms 108.170.248.65
16. 22 ms 20 ms 21 ms 108.170.237.207
17. 31 ms 22 ms 20 ms lga25s54-in-f14.1e100.net [172.217.6.206]



As you can see from the tracert:



First hop is exactly as expected - router responds 1ms



Second hop isn't too good - 40ms - and seems a little odd that Spectrum is taking me directly to 142.254.216.177 , a Charter switch in Dallas Texas first before routing me through a Saratoga Springs NY switch. (I live in / use Spectrum in the Saratoga Springs / Upstate NY area).



Third hop is consistently bad - 409ms worst ping in this case - as Spectrum routes my connect back up from their Texas switch back up to my local switch in Upstate NY at agg62.srspny0101h.northeast.rr.com [24.58.201.65]



After getting past that bad hop at agg62.srspny0101h.northeast.rr.com [24.58.201.65] pings return to normal / expected times (with the exception of a timeout between two Google switches, which is to be expected with certain secured switches) and as you can see my normal average pings / hops are around 20ms.



It's clear and obvious from the tracerts I've done, in both directions (tracing to servers, and doing MTR's from servers I control in NYC back to my local Saratoga IP) , that the consistent latency is happening at Hop 2 and Hop 3 - both Charter / Spectrum hops, that should be quick right out of the gate.



I could provide tracerts and MTR's between my home Spectrum connection to whatever site you want, and we're going to see the pattern - the second two hops (Charter / Spectrum) are bringing me off to a bad start.



As anyone reading this probably knows - it doesn't matter than I'm getting 20ms pings on later hops - if you've got a 20ms average on hop 10 but you're getting 150ms - 400ms on hope 3 from your ISP, it just means that the other hops are switching well and your ISP hops are the ones causing your latency. YOUR BEST CONNECTION IS ONLY AS GOOD AS YOUR WORST HOP.



I'm posting this here in hopes to make Spectrum techs aware and to avoid having to play phone games / speculation games in order to get someone to finally say "oh, yeah, definitely seeing some packet loss right out of the gate".



I'll certainly call in and go through the motions if I must, but with the hours I work it's really hard to justify the waste of time. The problem spot lights up like a firefly in the night sky, and any SysAdmin can see it.



Saratoga Springs, NY

100Mbps down / 10Mbps up plan (seriously though, 10Mbps up? COME ON!)

ARRIS MTA model TM602G

ARRIS CM820A DOCSIS 3

TP-Link C2300 Router

Windows 10 Pro, 32GB RAM, Core i7 8th gen, 512GB SSD

Actual PEDESTAL with 1-inch HARD LINE right near my porch with tap 200 feet away down at pole (long story short - Time Warner said my property / house was 100% "cable ready" when I moved here 2004, but it wasn't and I had to dig my own 10"x4" trench for them to run the hardline from the new tap to my house the week we moved in... they laid the cable and left the trench wide-open for me to have to bury myself). Signal was so hot that they stuck a 100 foot coax cable between the pedestal an indoor connection to act as resistor back in 2004 / 2005).

My systems are well maintained and kept updated (aside from Time Warner / Spectrum's ARRIS MTA model TM602G which they say I cannot update myself).



Maybe this post will go nowhere, or maybe there's a tech with enough knowledge, integrity, and ambition to actually look into the source of the problem instead of the usual "pin it on the end user".



I'd be happy to set up a custom linux server at a fully fledged data center here in New York and provide SSH shell access for an actual admin to run MTR's and test it both directions, whether via IP or actual Domain DNS. Choose your flavor - I just want to deal directly with an actual SysAdmin who knows what a tracert / cidr / class c / nameserver / etc... means. The last thing I want to hear is "we're not seeing a problem" from someone who wouldn't know a hostname from from an actual domain.



I honestly mean no offense at all. I'm just frustrated with the system of getting to an actual tech who knows that these things can be diagnosed remotely and that the problem is clearly not in my home or with my devices. Guess we'll see...



Many thanks for anyone who took the time to read,

Dargo

agg62.srspny0101h.northeast.rr.com 24.58.201.65 switch latency?
 
Top