1

Latency Issue

Hello Team, I came here to comment on a latency problem that is happening to me on the edgeuno-for server that is located in the same city that I live in. 

The latency for this server rose from 1 ms to 45-50 ms from nothing a month ago and continues to this day like this. At first I thought it was a temporary problem, some change or update that was being made that would be solved in 1 to 2 weeks, so I decided to wait, but it has already been 1 month with high latency, so I decided to come here to notify you about it.

The link to the diagnosis: https://nextdns.io/diag/1f2a5350-0619-11ed-b42d-811dcfce158f

12 replies

null
    • NextDNs
    • 2 yrs ago
    • Reported - view

    From the various traceroutes in the diag, it looks like an issue on your ISP's network.

      • Shadow_Colossus
      • 2 yrs ago
      • Reported - view

      NextDNS So, in this situation, what am I supposed to do?

      • NextDNs
      • 2 yrs ago
      • Reported - view

      Shadow Colossus only your ISP can fix this.

      • Shadow_Colossus
      • 2 yrs ago
      • Reported - view

      NextDNS 

      ... So I'm basically screwed, I live in a country where internet companies are not required to offer services like fixing routes. 

      Which means it's like playing roulette to do any ISP in my country to fix something like that, and even if they do, they make it hell to do the job (it means more than 2 hours on mobile or phone just trying to convince them to do such a thing or explain the problem)

      Isn't there really something you can do to fix this or put in some kind of patch that can slow the latency? I really don't want to go through hell trying to convince my ISP to fix this kind of thing.

      Also, thanks for the quick answers.

      • NextDNs
      • 2 yrs ago
      • Reported - view

      Shadow Colossus we can't act on your ISP's network. You most likely have latency issue with many other services.

      • Shadow_Colossus
      • 2 yrs ago
      • Reported - view

      NextDNS 

      Worse than not, this latency problem is only happening with NextDNS because before I came here, I decided to switch to Cloudflare and CONTROLD to test how the latency was in them (since they are the only two other services besides NextDNS that offer servers in Brazil), my ping in Control is from 16 ms to 26 ms (their server is basically in the middle of the country where I live) and in CloudFlare it was 10 ms (their server is also basically in the middle of the country).

      So, from what I understand, this issue is a network issue for my ISP that is only affecting NextDNS. Which is weird, come to think of it. 

      So basically the solution is to call my ISP or switch to a different DNS service that is not being affected by this.

      Thank you for your time and thank you for answering my questions. You have a nice day.

      • NextDNs
      • 2 yrs ago
      • Reported - view

      Shadow Colossus please show a traceroute to the two other services.

      • Shadow_Colossus
      • 2 yrs ago
      • Reported - view

      NextDNS 

      The first image is from here: https://controld.com/network

      The second is the Traceroutes for Cloudflare and ControlD servers.

      I hope this information is useful.

      • NextDNs
      • 2 yrs ago
      • Reported - view

      Shadow Colossus the traceroutes show the same latency issue from the same hops. This confirms our theory. Changing resolver service will have no effect.

      • Shadow_Colossus
      • 2 yrs ago
      • Reported - view

      NextDNS 

      Wait, so why on https://controld.com/network/ - do you keep saying that if I use their service, my latency will be between 16 ms to 26 ms? Like, every time I hover over Sao Paulo, the latency keeps changing, like on the NextDNS homepage that is down there on ULTRA-LOW LATENCY NETWORK. But unlike the homepage of NextDNS that keeps saying my real ping which is between 45 to 50 MS, in Controld it keeps saying that my ping will be between 16 to 26 ms and Traceroutes shows that the ping will be the same no matter what are the three services chosen...

      Now I'm confused, can you tell me how these Network pages work and the difference between them for a real Traceroutes test? Why I got confused now on how latency is appearing differently in both types of latency visualization in ControlD.

      I would appreciate it if you could give an answer to that. Also thank you for letting me know that I was looking at things the wrong way and that if I changed service it would be the same. I'll stick with NextDNS then and try to call my ISP...

      May God give me patience and a very free day to make that call, I'm going to need it.

      • NextDNs
      • 2 yrs ago
      • Reported - view

      Shadow Colossus our network page computes the RTT based on the RTT as seen by the server’s network stack that gets the HTTP request from your browser (the one used to get this value). This can be considered more accurate than a traceroute or a ping because it is measuring an actual TCP connection to the DNS server instead of relying on ICMP. Note that a difference of 20ms is considered to be noise. You won’t notice the difference.

      • Shadow_Colossus
      • 2 yrs ago
      • Reported - view

      juliank From the response of the NextDNS team about my question, it seems the test they do is different from the test ControlD does. From what it seems, NextDNS 'ping test is much more accurate than ControlD's, so I understood their answer. Still, thank you for your attempt to answer my question.

       

      NextDNS I understand now, thank you for answering my question. Thank you also for your patience in answering my questions.

Content aside

  • 1 Likes
  • 2 yrs agoLast active
  • 12Replies
  • 210Views
  • 4 Following