0

Horrible DNS latencies since yesterday - family is not happy.

Hi Team:  Long time NextDNS user with ASUS Merlin router.   NO changes on router and I do a manual setup using stubby.yml which has not changed.

Starting yesterday, DNS latencies went horrible and barely resolve.  Normally, boom and all in the low 20ms.  Family is screaming about how horrible DNS is.  What's going on? 

Oh I already tried to DL the "diag" script and 2+ antiviral / malware programs wiped it out immediately without even opening it.   So I doubt that's going to get passed those scanners. 

Also, I've already rebooted the router and checked my stubby.yml file for any changes.  

ping.nextdns.io yields (multiple times) I might get 1 to resolve to 25-50 ms if lucky.

 hydron-clt                error

  tier-clt                  error

  anexia-mnz                error

  zepto-xrs                 error

  zepto-iad                 error

  wlvrz-was                 error

  teraswitch-pit            error

  router-pit                error

  anexia-atl                error

  vultr-atl                 error

anycast.dns1.nextdns.io error (anycast1)

anycast.dns2.nextdns.io error (anycast2)

dns1.nextdns.io error (ultralow1)

dns2.nextdns.io error (ultralow2)

58 replies

null
    • olivier
    • 3 yrs ago
    • Reported - view

    Can you run the diag from a non windows machine?

    You may try ping with nextdns disabled to understand better what is going on.

    Also please try a traceroute to 45.90.28.0 and 45.90.30.0.

    • G_Mobley
    • 3 yrs ago
    • Reported - view

    Things have substantially improved this AM without explanation or root-cause.  My stubby.yml and router settings were verified right and unchanged for 6+ months.   Family has stopped complaining for now.  Whatever you guys did, thanks! 

    I'm sorry - I have no linux systems at home. Maybe next time my wife will let me near her Mac but it too is running anti-viral/anti-malware so I'll not be surprised if it's flagged there too.  If there was a way to run it on the router (ASUS / Merlin) then that I can do. 

    vultr-atl        24 ms  (anycast1, ultralow1)

      zepto-iad        26 ms

      tier-clt         27 ms

      zepto-xrs        29 ms

      hydron-clt       30 ms

      teraswitch-pit   31 ms

      anexia-atl       32 ms

      anexia-mnz       33 ms

      router-pit       34 ms

      anexia-rio      149 ms  (anycast2, ultralow2)

      wlvrz-was        error

    • olivier
    • 3 yrs ago
    • Reported - view

    We did nothing but you're welcome :) It has probably something to do with your ISP or something.

    • Chris_Dunn
    • 3 yrs ago
    • Reported - view

    We've been having issues all morning. The only way I was able to resolve it was to remove NextDNS and move back to Quad9. In my case it's high packet loss to 45.90.28.40 which is our primary DNS IP per our settings page.

      • Chris_Dunn
      • 3 yrs ago
      • Reported - view

      Chris Dunn Primary IP for us, in this case, is 45.90.28.40. When 45.90.30.40 was experiencing the high packet loss, our secondary IP, 45.90.30.40 wasn't impacted.

      • olivier
      • 3 yrs ago
      • Reported - view

      Chris Dunn is it still happening? What is your ISP?

      • Chris_Dunn
      • 3 yrs ago
      • Reported - view

      Olivier Poitrey Haven't tried again as the kids needed access to their school work and I needed access to work. I'm connected to a local WISP but I didn't see any interruption to the 45.90.30.40 IP. If it helps I'm using YogaDNS on most of the PCs. One of the kids PC doesn't have that loaded and he never had an issue. All the PC's running YogaDNS with the NextDNS settings loaded were impacted. 

      • olivier
      • 3 yrs ago
      • Reported - view

      Chris Dunn In yoga, did you try with and without the "Ultra Low Latency" option? Does it change something?

      • Chris_Dunn
      • 3 yrs ago
      • Reported - view

      Olivier Poitrey That option is not enabled on any of the PCs here. Should it be enabled? I've re-enabled it on my PC and so far it's been stable and not seeing the packet loss to that default DNS IP.

      • olivier
      • 3 yrs ago
      • Reported - view

      Chris Dunn yes, it’s still very new but should give best results.

    • G_Mobley
    • 3 yrs ago
    • Reported - view

    Hi Olivier and team: 

    I had remove NextDNS yesterday afternoon from my ASUS router and replace with Quad9/Cloudflare b/c NextDNS died again with no DNS resolution. I left NextDNS off all night with no issues and am currently still using the others since I cannot disrupt the WAH.   

    This AM, here's the tracert -  still not good with those timeouts - usually saw many more yesterday.  

    Thanks for any recommendations.. I suspect this is all ISP routing issues but if there's something you guys need to kick, by all means please. 

    >tracert 45.90.28.114

    Tracing route to dns1.nextdns.io [45.90.28.114]
    over a maximum of 30 hops:

    1 2 ms 1 ms 1 ms 3622-10007-AC1900-FA38.xxx [192.168.100.7]
      2     2 ms     2 ms     3 ms  192.168.222.7
      3     2 ms     1 ms     1 ms  192.168.111.7
    4 15 ms 13 ms 23 ms 065-190-080-001.inf.spectrum.com [65.190.80.1]
      5    14 ms    33 ms    19 ms  174.111.102.224
    6 16 ms 11 ms 20 ms cpe-024-025-062-048.ec.res.rr.com [24.25.62.48]
    7 20 ms 15 ms 14 ms be31.drhmncev01r.southeast.rr.com [24.93.64.184]
      8    23 ms    15 ms    29 ms  66.109.10.176
    9 28 ms 22 ms 31 ms bu-ether12.vinnva0510w-bcr00.tbone.rr.com [66.109.6.31]
    10 23 ms 21 ms 32 ms ae-11.edge5.WashintonDC12.Level3.net [4.68.37.213]
     11     *        *        *     Request timed out.
    12 30 ms 35 ms 40 ms CHOOPA-LLC.ear3.NewYork1.Level3.net [4.15.213.214]
     13     *        *        *     Request timed out.
     14     *        *        *     Request timed out.
     15     *        *        *     Request timed out.
    16 26 ms 26 ms 38 ms dns1.nextdns.io [45.90.28.114]

    Trace complete.

    >tracert 45.90.30.114

    Tracing route to dns2.nextdns.io [45.90.30.114]
    over a maximum of 30 hops:

    1 1 ms 1 ms 2 ms 3622-10007-AC1900-FA38.xxx [192.168.100.7]
      2     1 ms     2 ms     1 ms  192.168.222.7
      3     2 ms     1 ms     1 ms  192.168.111.7
    4 17 ms 15 ms 32 ms 065-190-080-001.inf.spectrum.com [65.190.80.1]
      5    13 ms    14 ms    14 ms  174.111.102.224
    6 27 ms 14 ms 14 ms cpe-024-025-062-048.ec.res.rr.com [24.25.62.48]
    7 22 ms 25 ms 12 ms be31.drhmncev01r.southeast.rr.com [24.93.64.184]
      8    26 ms    31 ms    23 ms  66.109.10.176
    9 21 ms 31 ms 29 ms 209-18-43-59.dfw10.tbone.rr.com [209.18.43.59]
    10 23 ms 23 ms 21 ms ash-b2-link.ip.twelve99.net [62.115.188.210]
    11 19 ms 24 ms 17 ms voxility-svc071266-ic357612.ip.twelve99-cust.net [195.12.254.137]
    12 22 ms 22 ms 35 ms ash-eqx-01c.voxility.net [5.254.81.22]
     13     *        *        *     Request timed out.
    14 26 ms 25 ms 20 ms c0010.mc2.iad01.us.misaka.io [45.11.106.10]
    15 27 ms 28 ms 21 ms dns2.nextdns.io [45.90.30.114]

    Trace complete.

    The ping.nextdns.io shows just now (not using NextDNS for DNS)

    zepto-iad        22 ms  (anycast2)

      vultr-atl        23 ms  (ultralow2)

      anexia-atl       23 ms  (ultralow1)

      zepto-xrs        27 ms

      vultr-ewr        28 ms  (anycast1)

      tier-clt         34 ms

      anexia-mnz       36 ms

      teraswitch-pit   36 ms

      router-pit       37 ms

      hydron-clt       39 ms

      smarthost-jax    45 ms

    Thanks!  Stay safe, stay alive! 

     

    It would really be super if we could setup the configs so that if NextDNS was not responding, our setups would automatically fallover to one of the other DNS providers like: QUAD9/CloudFlare/Google, ... and then send us an alert maybe.  It would have to be an "opt-in" setting b/c I'm sure there are people who do not want to use any of those options under any circumstances.    We can configure the listing in the router but then that doesn't play nice with the nextDNS setups as far as I know/read a year or so back. 

    THANKS!  

      • olivier
      • 3 yrs ago
      • Reported - view

      G Mobley your traceroute and ping look good. Are you sure the issue isn’t elsewhere? How did you setup nextdns?

    • Hans_Geiblinger
    • 3 yrs ago
    • Reported - view

    I am seeing similar behavior this morning as well too. Olivier Poitrey  here is a screenshot from just now. I have since moved back to cloudflare for the moment as it was unbearable.

      • olivier
      • 3 yrs ago
      • Reported - view

      Hans Geiblinger please provide a http://nextdns.io/diag

      • Hans_Geiblinger
      • 3 yrs ago
      • Reported - view

      Olivier Poitrey  I just switched back this morning, and it seems to be working okay at the moment. If the issue comes back again, I will start my own thread and tag you.

      • Hans_Geiblinger
      • 3 yrs ago
      • Reported - view

      Olivier Poitrey I opened a new issue on this: https://help.nextdns.io/t/60hzvs5

    • G_Mobley
    • 3 yrs ago
    • Reported - view

    Hi Olivier:  My setup has been stable until a few weeks ago.  I have a manual integration (no agent) on my ASUS router running Merlin 384.19.  I've been working with NextDNS and a paying customer since you launched by helping many Merlin users in the SNB forums.

    Nothing changed in my config. NextDNS totally stopped working at 16:18 on 2/25 - dead stop from DNS logs.   I rebooted my Main ASUS router - not DNS resolved.  My wife was standing in my office door so I quickly replaced the NextDNS config with QUAD9/Cloudflare and reset the 1 stubby file - everyone was back happy. 

    This is twice in ~ 2 weeks (see earlier report) NextDNS cold-stopped working using a config that's been stable for 6-12 months - untouched.  

    I've not tried switching back to NextDNS this AM as the wife is already up. Not much tinkering I can do until "off-peak" hours now.      I cannot provide that diag b/c the anti-viral and anti-malware software will just remove it from the setups.  I did provide the ping and other PD tools. THANKS!  

    • saffron_spring
    • 3 yrs ago
    • Reported - view

    Also similar issues taking ages to resolve sites yesterday, 10+ seconds.

    • Chris_Dunn
    • 3 yrs ago
    • Reported - view

    I keep having off-and-on issues with resolution. It's almost to the point I need to find another service.  In my YogaDNS I'm seeing blocks of time where I get "Error in getaddrinfo: No such host is known." or "request timeout"

      • John_DeCarlo
      • 3 yrs ago
      • Reported - view
      G Mobley said:
      that's

       Chris Dunn  

      Chris I had a similar problem. I fixed by creating a new dns rule and adding these domain's to the rule. 

      dns.nextdns.io

      steering.nextdns.io

      anycast.dns.nextdns.io

      dns1.nextdns.io

      dns2.nextdns.io

      Also add a new DOH dns resolver.

      https://1dot1dot1dot1.cloudflare-dns.com/dns-query

      The dns server address is 1.1.1.1

       

      This configuration should fix your time outs.

       

      Also for your nextdns cli you must add a dns forwarder

       

      sudo nextdns config set     -forwarder mycompany2.com=https://doh.mycompany.com/dns-query#1.2.3.4
      sudo nextdns restart
      • Chris_Dunn
      • 3 yrs ago
      • Reported - view

      John DeCarlo I'm giving this a try, thank you. 

      • John_DeCarlo
      • 3 yrs ago
      • Reported - view

      Chris Dunn 

      Let be know how it works out for you.

    • G_Mobley
    • 3 yrs ago
    • Reported - view

    John DeCarlo  I'm trying to figure out WHERE you created "a new DNS rule..." Is that on the router, in NextDNS, both, neither?    I'm going to try going back to NextDNS.... Thanks.  

      • John_DeCarlo
      • 3 yrs ago
      • Reported - view

      G Mobley  What kind of problems are you having.   

Content aside

  • 3 yrs agoLast active
  • 58Replies
  • 1754Views
  • 9 Following