NextDNS cutting out, getting errors on servers
Since I have set up NextDNS on my Asus router maybe once a day for 10-30 minutes the domain resolution is really slow, it's at the point of not working and most of the time I'm getting errors from websites but if I refresh them a lot of times they pop im with a few broken elements.
I have checked ping.nextdns.io and I'm getting errors on most of the servers
■ zepto-prg 37 ms (anycast1)
anexia-bts error
enahost-bts error
anexia-vie error
exoscale-vie error
edis-bud error
anexia-bud error
edis-zag error
anexia-prg error
edis-lju error
anycast.dns2.nextdns.io error (anycast2)
dns1.nextdns.io error (ultralow1)
dns2.nextdns.io error (ultralow2)
but after a while everything goes back to normal
On my PC there is a different NextDNS configuration but this is happening on my phone too at the same time so that shouldn't be the problem.
My guess is that it's a router miss-configuration but I haven't figured out how why where...
My router dns config:
I have totally ran out of ideas what might cause, I really appreciate any help
-
- Pro Subscriber
Same here (also using DoT)... Slow, freezing and getting errors like you. My AnyCast/UltraLow servers are changing all over Europe. Usually I'm connected to Copenhagen and Stockholm, but today I'm seeing Oslo, Frankfurt, Amsterdam and Prague.
Maybe some maintenance work going on by NextDNS team?!?!
-
Jörgen no maintenance in progress. Please send a https://nextdns.io/diag so we can investigate. Ultralow should be very stable. What is your system resolver on the machine you are experiencing this?
-
- Pro Subscriber
NextDNS Servers appear stable and back to normal now again with connections to Copenhagen and Stockholm. Maybe a local outbreak with my ISP?!
I'm still having big issues with the CLI every 30 mins like I wrote in another thread, so have given up using that for now https://help.nextdns.io/t/g9hd8jf/log-errors-doh-resolve-context-deadline-exceeded. Any clue what that is about?!
Now I have reverted back to the built-in DoT in ASUS Merlin (using Stubby 0.4.0). My issue now is seeing a lot of DNS-retransmissions on almost every DNS-query. Are the servers too slow in responding or why is this happening? Have tried changing timeouts, retries and Round Robin setup in Stubby, but no change to the better so back to all default config now.
Every yellow dot (Warning) below are retransmissions. The domain resolves after 3-4 retransmissions shown in WireShark, but something don't feel right with this overhead.
- Pro Subscriber
-
Jörgen both issues are probably related and it is most likely a routing issue. A diag would be helpful.
-
- Pro Subscriber
NextDNS Sending my diag in a PM right now. Thanks for any help!!
- Pro Subscriber
-
Jörgen looks very good. Please run another diag while it fails again if it does.
-
- Pro Subscriber
NextDNS Thanks for checking!! It does look good "on paper", but still lots of DNS-retransmissions and a CLI bugging out every 30 min for me. Strange...
- Pro Subscriber
-
Jörgen looking at your screenshot, why do you have retransmits for the same domain with a few milliseconds difference?
-
- Pro Subscriber
NextDNS That puzzles me too... Using Windows 11 so wondering if something is new/changed with it?! There should be some timeout before doing retransmission, but it seems to send bursts of DNS-queries. Same retransmissions is seen either if I use Ethernet or WiFi.
- Pro Subscriber
-
Jörgen windows is known for sending duplicate DNS packets on all available interfaces, so it is probably what you are seeing.
- Pro Subscriber
