0

lightnode-tpe PoP is down

Hi NextDNS team,

The lightnode-tpe PoP is currently returning errors, while all other nearby PoPs (including zepto-tpe, zepto-hkg, anexia-hkg, etc.) are responding normally.

Latency test results from my location (Taiwan):

 

Diagnostic: https://nextdns.io/diag/b6623d30-589c-11f1-9f78-ff2e99e20ba3

Could you please take a look? Thanks!

1 reply

null
    • Tim.17
    • 3 days ago
    • Reported - view

    Hey, quick update — lightnode-tpe is back online today thanks for the fix!

    But while testing I bumped into something that looks like a separate bug: EDNS Client-Subnet (ECS) doesn't seem to work on lightnode-tpe. Same profile works fine on zepto-tpe, so it doesn't look like a config issue on my side.

    Two back-to-back DoH3 queries to text.nextdns.io CHAOS , about a minute apart, full dig output below.

    Query 1 → routed to lightnode-tpe-1 (ECS = not sent):

    ; <<>> DiG 9.16.28 <<>> text.nextdns.io CHAOS
    ;; global options: +cmd
    ;; Got answer:
    ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 59824
    ;; flags: qr rd ra ad; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 10
    ;; OPT PSEUDOSECTION:
    ; EDNS: version: 0, flags:; udp: 1232
    ;; QUESTION SECTION:
    ;text.nextdns.io.               CH      A
    ;; AUTHORITY SECTION:
    nextdns.io.             1761    IN      SOA     dawn.ns.cloudflare.com. dns.cloudflare.com. 2405167425 10000 2400 604800 1800
    ;; ADDITIONAL SECTION:
    client-name.nextdns.io. 0       CH      TXT     "unknown-doh3"
    device-name.nextdns.io. 0       CH      TXT     "<redacted>"
    device-id.nextdns.io.   0       CH      TXT     "<redacted>"
    device-model.nextdns.io. 0      CH      TXT     "ASUS"
    proto.nextdns.io.       0       CH      TXT     "DOH3"
    server.nextdns.io.      0       CH      TXT     "lightnode-tpe-1"
    profile.nextdns.io.     0       CH      TXT     "<redacted>"
    client.nextdns.io.      0       CH      TXT     "<my-public-ip>"
    smart-ecs.nextdns.io.   0       CH      TXT     "not sent"
    ;; Query time: 113 msec
    ;; SERVER: 192.168.1.1#53(192.168.1.1)
    ;; WHEN: Wed May 27 08:53:53
    ;; MSG SIZE  rcvd: 392

    Query 2 → routed to zepto-tpe-1 (ECS works as expected):

    ; <<>> DiG 9.16.28 <<>> text.nextdns.io CHAOS
    ;; global options: +cmd
    ;; Got answer:
    ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 35110
    ;; flags: qr rd ra ad; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 10
    ;; OPT PSEUDOSECTION:
    ; EDNS: version: 0, flags:; udp: 1232
    ;; QUESTION SECTION:
    ;text.nextdns.io.               CH      A
    ;; AUTHORITY SECTION:
    nextdns.io.             1800    IN      SOA     dawn.ns.cloudflare.com. dns.cloudflare.com. 2405167425 10000 2400 604800 1800
    ;; ADDITIONAL SECTION:
    proto.nextdns.io.       0       CH      TXT     "DOH3"
    server.nextdns.io.      0       CH      TXT     "zepto-tpe-1"
    device-name.nextdns.io. 0       CH      TXT     "<redacted>"
    profile.nextdns.io.     0       CH      TXT     "<redacted>"
    client.nextdns.io.      0       CH      TXT     "<my-public-ip>"
    client-name.nextdns.io. 0       CH      TXT     "unknown-doh3"
    device-id.nextdns.io.   0       CH      TXT     "<redacted>"
    device-model.nextdns.io. 0      CH      TXT     "ASUS"
    smart-ecs.nextdns.io.   0       CH      TXT     "1.160.0.0/24"
    ;; Query time: 8 msec
    ;; SERVER: 192.168.1.1#53(192.168.1.1)
    ;; WHEN: Wed May 27 08:55:13
    ;; MSG SIZE  rcvd: 392

    Same profile, same client, same protocol — only difference is which PoP picked up the query. Smart ECS just isn't kicking in on lightnode-tpe for some reason. Could you take a look when you get a chance? 🙏

Content aside

  • 2 days agoLast active
  • 1Replies
  • 20Views
  • 1 Following