C
Coder.com•2mo ago
finomen_

Hot to configure STUN on kubernetes?

I've deployed coder to my kubernetes cluster using helm chart but DERP is shown as unhealthy(ESTUN01 No STUN servers are available.). What I'm missing in configuration? It is strange to have IPv6 red too because cluster has ipv6 netcheck: netcheck.runProbe: got STUN response for 1004stun0 from 46.4.36.142:39538 (9d0a6427fde4dd8c8068781b) in 5.429369ms netcheck: netcheck.runProbe: got STUN response for 1004stun0 from 46.4.36.142:39538 (477fdf472350cf0d1ff18396) in 5.299449ms netcheck: netcheck.runProbe: got STUN response for 1004stun0 from 46.4.36.142:39538 (3a797439a786b219bd4e7982) in 5.238018ms netcheck: netcheck.runProbe: got STUN response for 1000stun0 from 46.4.36.142:39538 (0bd0cd290cdc6027c056a8b1) in 5.293097ms netcheck: netcheck.runProbe: got STUN response for 1001stun0 from 46.4.36.142:39538 (3327e88b54569309d559229b) in 81.721878ms netcheck: [v1] measuring ICMP latency of coder (999): no address for node 999b netcheck: netcheck.runProbe: got STUN response for 1001stun0 from 46.4.36.142:39538 (85baad7375e730a1f6ffbb96) in 81.742204ms netcheck: netcheck.runProbe: got STUN response for 1001stun0 from 46.4.36.142:39538 (fbd49d855a8cf6de0a8c069f) in 81.765689ms netcheck: [v1] report: udp=true v6=false v6os=true mapvarydest=false hair=false portmap= v4a=46.4.36.142:39538 derp=999 derpdist=999v4:1ms
No description
5 Replies
Codercord
Codercord•2mo ago
<#1340645692405710990>
Category
Help needed
Product
Coder (v2)
Platform
Linux
Logs
Please post any relevant logs/error messages.
Phorcys
Phorcys•2mo ago
hey @finomen_, apologies for the lack of reply, have you figured this out? for the IPv6 side of things, does the Coder pod have an IPv6?
finomen_
finomen_OP•2mo ago
No, but except for warnings coder looks working Pod has both v4 and v6 
  hostIP: 172.16.0.1
  hostIPs:
  - ip: 172.16.0.1
  - ip: fd12:3456:789a:1::1
  phase: Running
  podIP: 10.244.3.240
  podIPs:
  - ip: 10.244.3.240
  - ip: 2001:db8:42:2::b5d7
  hostIP: 172.16.0.1
  hostIPs:
  - ip: 172.16.0.1
  - ip: fd12:3456:789a:1::1
  phase: Running
  podIP: 10.244.3.240
  podIPs:
  - ip: 10.244.3.240
  - ip: 2001:db8:42:2::b5d7
Phorcys
Phorcys•2mo ago
would you be able to send the full logs for your coder instance? to me, it seems fine
Scott
Scott•2mo ago
I'm not sure if it will help, but in order for me to properly use Coder in k8s, I had to make sure my ingress was "upgrading" the connection, much like is also necessary for websocket connections. I have this in my ingress mapping YAML for the Coder route: 
  allow_upgrade:
    - derp
    - websocket
  allow_upgrade:
    - derp
    - websocket
This is using emissary-ingress, but maybe your ingress controller/ gateway has something similar.

Did you find this page helpful?