Open eoli3n opened 4 years ago
Strangly ( :) ), fixing my routing tables to reach pods subnets from workers did the trick. See https://github.com/kelseyhightower/kubernetes-the-hard-way/issues/539#issuecomment-716106281
I can now resolve but it is long, between 10 to 30 seconds and sometimes it fails to resolve, i will figure this out
➜ time k exec -ti busybox -- nslookup kubernetes
Server: 10.32.0.10
Address 1: 10.32.0.10 kube-dns.kube-system.svc.cluster.local
Name: kubernetes
Address 1: 10.32.0.1
________________________________________________________
Executed in 20,82 secs fish external
usr time 123,06 millis 1124,00 micros 121,93 millis
sys time 35,19 millis 374,00 micros 34,81 millis
I solved long delay resolution with this -> https://github.com/kubernetes/kubernetes/issues/21613#issuecomment-343190401
It should be added ? Does ubuntu server from GKE load that kernel module defaultly ? https://github.com/kelseyhightower/kubernetes-the-hard-way/issues/561
Hi,
Internal DNS resolution is not working, and as forward is not set, external doesnt work too with a warning.
When i set
/etc/resolv.conf
in Corefilelog
plugin in Corefile -resolvConf: "/run/systemd/resolve/resolv.conf"
in kubelet config file It seems that service can't be reach, nothing appears in logs.Any help would be appreciated.