rancher: Kubernetes 1.6 showing DNS errors on deployments

Rancher versions: rancher/server: 1.6.3 rancher/agent: rancher/agent:v1.2.5

Infrastructure Stack versions: healthcheck: 0.3.0 ipsec: 0.1.1 network-services: 0.2.2 scheduler: rancher/k8s:v1.6.6-rancher1-4 kubernetes (if applicable): rancher/k8s:v1.6.6-rancher1-4

Docker version: (docker version,docker info preferred)

Client:
 Version:      1.12.6
 API version:  1.24
 Go version:   go1.6.4
 Git commit:   78d1802
 Built:        Tue Jan 10 20:38:45 2017
 OS/Arch:      linux/amd64

Server:
 Version:      1.12.6
 API version:  1.24
 Go version:   go1.6.4
 Git commit:   78d1802
 Built:        Tue Jan 10 20:38:45 2017
 OS/Arch:      linux/amd64

Operating system and kernel: (cat /etc/os-release, uname -r preferred)

ubuntu@ip-172-31-11-231:~$ cat /etc/os-release
NAME="Ubuntu"
VERSION="16.04.2 LTS (Xenial Xerus)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 16.04.2 LTS"
VERSION_ID="16.04"
HOME_URL="http://www.ubuntu.com/"
SUPPORT_URL="http://help.ubuntu.com/"
BUG_REPORT_URL="http://bugs.launchpad.net/ubuntu/"
VERSION_CODENAME=xenial
UBUNTU_CODENAME=xenial
ubuntu@ip-172-31-11-231:~$ uname -r
4.4.0-1013-aws
ubuntu@ip-172-31-11-231:~$

Type/provider of hosts: (VirtualBox/Bare-metal/AWS/GCE/DO) AWS

Setup details: (single node rancher vs. HA rancher, internal DB vs. external DB) single node rancher with external DB

Environment Template: (Cattle/Kubernetes/Swarm/Mesos) Kubernetes

Steps to Reproduce: Upgrade from 1.5.4 k8s to 1.6.6. Scale a deployment up by 1 pod.

Results: Deployment prints this error:

Search Line limits were exceeded, some dns names have been omitted, the applied search line is: default.svc.cluster.local svc.cluster.local cluster.local kubelet.kubernetes.rancher.internal kubernetes.rancher.internal rancher.internal

About this issue

Original URL
State: closed
Created 7 years ago
Reactions: 5
Comments: 19 (8 by maintainers)

Most upvoted comments

@deniseschannon - Thanks! Do you have any ETA?

moshe0076 on Sep 18, 2017

@ryanwalls @superseb the problem with the domain names list being too long will be that the user search domains (that come last), will never exercise. So although the warning is seemingly harmless (besides creating “noise” in the UI) , we’d have to fix it by removing the entries generated by rancher-dns code + some changes to rancher-dns microservice.

alena1108 on Jul 18, 2017

I get this message on Rancher 1.6.7 running k8s:v1.7.2-rancher7 when a new pod is launched.

moshe0076 on Aug 21, 2017

This is fixed in v1.6.11. You should be able to run a kubectl describe pod to see what the problem is, or use the Kubernetes Dashboard.

superseb on Nov 29, 2017

@moshe0076 This has not been released and will be in the next Rancher and k8s release.

deniseschannon on Sep 11, 2017

Rancher server: v1.6-development

I tested with normal services and in kubernetes, i don’t see the problem anymore, and resolve.conf doesn’t have rancher specific search domain set, skydns resolution still works just fine.

galal-hussein on Sep 7, 2017

To test, a) use the k8s catalog template which starts kubelet with this label set. b) login to kubelet and see no rancher specific domains are set. c) create a simple replication controller, and see that the POD’s /etc/resolv.conf doesn’t have any rancher specific search domains set. d) verify that the skydns resolution still works fine.

kinarashah on Sep 7, 2017