kubectl cannot communicate with new cluster

[xstevens]

I was trying out KIND on macOS. The cluster seems to startup fine and there's no indication of errors in the logs. However, kubectl cannot communicate with the server using the generated kubeconfig.

$ kubectl cluster-info

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
Unable to connect to the server: EOF

Here's a log of shell commands:

https://gist.github.com/xstevens/f43236505e7849d4079f8204c4a3f52b

[BenTheElder]

kind export logs will give us logs to look at, it does look like the cluster came up fine initially at least from what you posted.

/assign

[BenTheElder]

how much RAM does your docker daemon have configured?

docker for mac being low on RAM or disk space in the hyperkit vm is usually why something like this happens, that's something we need to cover more in the docs and maybe try to detect.

[xstevens]

I killed the previous cluster and tried 4 CPU / 4GB RAM for Docker and ended with the same result.

$ docker ps
CONTAINER ID        IMAGE                  COMMAND                  CREATED              STATUS              PORTS                      NAMES
c0261f7512fd        kindest/node:v1.12.2   "/usr/local/bin/entr…"   About a minute ago   Up About a minute   0.0.0.0:64907->64907/tcp   kind-1-control-plane
$ docker exec -it c0261f7512fd /bin/sh
# docker ps -a
CONTAINER ID        IMAGE               COMMAND             CREATED             STATUS              PORTS               NAMES
#

Attaching the export logs.
kubelet.log
journal.log
kubernetes-version.txt
docker.log
docker-info.txt

[BenTheElder]

Thanks! looks like disk space from the kubelet log:

Dec 07 00:37:53 kind-1-control-plane kubelet[688]: I1207 00:37:53.229561     688 eviction_manager.go:340] eviction manager: must evict pod(s) to reclaim ephemeral-storage
Dec 07 00:37:53 kind-1-control-plane kubelet[688]: E1207 00:37:53.229638     688 eviction_manager.go:351] eviction manager: eviction thresholds have been met, but no pods are active to evict

If you EG build kubernetes (or kind node images) often, it will eat up all of the space in the docker for mac disk, something like docker system prune usually helps.

[xstevens]

Cleared up my images with prune and removed some other old ones. Works now, thanks!

[BenTheElder]

Awesome!

I'll improve the docs around this. I think we can also probably make kind detect and warn on this situation.