Problem starting Docker/Kubernetes after upgrade MacOS to 10.13.5

[raymondvalentino]

• [x ] I have tried with the latest version of my channel (Stable or Edge)
• [x ] I have uploaded Diagnostics
• Diagnostics ID: 459DBC40-39A7-4D1D-B2B6-8F7DE23F3F15

Expected behavior

Start or restart Docker will start both Docker for Mac and Kubernetes

Actual behavior

Docker starts but Kubernetes reports "Kubernetes is starting ..." forever
20180608-112126.tar.gz

Information

• macOS Version:
  10.13.5

Diagnostic logs

Note that a Console display of Docker logs shows the continuous adding of a error line "com.docker.driver.amd64-linux Node status error: Unauthorized"```
Docker for Mac: version...
18.05.0-ce-mac67 (25042) (Edge)

### Steps to reproduce the behavior
<!--
A reproducible case, Dockerfiles FTW.
-->

  1. Restart Docker or Reboot my Mac
  2.Docker starts but Kubernetes never starts - reports "Kubernetes is starting .." forever

[raymondvalentino]

I resolved the issue by resetting to factory defaults. This means I get to re-create all my docker and kubernetes configurations (sigh). I am not sure if this is a clue, but I noticed that my Mac disk image file was a qcow2 file and after resetting to factory defaults, it is a raw file.

[norbertmocsnik]

Same issue here after upgrading to macOS 10.13.5:

• "Kubernetes is starting" forever
• Docker diagnostics reports everything is OK
• In the Console app I can see a "Node status error: Unauthorized" error being logged by com.docker.driver.amd64-linux every second.

Docker 18.05.0-ce-mac67

[deepybee]

I had the same issue. I solved it by quitting Docker, then moved the Docker.qcow2 disk image out of $HOME/Library/Containers/com.docker.docker/Data/com.docker.driver.amd64-linux/, restarted Docker which spawned a new disk image and within a minute or so Kubernetes was showing as running again.

NB: make sure your existing containers are writing to explicitly defined volumes if you require persistent storage before doing this. Upon restart Docker will need to pull down the images again so make sure you've got any important data backed up. For me this was not a concern (thankfully).

[barshow]

I know this sounds really dumb, but some folks on my team ran the OSX migration assistant and ended up with an empty /etc/hosts/ file. Setting it to the default seemed to fix the issue.

##
# Host Database
#
# localhost is used to configure the loopback interface
# when the system is booting.  Do not change this entry.
##
127.0.0.1    localhost
255.255.255.255    broadcasthost
::1             localhost

[spackle0]

Not dumb. Worked for me. I migrated between laptops and /etc/hosts was empty on the new one. Added the default content, restarted Docker, and Kubernetes started up in a few seconds.

It does seem though, still, when I disable then enable Kubernetes, I need to do a Docker restart to have Kubernetes come up, but before your solution it never came up.

[docker-robott]

Closed issues are locked after 30 days of inactivity.
This helps our team focus on active issues.

If you have found a problem that seems similar to this, please open a new issue.

Send feedback to Docker Community Slack channels #docker-for-mac or #docker-for-windows.
/lifecycle locked