2

I am getting the following error when trying to deploy a local Docker image to a pod in my local Minishift instance:

Failed to pull image "repo-name:port/app-name:1.0.0-SNAPSHOT-20190103151332485": 
rpc error: code = Unknown desc = Error: 
Status 400 trying to pull repository app-name: 
"{\n \"errors\" : [ {\n \"status\" : 400,\n \"message\" : 
\"Unsupported docker v1 repository request for 'docker'\"\n } ]\n}"

I have successfully deployed other pods locally using the same Docker environment.

Running docker images I can confirm my image exists locally.

Sam Berry
  • 7,394
  • 6
  • 40
  • 58

1 Answers1

2

It appears that this error can appear for a variety of reasons. Most commonly it seems to be due to invalid credentials or an issue communicating with a remote repository.

If you are certain your issue is not with remote communication, the problem might be that your image is failing to start. I have not had a chance to look into it but a better error message for this would be a great contribution to Minishift, if possible.

Test Docker Image Manually

To verify your docker image is working, try to run it manually.

Find the Image

Run: $ docker images

You should see a list that contains the image Minishift is trying to deploy. For example:

REPOSITORY                TAG                                IMAGE ID            CREATED             SIZE
repo-name:port/app-name   1.0.0-SNAPSHOT-20190103151332485   3e050126264c        14 minutes ago      704MB
repo-name:port/app-name   1.0.0-SNAPSHOT-20190103150418331   4293956e114f        23 minutes ago      704MB
repo-name:port/app-name   1.0.0-SNAPSHOT-20190103145227835   81fc7783e38f        35 minutes ago      704MB
repo-name:port/app-name   1.0.0-SNAPSHOT-20190103145203603   c5fa4815ee97        35 minutes ago      704MB

Run the Image

Once you have identified the Image ID of the image you are trying to deploy, run:

$ docker run -i -t <Image ID> /bin/bash

For example:

$ docker run -i -t 3e050126264c /bin/bash

Diagnosis

If the image fails with an error, you have likely located the problem with your Minishift deploys. Diagnose the issue until you have a healthy image then try to deploy again.

In my personal instance, I find it most reliable to delete the old application from Minishift before deploying a healthy version.

Sam Berry
  • 7,394
  • 6
  • 40
  • 58