Thanks buddy!

Thanks for the sub! Sorry buddy but I've never messed with docker on windows nor have I ever created a windows gitlab runner. As far as I'm aware, docker can't create windows containers, it can only create linux containers, but can run those containers on a windows host. Is your testing platform windows or linux? If it's windows you might not want docker at all, just have a windows gitlab runner instead. Thanks for watching!

@@DanGitschooldude Thanks.Hyper-V enabled windows machine both type of containers . I m doing testing on windows. Yes we can work on window git lab runner as shell executor,but i m doing my testing with docker executor git lab runner which in my case is windows.So in my git lab runner config (config.toml) i m specifying the base image which git lab is using to create it work space.If possible please share make videos on docker executor git lab runner for windows. Thanks.

As far as I'm aware, there is no docker or registry capability withheld from Gitlab CE so there should be no difference between CE vs EE. It seems all Gitlab-CI-related capabilities are free, which is a smart move in my opinion. Thanks for watching! If you have any suggestions for future videos let me know.

There's no way to update an image without building it locally and pushing to the docker registry as far as I'm aware. Thanks for watching!

Well, if you want to avoid rebuilding the image locally, then you can write the .gitlab-ci.yaml file in which you can specify scripts to login to the registry, building the image and push it to the registry. Something like this: --- stages: - build publish_docker_image: stage: build image: docker:stable services: - docker:stable-dind script: - docker login -u "Insert Docker ID here" -p $DOCKER_HUB_PASSWORD - docker build -t "Insert Docker ID here"/"name of your image":$CI_COMMIT_REF_SLUG . - docker push "Insert Docker ID here"/"name of your image":$CI_COMMIT_REF_SLUG The example above uses the pre-made docker:stable-dind image as a service for the scripts. The first script logins to Docker Hub in this case using the written username and a password which comes from a environment variable $DOCKER_HUB_PASSWORD (defined for the pipeline at the project's repository in Gitlab) The second script builds the image using the files in the current directory and tags the image with the username, image name and the $CI_COMMIT_REF_SLUG tag which is automatically given by Gitlab (usually this is the branch name you are working on). The third script pushes the built image to the Docker Hub repository. (of course, if you change the login part to "docker login registry.gitlab.com" then you would be pushing to Gitlab registry).