Generates fig.lock files based on fig.yml files. Gets the latest versions of all images defined in a fig.yml file, rewrites them to include the specific tags of those versions, generates a fig.lock file.
fig.lock is to fig.yml as Gemfile.lock is to Gemfile
Okay so, suppose you have a bunch of docker images. You want these images to work together to do something. You want whatever something is to be versioned because versioning is important - it helps you to figure out when a bug were introduced. It helps you roll back to a good version if something goes terribly wrong. All sorts of reasons!
Fig lets you describe your images and how they interact. But if you use latest tags in your image (or really anything other than explicit version locking that you do manually by always using specific tags for your images), it doesn't help you keep track of your versions - it just uses whatever docker images are installed locally, or fetches the specific tags from registries. This approach is fine if you're dealing with images that don't change very often. But if you're working in an environment where several images in different source repositories are under heavy development, it can be a problem to have to manually update the fig.yml file, and integration tests that auto-update might not be feasible.
This is why Fig::Lock exists. Fig::Lock lets you specify a fig.yml file containing images that use the latest tag. It generates fig.lock files by figuring out what explicit tag these "latest" tags actually map to, and then generating a fig.lock file that uses these exact images. Then you can run your integration tests using this file, check it into source control, etc. The point is that you get a file that represents all the versions of the things in your system, so you can use this to reproduce it exactly.
Add this line to your application's Gemfile:
gem 'fig-lock'
And then execute:
$ bundle
Or install it yourself as:
$ gem install fig-lock
fig-lock produces fig.lock files from fig.yml files, locking images to specific versions. eg:
fig.yml:
---
image: ubuntu
image: my.docker-registry.com/repo/myimage
resulting fig.lock:
---
image: ubuntu:14.04
image: my.docker-registry.com/repo/myimage:0.0.1
Currently, fig-lock only works properly with images that have been explicitly tagged. However, the next iteration should use hashes if tags are not available.
The install command uses docker to retrieve all images defined in fig.lock. If no fig.lock file exists, one will be generated by invoking the update command.
fig-lock install </path/to/fig.yml>
The update command uses docker to retrieve the latest versions of all images defined in the fig.yml file and generates/updates a corresponding fig.lock file in the same directory as the fig.yml file, which uses specific version tags for these images.
fig-lock update </path/to/fig.yml>
- Fork it ( http://github.com//fig-lock/fork )
- Create your feature branch (
git checkout -b my-new-feature) - Commit your changes (
git commit -am 'Add some feature') - Push to the branch (
git push origin my-new-feature) - Create new Pull Request