docker-update

A small, simple bash utility for easily updating Docker images

Usage

This utility has been designed to be as simple and easy to use as possible, and accepts only a container name, image tag, or a list of container names/image tags, space-separated. There are no options or arguments.

Examples

• docker-update container
• docker-update author/image:latest
• docker-update container-1 container-2 docker.io/author/image:latest

This utility can also be sourced as a script to make use of the functions directly. There are two functions:

• docker_update(): The actual update function. Handles a space-separated list of container names only as arguments. Can differentiate between standalone containers and Docker Compose stacks, and prioritizes Compose stacks over standalone containers.
• get_container_names(): Builds a list of space-separated container names out of either container names directly, image tags, or a combination thereof. Prints this list, space-separated, to stdout.

Installation

To install this utility, place or symlink it anywhere in your PATH, i.e. /usr/local/bin, or any other directory you have added to your PATH. For example, I prefer to symlink it to ~/.local/bin/docker-update:

1. git clone https://github.com/xBelladonna/docker-update ~/scripts/docker-update
2. ln -s ~/scripts/docker-update/docker-update ~/.local/bin/docker-update

This utility makes three assumptions about your infrastructure:

• That standalone containers have matching scripts that create/start their respective containers and that they all have the same name. For example, a series of subdirectories named after the container (not necessarily the same as the container name), containing files named run.sh
• That these scripts and any Docker Compose stacks explicitly specify container names
• That all these scripts and/or Compose stacks are contained within one directory or its subdirectories

Therefore, the utility will need to know which directory these files are contained within. This is set using the environment variable DOCKER_UPDATE_DIR, i.e. in your ~/.profile, ~/.bashrc, ~/.zshrc, etc.

• export DOCKER_UPDATE_DIR=~/docker

Additionally, it needs to know the name of the script used to create/start your containers, which is set using the environment variable DOCKER_UPDATE_RUNSCRIPT, defaulting to run.sh.

• export DOCKER_UPDATE_RUNSCRIPT=start.sh

The utility will recurse through the directories contained within DOCKER_UPDATE_DIR and its subdirectories, searching the contents of files named the value of DOCKER_UPDATE_RUNSCRIPT and docker-compose.yml for container names, and use those scripts to recreate their respective containers once their parent images are updated.

Tab-completion

This utility is able to tab-complete container names once the appropriate completion mechanism has been installed for your respective shell(s).

• To enable tab completion for bash, add the contents of .bash_completion to your ~/.bash_completion file: cat .bash_completion >> ~/.bash_completion
• To enable tab completion for zsh, source _docker-update.zsh through any method you prefer, e.g. your ~/.zshrc file.
  • If you use Oh My Zsh, you can add this file as a plugin:
    1. Create a directory named docker-update under your ZSH_CUSTOM directory, ~/.oh-my-zsh/custom by default:
       • mkdir $ZSH_CUSTOM/docker-update
    2. Copy _docker-update.zsh to _docker-update inside that directory:
       • cp _docker-update.zsh $ZSH_CUSTOM/docker-update/_docker-update
    3. Add docker-update to your list of plugins in ~/.zshrc:
       • plugins=(git zsh-syntax-highlighting docker docker-compose docker-update)