⚠️⚠️⚠️ ouroboros is no longer in development. It does its job (more or less) and the devs have succumb to real life! Please feel free to fork and maintain as you wish. We appreciate all of the support in the last year :) ⚠️⚠️⚠️

Automatically update your running Docker containers to the latest available image.

The de-facto standard for docker update automation

Overview

Ouroboros will monitor (all or specified) running docker containers and update them to the (latest or tagged) available image in the remote registry. The updated container uses the same tag and parameters that were used when the container was first created such as volume/bind mounts, docker network connections, environment variables, restart policies, entrypoints, commands, etc.

• Push your image to your registry and simply wait your defined interval for ouroboros to find the new image and redeploy your container autonomously.
• Notify you via many platforms courtesy of Apprise
• Serve metrics for trend monitoring (Currently: Prometheus/Influxdb)
• Limit your server ssh access
• ssh -i key server.domainname "docker pull ... && docker run ..." is for scrubs
• docker-compose pull && docker-compose up -d is for fancier scrubs

Getting Started

More detailed usage and configuration can be found on the wiki.

Docker

Ouroboros is deployed via docker image like so:

docker run -d --name ouroboros \
  -v /var/run/docker.sock:/var/run/docker.sock \
  pyouroboros/ouroboros

This is image is compatible for amd64, arm32, and arm64 CPU architectures

or via docker-compose:

Official Example

Pip

Ouroboros can also be installed via pip:

pip install ouroboros-cli

And can then be invoked using the ouroboros command:

$ ouroboros --interval 300 --log-level debug

This can be useful if you would like to create a systemd service or similar daemon that doesn't run in a container

Examples

Per-command and scenario examples can be found in the wiki

Contributing

All contributions are welcome! Contributing guidelines are in the works