What is Komodo?
Komodo is a web application for managing servers, builds, deployments, and automated procedures.
- Connect servers. Monitor CPU, memory, and disk usage with alerts. Connect to shell sessions.
- Deploy containers. Create, start, stop, and redeploy Docker containers. View status, logs, and exec into shells.
- Deploy compose stacks. Define compose files in the UI, on the host, or in a git repo with auto-deploy on push.
- Manage Docker Swarms. Connect swarm managers and deploy services and stacks across your cluster.
- Build images. Define the dockerfile in UI or clone a git repo. Supports AWS EC2 spot instances for scalable build capacity.
- Run automation. Orchestrate multi-step workflows with Procedures and Actions. Schedule automations to run regularly.
- Manage configuration. shared variable and secret with interpolation across all resources.
- Full audit trail. every change is recorded, with who made it and when.
There is no limit to the number of servers you can connect, and there never will be.
Architecture
Komodo is composed of two components: Core and Periphery.
| Component | Role |
|---|---|
| Core | Web server hosting the API and browser UI. All user interaction flows through Core. |
| Periphery | Small, stateless agent running on each connected server. Called by Core to perform actions, report system usage, and retrieve container logs. |
API
Core exposes a REST and WebSocket API for programmatic access. Client libraries are available:
Permissioning
Komodo has a granular, role-based permissioning system so teams of developers, operators, and administrators can collaborate safely. See Permissioning for details.
User sign-on supports username/password and OAuth (GitHub, Google, and generic OIDC). See Core Setup.
Docker
Komodo uses Docker as the container engine for building and deploying.
info
Podman is also supported via the podman → docker alias.