What it does
CapRover puts a browser dashboard and deployment CLI in front of Docker-based app hosting on your own server. It can build projects from a Git working tree or deploy an existing image, then route web apps through its server stack and manage settings such as domains and HTTPS. It suits developers who want repeatable app deployments without handing their runtime to a hosted PaaS, but who are willing to operate the underlying machine.
Put the server in place
CapRover is the control plane, not the computer that hosts your app. Start with a Linux server that has a public IP, Docker 25 or newer, and a wildcard DNS record if you want apps under separate subdomains. The project recommends Ubuntu 24.04; its guide warns that small machines can run short of memory during builds and says 512 MB may not be enough. You can install through a marketplace image or follow the self-managed setup.
The deploy loop
Create an app in the dashboard, add a captain-definition file at the project root, and deploy from the project directory with the documented CLI command:
caprover deploy
The CLI transfers the current Git commit and reports upload, build, and deployment progress. Uncommitted files and files ignored by Git are not sent, so check the commit you intend to ship before deploying. The definition can direct CapRover to build source or use a container image. A connected repository branch can also trigger builds automatically.
For a small service, the dashboard handles app-level configuration while CapRover supplies routing and automatic HTTPS. This keeps routine deployment work in one place, but the operator still chooses the server size, configures DNS and firewall access, and maintains the host. App data and the free CapRover software are separate from the provider's server charges. The documented rollback rebuilds an earlier app image; it does not restore environment variables or other app configuration.
A boundary worth knowing
CapRover is attractive when keeping control of the host matters more than avoiding server administration. Its one-click apps and webhook-triggered deployments help with common tasks, but you still own the Docker host and should budget memory for builds. The deployment webhook is a documented integration point; the pages reviewed did not establish a broader general-purpose management API.
Strengths
- Deploy from a Git project or a prebuilt image while keeping the runtime on a server you control.
- Provides a web dashboard, routing, and automatic HTTPS for apps on the managed server.
- Supports one-click apps, repository-triggered builds, and a deployment webhook.
Watch out for
- Requires a server with a public IP, Docker, suitable DNS, and reachable firewall ports.
- Builds can need more memory than a very small VPS provides, and the host remains your responsibility.
- An image rollback does not restore environment variables or other app settings.
Best for
- Developers who want a self-hosted alternative to a managed PaaS.
- Small teams deploying Dockerized or supported-language web apps to a VPS they operate.
Sources
- CapRover · Scalable, Free and Self-hosted PaaS (Oct 26, 2026)
- Getting Started | CapRover (Oct 26, 2026)
- Deployment Methods | CapRover (Oct 26, 2026)
- CapRover Apache License 2.0 (Oct 26, 2026)