Server installation

Install a standalone CamusDB node to run SQL locally or alongside your application. Choose the package that fits your environment; all three options run the same server and expose the same SQL APIs.

Method	Requirements	Choose it when
NuGet tool	.NET 10 SDK to install the tool	You already use .NET and want a global camusdb command.
Homebrew	Homebrew on macOS or Linux	You want a native command or background service with its runtime included.
Docker	A running Docker engine	You want a container with data kept in a named volume.

Install and start

Choose one method. The examples start a standalone node with REST on port 5095 and gRPC on 5096. Stop an existing node before starting another on the same ports.

NuGet tool

Install the .NET 10 SDK, then install the server from NuGet:

dotnet tool install --global CamusDB.Server
camusdb

With no configuration file, the node starts on built-in defaults and stores data in your user data directory. To keep settings of your own:

camusdb init

This creates a starter configuration and the default data directory. Edit the file and restart camusdb; see configuration and data for the paths. Stop a foreground node with Ctrl+C.

To update, stop the node, update the tool, and start it again:

dotnet tool update --global CamusDB.Server
camusdb

Homebrew

On macOS or Linux, install with Homebrew. The package includes the .NET runtime, so the server needs no separate .NET installation:

brew install camusdb/tap/camusdb
camusdb

This runs in the foreground with your user configuration. Stop it with Ctrl+C. To run a background service instead:

brew services start camusdb

The service uses its own configuration and data directory under $(brew --prefix)/var/camusdb; it does not use the foreground command's user configuration. Homebrew creates the service configuration on installation and preserves your edits on upgrade.

Manage the service with:

brew services stop camusdb
brew services restart camusdb

Update the package and restart the service to load the new version:

brew upgrade camusdb
brew services restart camusdb

If you run the foreground command instead, stop it before upgrading and run camusdb again afterward.

Docker

With Docker running, start the published image:

docker run --rm \
  -p 5095:5095 \
  -p 5096:5096 \
  -v camus-data:/data \
  --name camusdb camusdb/camusdb:latest

The image includes the runtime. The camus-data volume keeps database files under /data, independently of the container. --rm removes the stopped container, while the named volume remains for the next run.

Stop the node with Ctrl+C or, from another terminal:

docker stop camusdb

To update, stop the container, pull the image, and repeat the start command with the same volume:

docker pull camusdb/camusdb:latest

Use a version tag in place of latest when you need a fixed server version.

Check the node

From another terminal, check that the standalone node is responding:

curl --fail http://localhost:5095/health

To run SQL, connect with camus-cli or the Web Console. They are separate installations. Follow the tutorial to create a database and run your first queries. A database must be created explicitly before use.

Configuration and data

These are the default locations; explicit configuration and environment settings can override them:

Installation	Configuration	Data
Foreground command on macOS/Linux	~/.camusdb/config.yml, created by camusdb init	$XDG_DATA_HOME/camusdb, or ~/.local/share/camusdb when unset
NuGet tool on Windows	%APPDATA%\camusdb\config.yml, created by camusdb init	%LOCALAPPDATA%\camusdb
Homebrew service	$(brew --prefix)/var/camusdb/config.yml	$(brew --prefix)/var/camusdb/data
Docker command above	Built-in/image settings unless overridden	/data, backed by the camus-data volume

The Homebrew service writes its log to $(brew --prefix)/var/log/camusdb.log. A foreground process or Docker container writes to its console; use docker logs camusdb while the container exists.

For installed commands, CAMUS_HOME relocates the user configuration to $CAMUS_HOME/config.yml and the default data directory to $CAMUS_HOME/data. --config <path> selects a configuration file explicitly. See configuration discovery for precedence and the settings reference for ports, authentication, TLS, and storage options.

Run a cluster

The commands above run one standalone node. For a local cluster, start with the Aspire sample, or follow cluster mode for manual setup. For Kubernetes, use the Kubernetes Operator.