Deploy Server

TLDR

The Labtasker server can be deployed in two ways, depending on the database backend and the chosen deployment method.

Deployment Method	Server	Database
Python Native labtasker-server	Local Environment	A Python Emulated Embedded DB
docker compose	Run inside a container	MongoDB Service

Method 1. Python Native (Easy)

This is the simplest way to get started with Labtasker using only Python dependencies. The embedded database makes setup fast and straightforward.

pip install labtasker

Then, to start a Labtasker server (with embedded database) in the background, run the following command:

labtasker-server serve --host 0.0.0.0 --port 9321 &

Method 2. Docker Compose (Advanced)

This method is recommended for scenarios where you need more robust database capabilities and containerized deployment.

Prerequisites

• Docker Compose

Step 1: Configuration

1. Clone the repository:

   git clone https://github.com/luocfprime/labtasker.git
   cd labtasker
   

2. Create your environment file:

   cp server.example.env server.env
   

3. Configure your settings in server.env:

   • Configure MongoDB.
   • Configure server ports.
   • Configure how often you want to check for task timeouts.

Step 2: Start services

1. Start services (first time or update existing services):

   docker compose --env-file server.env up -d --pull always
   

2. Check status:

   docker compose --env-file server.env ps
   

3. View logs:

   docker compose --env-file server.env logs -f
   

Database Management

To expose MongoDB for external tools (this is potentially risky):

1. Set EXPOSE_DB=true in server.env
2. Optionally set DB_PORT to change the exposed port (default: 27017)
3. Use tools like MongoDB Compass to connect to the database.