Skip to content

Installation

To deploy mealie on your local network it is highly recommended to use docker to deploy the image straight from dockerhub. Using the docker-compose below you should be able to get a stack up and running easily by changing a few default values and deploying. You can deploy with either SQLite (default) or Postgres. SQLite is sufficient for most use cases. Additionally, with mealies automated backup and restore functionality, you can easily move between SQLite and Postgres as you wish.

Get Docker

Mealie on Dockerhub

  • linux/amd64
  • linux/arm64

Quick Start - Docker CLI

Deployment with the Docker CLI can be done with docker run and specify the database type, in this case sqlite, setting the exposed port 9925, mounting the current directory, and pull the latest image. After the image is up and running you can navigate to http://your.ip.address:9925 and you'll should see mealie up and running!

docker run \
    -p 9925:80 \
    -v `pwd`:'/app/data/' \
    hkotel/mealie:latest

Default Credentials

Username: changeme@email.com

Password: MyPassword

Docker Compose with SQLite

Deployment with docker-compose is the recommended method for deployment. The example below will create an instance of mealie available on port 9925 with the data volume mounted from the local directory. To use, create a docker-compose.yml file, paste the contents below and save. In the terminal run docker-compose up -d to start the container.

version: "3.1"
services:
  mealie:
    container_name: mealie
    image: hkotel/mealie:latest
    restart: always
    ports:
      - 9925:80
    environment:
      PUID: 1000
      PGID: 1000
      TZ: America/Anchorage

      # Default Recipe Settings
      RECIPE_PUBLIC: 'true'
      RECIPE_SHOW_NUTRITION: 'true'
      RECIPE_SHOW_ASSETS: 'true'
      RECIPE_LANDSCAPE_VIEW: 'true'
      RECIPE_DISABLE_COMMENTS: 'false'
      RECIPE_DISABLE_AMOUNT: 'false'

      # Gunicorn
      WEB_CONCURRENCY: 2
      # WORKERS_PER_CORE: 0.5
      # MAX_WORKERS: 8
    volumes:
      - ./mealie/data/:/app/data

Docker Compose with Postgres (BETA)

Postgres support was introduced in v0.5.0. At this point it should be used with caution and frequent backups.

version: "3.1"
services:
  mealie:
    container_name: mealie
    image: hkotel/mealie:latest
    restart: always
    ports:
      - 9090:80
    depends_on:
      - postgres
    environment:
      PUID: 1000
      PGID: 1000
      TZ: America/Anchorage

      # Database Settings
      DB_ENGINE: postgres # Optional: 'sqlite', 'postgres'
      POSTGRES_USER: mealie
      POSTGRES_PASSWORD: mealie
      POSTGRES_SERVER: postgres
      POSTGRES_PORT: 5432
      POSTGRES_DB: mealie

      # Default Recipe Settings
      RECIPE_PUBLIC: 'true'
      RECIPE_SHOW_NUTRITION: 'true'
      RECIPE_SHOW_ASSETS: 'true'
      RECIPE_LANDSCAPE_VIEW: 'true'
      RECIPE_DISABLE_COMMENTS: 'false'
      RECIPE_DISABLE_AMOUNT: 'false'

      # Gunicorn
      WEB_CONCURRENCY: 2
      # WORKERS_PER_CORE: 0.5
      # MAX_WORKERS: 8
    volumes:
      - ./mealie/data/:/app/data
  postgres:
    container_name: postgres
    image: postgres
    restart: always
    environment:
      POSTGRES_PASSWORD: mealie
      POSTGRES_USER: mealie

Env Variables

Variables Default Description
PUID 911 UserID permissions between host OS and container
PGID 911 GroupID permissions between host OS and container
DEFAULT_GROUP Home The default group for users
DEFAULT_EMAIL changeme@email.com The default username for the superuser
BASE_URL http://localhost:8080 Used for Notifications
DB_ENGINE sqlite Optional: 'sqlite', 'postgres'
POSTGRES_USER mealie Postgres database user
POSTGRES_PASSWORD mealie Postgres database password
POSTGRES_SERVER postgres Postgres database server address
POSTGRES_PORT 5432 Postgres database port
POSTGRES_DB mealie Postgres database name
TOKEN_TIME 2 The time in hours that a login/auth token is valid
RECIPE_PUBLIC True Default Recipe Settings - Make Recipe Public
RECIPE_SHOW_NUTRITION True Default Recipe Settings - Show Recipe Nutrition
RECIPE_SHOW_ASSETS True Default Recipe Settings - Show Recipe Assets
RECIPE_LANDSCAPE_VIEW True Default Recipe Settings - Set Landscape View
RECIPE_DISABLE_COMMENTS False Default Recipe Settings - Disable Comments
RECIPE_DISABLE_AMOUNT False Default Recipe Settings - Disable Amount
API_PORT 9000 The port exposed by backend API. Do not change this if you're running in Docker
API_DOCS True Turns on/off access to the API documentation locally.
TZ UTC Must be set to get correct date/time on the server
WORKERS_PER_CORE 1 Set the number of workers to the number of CPU cores multiplied by this value (Value * CPUs). More info here
MAX_WORKERS Set the maximum number of workers to use. Default is not set meaning unlimited. More info here
WEB_CONCURRENCY 2 Override the automatic definition of number of workers. More info here

Raspberry Pi 4

Fatal Python error: init_interp_main: can't initialize time

Some users experience an problem with running the linux/arm/v7 container on Raspberry Pi 4. This is not a problem with the Mealie container, but with a bug in the hosts Docker installation.

Update the host RP4 using instructions, summarized here:

wget http://ftp.us.debian.org/debian/pool/main/libs/libseccomp/libseccomp2_2.5.1-1_armhf.deb
sudo dpkg -i libseccomp2_2.5.1-1_armhf.deb

Advanced

Not Required

The items below are completely optional and are not required to manage or install your Mealie instance.

Custom Caddy File

The Docker image provided by Mealie contains both the API and the html bundle in one convenient image. This is done by using a proxy server to serve different parts of the application depending on the URL/URI. Requests sent to /api/* or /docs will be directed to the API, anything else will be served the static web files. Below is the default Caddyfile that is used to proxy requests. You can override this file by mounting an alternative Caddyfile to /app/Caddyfile.

{
  auto_https off
  admin off
}

:80 {
  @proxied path /api/* /docs /openapi.json

  root * /app/dist
  encode gzip
  uri strip_suffix /

  handle_path /api/recipes/image/* {
    root * /app/data/img/
    file_server
  }

  handle @proxied {
    reverse_proxy http://127.0.0.1:9000
  }

  handle {
    try_files {path}.html {path} /
    file_server
  }
}

Deployed without Docker

Unsupported Deployment

If you are experiencing a problem with manual deployment, please do not submit a github issue unless it is related to an aspect of the application. For deployment help, the discord server is a better place to find support.

Alternatively, this project is built on Python and SQLite so you may run it as a python application on your server. This is not a supported options for deployment and is only here as a reference for those who would like to do this on their own. To get started you can clone this repository into a directory of your choice and use the instructions below as a reference for how to get started.

There are three parts to the Mealie application

  • Frontend/Static Files
  • Backend API
  • Proxy Server

Frontend/ Static Files

The frontend static files are generated with npm run build. This is done during the build process with docker. If you choose to deploy this as a system application you must do this process yourself. In the project directory run cd frontend to change directories into the frontend directory and run npm install and then npm run build. This will generate the static files in a dist folder in the frontend directory.

Backend API

The backend API is build with Python, FastAPI, and SQLite and requires Python 3.9, and Poetry. Once the requirements are installed, in the project directory you can run the command poetry install to create a python virtual environment and install the python dependencies.

Once the dependencies are installed you should be ready to run the server. To initialize that database you need to first run python mealie/db/init_db.py. Then to start The web server, you run the command uvicorn mealie.app:app --host 0.0.0.0 --port 9000

Proxy Server

You must use a proxy server to server up the static files created with npm run build and proxy requests to the API. In the docker build this is done with Caddy. You can use the CaddyFile in the section above as a reference. One important thing to keep in mind is that you should drop any trailing / in the url. Not doing this may result in failed API requests.

Back to top