Skip to the article
Aniq-UI

VelraRun the agency locally

Run the agency locally

Get your package running on your computer, with Docker.

For the Full Stack package

DockerRecommendedManual setup

Before you start: install and open Docker Desktop, or Docker Engine with Compose.

View requirements

  1. Download your template

    Open My templates and download the ZIP for your package.

    Expected result: A file named like velra-real-estate-full-stack-….zip is in your downloads.

  2. Unzip and open the folder

    Open a terminal where the ZIP is, extract it and move into the velra-real-estate-full-stack folder. On Windows, you can extract it in File Explorer instead, then open a terminal in the folder.

    Terminal
    unzip -q velra-real-estate-full-stack-*.zip
cd velra-real-estate-full-stack

    Expected result: The folder lists admin-dashboard, back-end, storefront and docker-compose.yml, beside README.md, QUICKSTART.md and LICENSE.md.

    Having trouble?

  3. Start the apps

    Build and start the API, the public website and the agency dashboard. Nothing has to be configured first. The first start builds three images and fills the database with the demo agency, which takes a few minutes; later starts take seconds.

    Terminalin velra-real-estate-full-stack
    docker compose up --build

    A port is already in use?

    If the start stops with ports are not available, address already in use or port is already allocated, another program uses 3030, 3031 or 8000. Either stop it, or create a file named .env in the velra-real-estate-full-stack folder, beside docker-compose.yml, with the port you need, such as ADMIN_PORT=3041 (SITE_PORT for the website, API_PORT for the API). Run docker compose up --build again and open the app on its new port.

    A port is already in use

    Expected result: The three services keep running and the log stops scrolling. Leave this terminal open.

    Having trouble?

  4. Open the public website

    Open localhost:3030Local. English is at /en and Arabic at /ar. The website has no accounts: saved homes and the comparison are kept in the visitor's browser.

    Expected result: The website opens on its home page with the demo listings and their photos.

    Having trouble?

  5. Sign in to the agency dashboard

    Open localhost:3031Local. Sign in with admin@velra.example and the password Admin@123. For the agent, type agent@velra.example with Agent@123. Change both passwords before the stack is reachable from the internet.

    Expected result: The dashboard opens on its overview, with the published properties, the latest inquiries and the upcoming viewings.

    Having trouble?

Take a tour

What the demo agency looks like once everything runs.

AccountEmailPassword
Super Admin: every page and permissionadmin@velra.exampleAdmin@123
Agent: overview, properties, inquiries, viewings, the staging studio and the neighborhood listagent@velra.exampleAgent@123

The seed adds 17 properties (15 published, 2 drafts) with their photos, 6 neighborhoods, 2 approved staging previews, 5 inquiries and 8 viewing requests, so every screen opens full. These are local demo accounts only.

Stop and start again

Press Ctrl+C in the terminal to stop the apps, or run this from the velra-real-estate-full-stack folder. Your data, the database and the uploads, is kept in the velra-data volume.

Terminalin velra-real-estate-full-stack
docker compose down

Start again with the same command as step 3. Without --build it reuses the images you already built.

Terminalin velra-real-estate-full-stack
docker compose up

Keep your data

docker compose down -v also deletes the data volume. The next start seeds a fresh copy of the demo agency, and everything you created is gone.

Run the public website

This package is the public website on its own. It reads everything from the Velra API, so start the API first. Unless you build it with another address, the website expects the API at http://localhost:8000. The API is a separate package and is not in this download.

  1. Download your template

    Open My templates and download the Public Website ZIP.

  2. Unzip and open the folder

    Terminal
    unzip -q velra-website-*.zip
cd velra-website

    Expected result: The folder has a Dockerfile and a package.json.

    Having trouble?

  3. Build the image

    The NEXT_PUBLIC_* values are compiled into the JavaScript the browser loads, so they are build arguments. Without any, the website uses the API at http://localhost:8000/api and calls itself http://localhost:3030. For an API somewhere else, pass its address.

    Terminalin velra-website
    docker build -t velra-website .

    With your own API and domain:

    Terminalin velra-website
    docker build -t velra-website --build-arg NEXT_PUBLIC_API_BASE_URL=https://api.your-agency.com/api --build-arg NEXT_PUBLIC_SITE_URL=https://www.your-agency.com --build-arg NEXT_PUBLIC_MEDIA_HOSTNAME=media.your-agency.com .
    Having trouble?

  4. Start it

    Inside the container localhost is the container itself, so API_INTERNAL_URL tells the website's own server where to reach the API on your computer, for property page titles, the first render of a property page and the sitemap. --add-host makes host.docker.internal work on Linux; Docker Desktop already has it. If port 3030 is taken, change the number on the left of -p, for example -p 3041:3030, open that port instead, and add http://localhost:3041 to the API's CORS_ORIGIN.

    Terminalin velra-website
    docker run -p 3030:3030 -e API_INTERNAL_URL=http://host.docker.internal:8000/api --add-host=host.docker.internal:host-gateway velra-website

    Expected result: localhost:3030Local opens the website with the listings from your API. English is at /en, Arabic at /ar.

    Having trouble?

Run the agency dashboard

This package is the agency dashboard on its own. Every screen reads the Velra API, so start the API first. Unless you build it with another address, the dashboard expects the API at http://localhost:8000. The API is a separate package and is not in this download.

  1. Download your template

    Open My templates and download the Agency Dashboard ZIP.

  2. Unzip and open the folder

    Terminal
    unzip -q velra-agency-dashboard-*.zip
cd velra-agency-dashboard

    Expected result: The folder has a Dockerfile and a package.json.

    Having trouble?

  3. Build the image

    Without build arguments the dashboard uses the API at http://localhost:8000/api. For an API somewhere else, pass its address, its address without /api for the live notifications, and the website's address for the "View website" link.

    Terminalin velra-agency-dashboard
    docker build -t velra-dashboard .

    With your own API:

    Terminalin velra-agency-dashboard
    docker build -t velra-dashboard --build-arg NEXT_PUBLIC_API_BASE_URL=https://api.your-domain.com/api --build-arg NEXT_PUBLIC_WEBSOCKET_BASE_URL=https://api.your-domain.com --build-arg NEXT_PUBLIC_MEDIA_HOSTNAME=api.your-domain.com --build-arg NEXT_PUBLIC_SITE_URL=https://your-domain.com .
    Having trouble?

  4. Start it and sign in

    If port 3031 is taken, change the number on the left of -p, for example -p 3041:3031, open that port instead, and add http://localhost:3041 to the API's CORS_ORIGIN (and set its FRONTEND_URL to it, the address the live notifications connect from).

    Terminalin velra-agency-dashboard
    docker run -p 3031:3031 velra-dashboard

    Expected result: localhost:3031Local opens the sign-in form. Sign in with admin@velra.example and the password Admin@123, which a seeded API accepts.

    Having trouble?

Run the API

This package is the NestJS API on its own, on a SQLite file kept in a Docker volume with the uploads. The public website and the agency dashboard are separate packages and are not in this download.

  1. Download your template

    Open My templates and download the Backend API ZIP.

  2. Unzip and open the folder

    Terminal
    unzip -q velra-backend-*.zip
cd velra-backend

    Expected result: The folder has a Dockerfile, a docker-entrypoint.sh and a package.json.

    Having trouble?

  3. Build the image

    Terminalin velra-backend
    docker build -t velra-api .
    Having trouble?

  4. Start it with the demo agency

    A new volume gets empty tables and no accounts unless you ask for the demo agency, so keep -e SEED_DEMO_DATA=true for a first start. The velra-api-data volume keeps the database and the uploads between runs. If port 8000 is taken, change the number on the left of -p, for example -p 8010:8000, and add -e PUBLIC_MEDIA_URL=http://localhost:8010/media so the photo addresses use it too. Your frontends must then point at http://localhost:8010/api.

    Terminalin velra-backend
    docker run -p 8000:8000 -v velra-api-data:/data -e SEED_DEMO_DATA=true velra-api

    Expected result: localhost:8000/api/healthLocal answers with "status":"ok" and "database":"up".

    Having trouble?

The container accepts the website on http://localhost:3030 and the dashboard on http://localhost:3031. On other addresses, set CORS_ORIGIN (both, comma separated) and FRONTEND_URL (the dashboard) with -e.

To give the container your own keys (R2, AI, MCP), create .env from .env.example with cp .env.example .env, fill it in, and add --env-file .env -e NODE_ENV=production to docker run:

Terminalin velra-backend
docker run -p 8000:8000 -v velra-api-data:/data -e SEED_DEMO_DATA=true --env-file .env -e NODE_ENV=production velra-api

A value given with -e wins over the same name in the file, so the container runs in production whatever .env says, and it moves the file's relative paths (SQLITE_DATABASE, MEDIA_UPLOAD_DIR) into the volume. Without a JWT_SECRET of your own, or with the example one, the container generates a secret and keeps it in the volume, so sign-ins survive a restart.

Next

Stuck on a step?

Find a fix before you start over.

Troubleshooting

Cookie Preferences

We use cookies to enhance your browsing experience, analyze site traffic, and personalize content. By clicking "Accept All", you consent to our use of cookies for analytics and personalized advertising.