Zum Hauptinhalt springen

Docker Quick Start - Ever Teams Web

Two Available Modes​

  • ✅ Hot reload: Code changes reload automatically
  • ✅ No rebuild needed for each change
  • ✅ Ideal for rapid development and testing
  • ⚠️ Larger image size (~1GB)

Production Mode​

  • ✅ Optimized and lightweight image (~200MB)
  • ✅ Maximum performance
  • ❌ No hot reload (rebuild required for each change)
  • ✅ Ideal for deployment

Quick Start - Development Mode​

Simple Method: Use the automated script​

# 1. Make the script executable (once)
chmod +x docker-run.sh

# 2. Build the development image (once)
./docker-run.sh dev:build

# 3. Start the application in development mode
./docker-run.sh dev

# The application will start at http://localhost:3030
# Edit your code and see changes in real-time!

# 4. Stop the application (Ctrl + C in terminal)
# Or in another terminal:
./docker-run.sh stop  # Stops ALL containers (dev + prod)

How does hot reload work?

The development mode uses mounted volumes + Next.js file watching (with polling):

  • ✅ Automatic hot reload: Changes appear in real-time
  • ✅ Compatible: Works with all Docker Compose versions
  • ✅ Reliable: Proven method for Docker development
  • ⚠️ Polling-based: Uses CHOKIDAR_USEPOLLING=true for file detection

Manual Method: Use docker-compose directly​

# 1. Build the development image
docker-compose -f docker-compose.dev.yml --env-file .env.dev.docker build

# 2. Start the application
docker-compose -f docker-compose.dev.yml --env-file .env.dev.docker up

# 3. Access the application
# Open your browser at: http://localhost:3030

# 4. Edit your code - the application reloads automatically!

# 5. Stop the application (Ctrl + C in terminal)
# Then to clean up:
docker-compose -f docker-compose.dev.yml down

Quick Start - Production Mode​

Simple Method: Use the automated script​

# 1. Make the script executable (once)
chmod +x docker-run.sh

# 2. Build the production image
./docker-run.sh build

# 3. Start the application
./docker-run.sh start

# 4. View logs
./docker-run.sh logs

# 5. Stop the application
./docker-run.sh stop  # Stops ALL containers (dev + prod)

Manual Method: Use docker-compose directly​

# 1. Build the image
docker-compose -f docker-compose.build.yml --env-file .env.docker build

# 2. Start the application
docker-compose -f docker-compose.build.yml --env-file .env.docker up

# 3. Access the application
# Open your browser at: http://localhost:3030

# 4. Stop the application (Ctrl + C in terminal)
# Then to clean up:
docker-compose -f docker-compose.build.yml down

⚠️ Important: stop Command​

The ./docker-run.sh stop command stops ALL containers (development AND production).

# Stops ALL containers (dev + prod)
./docker-run.sh stop

If you want to stop only a specific container:

# Stop only the development container
docker-compose -f docker-compose.dev.yml down

# Stop only the production container
docker-compose -f docker-compose.build.yml down

Useful Commands​

Run in background (detached mode)​

docker-compose -f docker-compose.build.yml --env-file .env.docker up -d

View logs​

docker logs webapp

View logs in real-time​

docker logs -f webapp

Stop the application​

docker-compose -f docker-compose.build.yml down

Rebuild the image (after code changes)​

docker-compose -f docker-compose.build.yml --env-file .env.docker build --no-cache

View running containers​

docker ps

Enter the container​

docker exec -it webapp /bin/bash

Configuration​

All environment variables are in the .env.docker file at the project root.

Change the API URL​

Open .env.docker and modify:

GAUZY_API_SERVER_URL=http://localhost:3000
NEXT_PUBLIC_GAUZY_API_SERVER_URL=http://localhost:3000

Then rebuild the image:

docker-compose -f docker-compose.build.yml --env-file .env.docker build

Change the port​

Open .env.docker and modify:

UI_PORT=3031

Then restart the application:

docker-compose -f docker-compose.build.yml --env-file .env.docker up

The application will be accessible at http://localhost:3031

Common Issues​

"Port 3030 is already in use"​

Change the port in .env.docker:

UI_PORT=3031

"Cannot connect to the Docker daemon"​

Make sure Docker Desktop is running.

Application doesn't load​

Check the logs:

docker logs webapp

Build fails with "Unable to download sentry-cli binary" or timeout errors​

This happens when optional binaries (Sentry CLI, Cypress, Playwright) fail to download during yarn install.

Solution: The Dockerfile.dev now skips these optional downloads by default using:

  • SENTRYCLI_SKIP_DOWNLOAD=1 - Skips Sentry CLI binary
  • CYPRESS_INSTALL_BINARY=0 - Skips Cypress binary
  • PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=1 - Skips Playwright browsers

If you still encounter timeout errors:

  1. Increase network timeout (already set to 600 seconds in Dockerfile.dev):

    RUN yarn install --frozen-lockfile --non-interactive --network-timeout 600000

  2. Check your internet connection - Slow or unstable connections can cause timeouts

  3. Use Docker BuildKit cache - Rebuild to use cached layers:

    ./docker-run.sh dev:build

  4. If on ARM64/M1/M2 Mac - Some binaries may not be available for ARM architecture. The skip flags prevent these issues.

Build is very slow (5-10 minutes)​

This is normal for the first build. Docker caches layers, so subsequent builds are much faster (30 seconds to 2 minutes).

To speed up builds:

  • ✅ Use ./docker-run.sh dev:build which uses BuildKit caching
  • ✅ Don't change package.json files frequently (invalidates cache)
  • ✅ Keep Docker Desktop running (preserves cache)