BAZEL_SETUP.md

Bazel Build System Setup for EventStreamMonitor

This document explains how to use Bazel to build, test, and deploy the EventStreamMonitor microservices.

What Bazel Provides

Bazel offers several advantages over traditional Docker-based builds:

1. Fast, Incremental Builds

• Only rebuilds what changed
• Parallel execution of independent tasks
• Smart caching reduces build times significantly

2. Reproducible Builds

• Same inputs always produce same outputs
• Deterministic builds across different environments
• Eliminates "works on my machine" issues

3. Dependency Management

• Explicit dependency graph
• Automatic dependency resolution
• Prevents circular dependencies

4. Multi-Language Support

• Builds Python code, proto files, and Docker images
• Consistent build process across all components

5. Testing

• Run tests with proper isolation
• Parallel test execution
• Better test reporting

6. Remote Caching (Optional)

• Share build cache across team
• Faster CI/CD pipelines
• Reduced build times in cloud environments

Prerequisites

1. Install Bazel: Follow instructions at https://bazel.build/install

   # macOS
   brew install bazel
   
   # Or download from: https://github.com/bazelbuild/bazel/releases

2. Verify Installation:

   bazel version

Project Structure

EventStreamMonitor/
├── WORKSPACE              # Bazel workspace configuration
├── BUILD                  # Root BUILD file
├── .bazelrc              # Bazel configuration flags
├── requirements-lock.txt  # Pinned Python dependencies
├── common/               # Shared library
│   └── BUILD            # (to be created)
├── proto/               # Protocol buffer definitions
│   └── BUILD           # (included in root BUILD)
└── services/           # Microservices
    ├── taskprocessing/
    │   └── BUILD       # Service BUILD file
    ├── usermanagement/
    │   └── BUILD
    ├── notification/
    │   └── BUILD
    └── logmonitor/
        └── BUILD

Basic Bazel Commands

Build All Services

# Build all services
bazel build //services/...

# Build specific service
bazel build //services/taskprocessing:taskprocessing

# Build Docker image
bazel build //services/taskprocessing:taskprocessing_docker

Run Services Locally

# Run a service
bazel run //services/taskprocessing:taskprocessing

# Run with custom arguments
bazel run //services/taskprocessing:taskprocessing -- --port=9092

Test Services

# Run all tests
bazel test //...

# Run tests for specific service
bazel test //services/taskprocessing/...

# Run tests with output
bazel test //services/taskprocessing/... --test_output=all

Build Docker Images

# Build Docker image for a service
bazel build //services/taskprocessing:taskprocessing_docker

# Load image into local Docker (after building)
docker load -i bazel-bin/services/taskprocessing/taskprocessing_docker.tar

# Or use bazel run to build and load
bazel run //services/taskprocessing:taskprocessing_docker -- --norun

Query Build Graph

# See dependencies of a target
bazel query --output=graph //services/taskprocessing:taskprocessing | dot -Tpng > deps.png

# List all targets
bazel query //...

# Find what depends on common library
bazel query 'rdeps(//..., //:common)'

Integration with Docker Compose

You can use Bazel-built images with Docker Compose:

1. Build images with Bazel:

   bazel build //services/...
   bazel build //services/taskprocessing:taskprocessing_docker
   bazel build //services/usermanagement:usermanagement_docker
   bazel build //services/notification:notification_docker
   bazel build //services/logmonitor:logmonitor_docker

2. Load images into Docker:

   docker load -i bazel-bin/services/taskprocessing/taskprocessing_docker.tar
   docker load -i bazel-bin/services/usermanagement/usermanagement_docker.tar
   docker load -i bazel-bin/services/notification/notification_docker.tar
   docker load -i bazel-bin/services/logmonitor/logmonitor_docker.tar

3. Tag images (optional):

   docker tag bazel/services/taskprocessing:taskprocessing_docker eventstreammonitor-taskprocessing-service:latest

4. Use in docker-compose.yml (update image references)

Common Use Cases

Development Workflow

1. Make code changes
2. Build only what changed:

   bazel build //services/taskprocessing:taskprocessing

3. Run service locally:

   bazel run //services/taskprocessing:taskprocessing

CI/CD Pipeline

# Build all services
bazel build //services/...

# Run all tests
bazel test //...

# Build Docker images
bazel build //services/...:..._docker

# Push images (after loading)
docker push your-registry/eventstreammonitor-taskprocessing-service:latest

Debugging Builds

# Verbose output
bazel build //services/taskprocessing:taskprocessing --verbose_failures

# See what commands would run (dry run)
bazel build //services/taskprocessing:taskprocessing --nobuild

# Clean build (remove cache)
bazel clean
bazel build //services/taskprocessing:taskprocessing

Advanced Features

Remote Caching

Configure remote cache in .bazelrc:

build --remote_cache=https://your-cache-server.com

Build Performance

# Use more jobs (parallel builds)
bazel build //... --jobs=16

# Profile build
bazel build //... --profile=profile.json
# View: chrome://tracing -> load profile.json

Python Dependencies

Update requirements-lock.txt with pinned versions, then:

# Update pip dependencies
bazel sync --only=pip_deps

# Or manually update in WORKSPACE

Troubleshooting

Common Issues

1. "Module not found" errors:

   • Check deps in BUILD files
   • Ensure imports match package structure
   • Run bazel clean --expunge and rebuild

2. Proto compilation errors:

   • Verify proto file paths in BUILD
   • Check proto syntax and imports

3. Docker image build fails:

   • Ensure base image is available
   • Check file paths in container_image rule

4. Slow builds:

   • Use remote caching
   • Increase --jobs flag
   • Check for unnecessary dependencies

Next Steps

1. Fine-tune BUILD files for each service's specific needs
2. Add test targets using py_test rules
3. Set up remote caching for team/CI
4. Integrate with CI/CD (GitHub Actions, Jenkins, etc.)
5. Create deployment scripts using Bazel-built images

Resources

• Bazel Documentation
• Rules Python
• Rules Docker
• Rules Proto