FlagFlow • Prerequisites

Any container engine: Docker or Kubernetes

The FlagFlow system is delivered as a Docker image. Therefore, it is necessary to have a containerized runtime environment (Docker or Kubernetes) available. An enterprise environment is the ideal choice, but a simple Docker installation is sufficient for developing and testing FlagFlow. Before installing FlagFlow, please ensure that your Docker or Kubernetes environment is functioning correctly.

Docker Kubernetes

Storage Engine: etcd or Filesystem Storage

FlagFlow supports two storage options for storing its state: etcd (recommended for production) and filesystem storage (suitable for small companies or development).

etcd is the recommended storage engine for production environments, providing distributed, reliable key-value storage with real-time watching capabilities. It's ideal for environments requiring high availability, multiple replicas, and instant synchronization across instances.

Filesystem storage is available as an alternative for small companies without heavy infrastructure requirements or single-instance deployments. While it can work with replicas, changes have a few milliseconds delay across instances. When using filesystem storage, you must mount the /data volume to persist data across container restarts.

Hardware and System Requirements

FlagFlow has modest hardware requirements, making it suitable for various deployment scenarios from development to production environments.

Development Environment

CPU: 1 vCPU (minimum)
Memory: 1GB RAM (minimum)
Storage: 5GB available disk space
Network: Internet connection for container images

Production Environment

CPU: 2+ vCPUs (recommended)
Memory: 4GB+ RAM (recommended)
Storage: 20GB+ SSD storage
Network: Reliable network with low latency

These requirements scale based on your usage patterns, number of feature flags, and frequency of flag changes. Monitor resource usage and adjust accordingly.

Network and Port Requirements

FlagFlow requires specific network configurations and port access for proper operation.

Required Ports

Service	Port	Protocol	Purpose	External Access
FlagFlow	3000	HTTP	Web UI and API	Required
etcd (optional)	2379	HTTP	Client connections	Internal only
etcd (optional)	2380	HTTP	Peer communication	Internal only

Firewall Configuration

Only port 3000 needs external access for FlagFlow's web UI and API. If using etcd, do not expose etcd ports (2379, 2380) to external networks - this would create a significant security vulnerability.

Security Warning: If using etcd, etcd ports should remain internal to your container network or cluster. Exposing them externally allows unauthorized access to your feature flag data and configuration.

Security and Access Control

Proper security configuration is essential for production FlagFlow deployments.

Authentication

Configure strong passwords for etcd authentication (if using etcd)
Use TLS certificates for encrypted communication in production
Consider implementing network policies in Kubernetes environments
Secure filesystem permissions for /data volume (if using filesystem storage)

Data Protection

Enable encryption at rest for persistent storage
Use encrypted container registries for image storage
Implement backup strategies for your chosen storage engine (etcd data or /data volume)

Access Control

Limit network access to FlagFlow services
Use service accounts with minimal required permissions
Implement logging and monitoring for security events

Supported Software Versions

FlagFlow is tested and supported with the following software versions:

Container Platforms

Docker Engine: 20.0+ (recommended: latest LTS)
Docker Compose: 2.0+ (for compose deployments)
Kubernetes: 1.20+ (recommended: 1.25+)
Podman: 3.0+ (alternative to Docker)

Operating Systems

Linux: Any modern distribution
macOS: 10.15+ (with Docker Desktop)
Windows: Windows 10+ (with Docker Desktop)
Cloud: AWS, GCP, Azure, DigitalOcean

Pre-installation Verification Checklist

Before proceeding with FlagFlow installation, verify all prerequisites are met:

💡 Tip: For production deployments, consider running through this checklist in your staging environment first to validate the entire installation process.