How to deploy Flagr on Kubernetes?

Flagr can be deployed using its Docker image; for Kubernetes, you need to create deployment and service configurations, but persistence and high availability setups require additional manual configuration as the documentation doesn't provide out-of-the-box Helm charts.

Flagr vs Unleash: which is better for open-source feature flags?

Flagr is written in Go and emphasizes performance with low-latency APIs, while Unleash is Node.js-based with a larger plugin ecosystem. Choose Flagr for high-throughput Go environments or Unleash for more extensibility and community integrations.

Can Flagr handle percentage-based rollouts for feature flags?

Yes, Flagr supports percentage-based targeting for feature flags and A/B testing variants through its segment and distribution settings, allowing gradual rollouts to specific user groups without code changes.

How to secure Flagr APIs in production?

Flagr relies on standard HTTP security practices; you must implement authentication, authorization, and TLS externally, as it doesn't include built-in advanced security features like OAuth or role-based access control out-of-the-box.

What databases does Flagr support for persistence?

Flagr uses SQL databases; based on typical setups, it supports PostgreSQL or MySQL, but you need to check the documentation for specific configuration details and migration steps.

How to monitor Flagr performance and errors?

Flagr provides basic metrics via its APIs and benchmarks, but comprehensive monitoring requires integrating with external tools like Prometheus or custom logging, as it lacks native dashboards for real-time analytics.

Open-Awesome

Flagr

Apache-2.0Go1.1.20

Flagr is an open-source Go microservice for feature flagging, A/B testing, and dynamic configuration.

Visit Website GitHub

2.6k stars202 forks0 contributors

What is Flagr?

Flagr is an open-source Go microservice that enables teams to deliver targeted experiences through feature flags, A/B testing, and dynamic configuration. It provides clear REST APIs for managing flags and evaluating them in real-time, helping teams safely roll out features and monitor their impact without code deploys.

Target Audience

Engineering teams and developers building microservices or distributed applications who need a language-agnostic, API-first solution for feature management and experimentation.

Value Proposition

Developers choose Flagr for its high performance, low-latency evaluation optimized for thousands of requests per second, and its standalone, easy-to-integrate design with official multi-language SDKs and an intuitive web UI.

Overview

Flagr is a feature flagging, A/B testing and dynamic configuration microservice

Use Cases

Best For

Safely rolling out new features to specific user segments or percentages without deploying code.
Running A/B tests to compare different variants and measure performance or user engagement.
Managing dynamic runtime configuration changes across multiple services in real-time.
Integrating feature flagging into a polyglot tech stack with official SDKs for Go, JavaScript, Python, and Ruby.
Handling high-throughput flag evaluation with low latency in production environments.
Teams needing a self-hosted, open-source alternative to commercial feature flag services with a full web UI.

Not Ideal For

Projects that require a fully managed, serverless feature flag service without any self-hosting overhead
Small teams or startups with limited DevOps resources who need plug-and-play feature flagging with minimal setup
Applications needing advanced, real-time analytics dashboards for experiments out-of-the-box
Systems where sub-millisecond latency for every flag evaluation is critical, as network calls to a separate microservice add overhead

Pros & Cons

Pros

High Performance Evaluation

Benchmarks show Flagr handles thousands of requests per second with mean latency under 400µs, optimized for low-latency production environments as evidenced in the README.

Language-Agnostic APIs

Provides clear REST APIs and official SDKs for Go, JavaScript, Python, and Ruby, making it easy to integrate into diverse tech stacks without language constraints.

Comprehensive Web UI

Includes an intuitive interface for creating, managing, and monitoring flags and experiments visually, reducing reliance on API-only management.

Open Source Flexibility

Apache 2.0 licensed and self-hosted, offering full control over data and customization without vendor lock-in, as highlighted in the community-driven approach.

Cons

Operational Overhead

Requires self-hosting and maintenance of a Go microservice, adding complexity in deployment, scaling, and monitoring compared to managed SaaS solutions.

Limited Advanced Analytics

Focuses on basic flag evaluation and A/B testing but lacks built-in statistical analysis or machine learning features for deeper experiment insights.

Ecosystem Maturity

While core SDKs exist, the ecosystem may have fewer third-party integrations, plugins, or community tools compared to established commercial alternatives.

Frequently Asked Questions

Related Projects

Caddy

Fast and extensible multi-platform HTTP/1-2-3 web server with automatic HTTPS

MinIO is a high-performance, S3 compatible object store, open sourced under GNU AGPLv3 license.

Stars61,208

Forks7,564

Last commit1 month ago

etcd

Distributed reliable key-value store for the most critical data of a distributed system

Stars51,787

Forks10,389