OnionBalance — Load Balancing for .onion Services

• High availability — If one backend server goes down, others continue serving traffic.
• Load distribution — Spread traffic across multiple servers to handle more concurrent users.
• DDoS mitigation — Distributed backends are harder to overwhelm with attacks.
• Geographic distribution — Place backends in different locations for better performance.

OnionBalance works by having a frontend that publishes a single .onion address, with traffic routed to multiple backend Tor instances:

1. Each backend server runs its own Tor hidden service
2. OnionBalance collects the descriptor information from each backend
3. It publishes a combined descriptor under the main .onion address
4. Tor clients connecting to the main .onion are routed to different backends

# Install OnionBalance
pip3 install onionbalance

# Generate configuration
onionbalance-config --n 3  # 3 backend instances

# This creates:
# - Master .onion address and keys
# - 3 backend instance configurations
# - config.yaml for OnionBalance

Deploy each backend configuration to a separate server, then run OnionBalance on the frontend to coordinate them.

Setting up OnionBalance correctly is complex. AnubizHost offers managed Tor hosting with built-in load balancing for high-traffic .onion services. We configure OnionBalance, manage multiple backends, and ensure your .onion service stays online and fast.