Multi-Node & Delegate Setup

Multi-Node & Delegate Setup

Running multiple Flux nodes is a common strategy for maximizing rewards and diversifying risk. This guide covers multi-node configuration, the P2SH delegate system for shared node operation, and Proxmox-based setups for running multiple nodes on a single physical server.

Multi-Node Configuration

Each Flux node requires its own IP address and its own collateral UTXO. There are several approaches:

UPnP Multi-Node (Home Hosting)

If hosting multiple nodes behind a single router, each node needs a unique API port. The default port is 16127, and allowed alternative ports are: 16137, 16147, 16157, 16167, 16177, 16187, 16197. This gives you a maximum of 8 nodes per IP address.

1. 1

   Enable UPnP on your router

   Access your router settings and enable UPnP (Universal Plug and Play). Most modern routers support this.

2. 2

   Run Multitoolbox Option 14

   On each node, run Multitoolbox and select Option 14 to configure UPnP port mapping. Assign a unique port to each node.

3. 3

   Verify port accessibility

   From an external network, verify each port is reachable: `curl http://YOUR_PUBLIC_IP:16137/flux/version`

Proxmox Fractus Setup

Proxmox VE is a popular open-source hypervisor for running multiple Flux nodes on one physical server. The "Fractus" approach refers to splitting a single high-spec server into multiple VMs:

• •Hardware: A single Stratus-class server (16 cores, 64 GB RAM, 1 TB SSD) can host up to 4 Cumulus nodes or 2 Nimbus nodes
• •Storage: Use ZFS or LVM-thin for efficient storage allocation. Thunder storage pools provide best performance.
• •Networking: Each VM needs a unique IP (use multiple IPs from your hosting provider or VLAN configuration)
• •Resources: Allocate slightly more than minimum per VM — leave headroom for system overhead

P2SH Delegate System

The Flux delegate system allows multiple parties to collaborate on running a node using P2SH (Pay-to-Script-Hash) transactions. This is useful for shared ownership scenarios or managed hosting where the host doesn't hold the collateral:

1. 1

   Create a delegate key pair

   Use the `createdelegatekeypair` RPC command to generate a delegate keypair. The delegate key proves operational control without requiring collateral custody.

2. 2

   Create a P2SH start transaction

   The collateral holder creates a P2SH start transaction that includes the delegate's public key. This authorizes the delegate to operate the node.

3. 3

   Sign and broadcast

   Both parties sign the transaction. The collateral holder signs with their key, and the delegate signs with their delegate key.

4. 4

   Start the node

   The delegate can now start and manage the node using `startfluxnodeasdelegate` or `startp2shasdelegate` commands.

Fleet Management Tips

• •Stagger updates: Never update all nodes simultaneously. Stagger by 10–15 minutes to maintain availability.
• •Diversify hosting: Spread nodes across different hosting providers and geographic regions to avoid correlated failures.
• •Centralized monitoring: Use Fluxme.io's wallet-based view to monitor all nodes tied to a single payment address.
• •Automate with scripts: For large fleets, write SSH-based scripts to batch-check status and trigger updates across all nodes.
• •Separate wallets: Consider using different payment addresses for different groups of nodes for organizational clarity.