Plasma: RPC failover with eRPC

Set up fault-tolerant RPC access for Plasma using eRPC proxy with Chainstack as primary and public fallback endpoints for production reliability.

Overview

This tutorial shows how to configure eRPC as a fault-tolerant proxy for Plasma, combining your Chainstack endpoint with public RPCs for automatic failover. When your primary endpoint experiences issues, eRPC seamlessly routes requests to backup providers.

By the end, you will have a local proxy that:

• Routes Plasma RPC calls through Chainstack as the primary provider
• Falls back to public endpoints when the primary is unavailable
• Caches responses to reduce latency and RPC calls
• Retries failed requests with exponential backoff

Prerequisites

• Chainstack account with a Plasma node deployed
• Docker installed
• Basic familiarity with JSON-RPC

For general eRPC setup and Chainstack Platform API auto-discovery, see Using eRPC with Chainstack: Quickstart.

Network parameters

Network	Chain ID	Currency
Plasma Mainnet	9745	XPL
Plasma Testnet	9746	XPL

1. Get your endpoints

Chainstack endpoint

1. Navigate to your Plasma node in the Chainstack console.
2. Copy your HTTPS endpoint from Access and credentials.

Public endpoints

Plasma provides public RPC endpoints as backup:

• https://rpc.plasma.to — public mainnet
• https://rpc-testnet.plasma.to — public testnet

Public endpoints have rate limits and no SLA. Use them as fallback only, not as primary.

2. Create the configuration

Create a file named erpc.yaml:

logLevel: info

server:
  httpHostV4: 0.0.0.0
  httpPortV4: 4000
  maxTimeout: 30s

projects:
  - id: plasma
    networks:
      - architecture: evm
        evm:
          chainId: 9745
        failsafe:
          timeout:
            duration: 30s
          retry:
            maxAttempts: 3
            delay: 500ms
            backoffMaxDelay: 5s
            backoffFactor: 2
            jitter: 200ms
          hedge:
            delay: 2s
            maxCount: 2

    upstreams:
      # Primary: Chainstack (higher priority)
      - id: chainstack-plasma
        type: evm
        endpoint: YOUR_CHAINSTACK_PLASMA_ENDPOINT
        rateLimitBudget: chainstack-budget
        evm:
          chainId: 9745
        failsafe:
          timeout:
            duration: 15s
          retry:
            maxAttempts: 2
            delay: 300ms
            backoffMaxDelay: 3s
            backoffFactor: 1.5
            jitter: 100ms

      # Fallback: Public RPC (lower priority)
      - id: plasma-public
        type: evm
        endpoint: https://rpc.plasma.to
        rateLimitBudget: public-budget
        evm:
          chainId: 9745
        failsafe:
          timeout:
            duration: 20s
          retry:
            maxAttempts: 2
            delay: 500ms
            backoffMaxDelay: 5s
            backoffFactor: 2
            jitter: 200ms

rateLimiters:
  budgets:
    - id: chainstack-budget
      rules:
        - method: '*'
          maxCount: 25
          period: 1s
        - method: 'eth_getLogs'
          maxCount: 10
          period: 1s
    - id: public-budget
      rules:
        - method: '*'
          maxCount: 5
          period: 1s

Replace YOUR_CHAINSTACK_PLASMA_ENDPOINT with your actual Chainstack endpoint URL.

Never commit configuration files with real endpoint URLs to version control. Use environment variables in production.

3. Run eRPC

Start the proxy with Docker:

docker run --rm \
  -v $(pwd)/erpc.yaml:/erpc.yaml \
  -p 4000:4000 \
  -p 4001:4001 \
  ghcr.io/erpc/erpc:latest

You should see output indicating both upstreams are configured:

INF starting eRPC server...
INF registered upstream id=chainstack-plasma network=evm:9745
INF registered upstream id=plasma-public network=evm:9745
INF http server started addr=0.0.0.0:4000

4. Test the setup

Basic connectivity

curl -X POST http://localhost:4000/plasma/evm/9745 \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"eth_chainId","params":[],"id":1}'

Expected response:

{"jsonrpc":"2.0","id":1,"result":"0x2611"}

The hex value 0x2611 equals 9745 (Plasma Mainnet chain ID).

Get latest block

curl -X POST http://localhost:4000/plasma/evm/9745 \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"eth_blockNumber","params":[],"id":1}'

Check USDT balance

curl -X POST http://localhost:4000/plasma/evm/9745 \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc":"2.0",
    "method":"eth_call",
    "params":[{
      "to":"0xB8CE59FC3717Ada4C02eadf9682A9e934F625ebb",
      "data":"0x70a08231000000000000000000000000YOUR_ADDRESS_HERE"
    },"latest"],
    "id":1
  }'

5. Use with your application

Point your application to the eRPC proxy instead of directly to an RPC endpoint.

ethers.js

const { ethers } = require("ethers");

const provider = new ethers.JsonRpcProvider("http://localhost:4000/plasma/evm/9745", {
  chainId: 9745,
  name: "plasma"
});

const blockNumber = await provider.getBlockNumber();
console.log("Current block:", blockNumber);

web3.py

from web3 import Web3

w3 = Web3(Web3.HTTPProvider("http://localhost:4000/plasma/evm/9745"))
print("Connected:", w3.is_connected())
print("Block:", w3.eth.block_number)

Hardhat

In hardhat.config.js:

module.exports = {
  networks: {
    plasma: {
      url: "http://localhost:4000/plasma/evm/9745",
      chainId: 9745,
      accounts: [process.env.PRIVATE_KEY]
    }
  }
};

6. Add testnet support

To also proxy Plasma Testnet, add another network to the configuration:

projects:
  - id: plasma
    networks:
      - architecture: evm
        evm:
          chainId: 9745
        # ... mainnet failsafe config
      - architecture: evm
        evm:
          chainId: 9746
        failsafe:
          timeout:
            duration: 30s
          retry:
            maxAttempts: 3
            delay: 500ms
            backoffMaxDelay: 5s
            backoffFactor: 2
            jitter: 200ms

    upstreams:
      # Mainnet upstreams
      - id: chainstack-plasma-mainnet
        type: evm
        endpoint: YOUR_CHAINSTACK_PLASMA_MAINNET_ENDPOINT
        evm:
          chainId: 9745
        rateLimitBudget: chainstack-budget
        # ... failsafe config

      - id: plasma-public-mainnet
        type: evm
        endpoint: https://rpc.plasma.to
        evm:
          chainId: 9745
        rateLimitBudget: public-budget
        # ... failsafe config

      # Testnet upstreams
      - id: chainstack-plasma-testnet
        type: evm
        endpoint: YOUR_CHAINSTACK_PLASMA_TESTNET_ENDPOINT
        evm:
          chainId: 9746
        rateLimitBudget: chainstack-budget
        # ... failsafe config

      - id: plasma-public-testnet
        type: evm
        endpoint: https://rpc-testnet.plasma.to
        evm:
          chainId: 9746
        rateLimitBudget: public-budget
        # ... failsafe config

Access testnet at http://localhost:4000/plasma/evm/9746.

How failover works

eRPC handles failures automatically:

1. Primary attempt — Request goes to Chainstack first
2. Timeout/error — If Chainstack fails or times out, eRPC retries based on failsafe config
3. Fallback — After exhausting retries, request routes to the public endpoint
4. Hedging — For latency-sensitive calls, eRPC can send parallel requests to multiple upstreams

The hedge configuration sends a backup request after 2 seconds if the primary hasn’t responded, using whichever completes first.

Production deployment

For production, consider:

• Environment variables — Use ${ENV_VAR} syntax in eRPC config
• Docker Compose — Run eRPC alongside your application
• Health checks — Monitor eRPC metrics on port 4001
• Multiple Chainstack nodes — Add nodes in different regions for geographic redundancy

Example with environment variables:

upstreams:
  - id: chainstack-plasma
    endpoint: ${CHAINSTACK_PLASMA_URL}

Run with:

docker run --rm \
  -e CHAINSTACK_PLASMA_URL="https://your-endpoint..." \
  -v $(pwd)/erpc.yaml:/erpc.yaml \
  -p 4000:4000 \
  ghcr.io/erpc/erpc:latest

Conclusion

You now have fault-tolerant RPC access to Plasma. When your Chainstack endpoint experiences issues, eRPC automatically fails over to public endpoints, ensuring your application stays connected.

For advanced configuration options, see the eRPC documentation.