How do you implement a read replica strategy in Node.js?

Question

Accepted Answer

**Read replicas** — distribute read queries across multiple DB instances.

```javascript
const { Pool } = require("pg");

const primary = new Pool({ host: "primary.db" });
const replicas = [
    new Pool({ host: "replica1.db" }),
    new Pool({ host: "replica2.db" }),
];

let replicaIndex = 0;

function getReplica() {
    const replica = replicas[replicaIndex % replicas.length];
    replicaIndex++;
    return replica;
}

async function query(sql, params, options = {}) {
    const { write = false } = options;

if (write) {
        return primary.query(sql, params);
    }

// Reads with retry on replica failure
    try {
        return await getReplica().query(sql, params);
    } catch (err) {
        console.error("Replica failed, falling back to primary");
        return primary.query(sql, params);
    }
}

// Usage
const users = await query("SELECT * FROM users"); // goes to replica
await query("INSERT INTO users ...", [], { write: true }); // primary
```

**Key points:**
- Route reads to replicas, writes to primary.
- Handle replica lag (read-your-writes).
- Fallback to primary on failure.

Node.js: How do you implement a read replica strategy in Node.js?