Mongos router behavior in MongoDB

Introduction

Mongos acts like a traffic controller that directs database requests to the right place in a sharded MongoDB setup.

When your database is split into parts (shards) to handle lots of data.

When you want to make sure queries go to the correct shard automatically.

When you need to balance the load across multiple servers.

When you want to hide the complexity of multiple shards from your application.

When you want to scale your database horizontally.

Syntax

MongoDB

mongos --configdb <configReplSet>/<host1>:<port1>,<host2>:<port2>,...

// Connect to mongos like a normal MongoDB client
mongo --host <mongos-host> --port <mongos-port>

Mongos is not a database server but a router for sharded clusters.

You connect your application to mongos, not directly to shards.

Examples

Starts a mongos instance connected to the config server replica set.

MongoDB

mongos --configdb configReplSet/localhost:27019,localhost:27020,localhost:27021

Connects to mongos on port 27017 to run queries.

MongoDB

mongo --host localhost --port 27017

Sample Program

This example shows inserting and querying data through mongos. Mongos routes the insert and find commands to the correct shard automatically.

MongoDB

// Assume mongos is running on localhost:27017
// Insert a document into a sharded collection
use shop

// Insert a product
 db.products.insertOne({ _id: 1, name: "Pen", category: "Stationery" })

// Query the product
 db.products.find({ _id: 1 })

Important Notes

Mongos does not store data itself; it only routes requests.

It uses metadata from config servers to know where data lives.

If mongos is down, your application cannot access the sharded cluster.

Summary

Mongos routes queries to the right shard in a sharded MongoDB cluster.

Applications connect to mongos, not directly to shards.

Mongos helps scale and balance database load transparently.