How to Set Up a MongoDB Replica Set on CentOS – Step‑by‑Step Guide

Introduction

MongoDB is a C++‑written, distributed, NoSQL database that offers scalable high‑performance storage for web applications. This guide focuses on MongoDB replica sets and sharding.

Overview

MongoDB bridges relational and non‑relational databases, supporting flexible data structures, powerful query language, indexing, replication, and automatic sharding. It stores data in BSON (an extended JSON) and provides drivers for many languages.

Key Features

High performance, easy deployment and use

Document‑oriented storage, schema‑free

Dynamic queries and full indexing, including embedded objects

Replication and automatic failover

Binary storage for large objects (e.g., videos)

Automatic sharding for cloud‑scale expansion

Multi‑language support: Ruby, Python, Java, C++, PHP, etc.

Advantages and Disadvantages

Advantages

Eventual consistency for faster reads

Document model simplifies data access

Built‑in GridFS for large file storage

Native sharding support

Rich third‑party ecosystem

Strong performance

Disadvantages

Limited transaction support

Higher storage footprint

Immature maintenance tools

Replica Set Types

MongoDB offers Master/Slave (deprecated) and Replica Set replication. The modern approach uses Replica Sets, which require at least three odd‑numbered nodes and may include arbiter nodes for elections.

Replica Set Architecture

Configuration Steps

1. Install Required Packages

# cd mongodb/
# yum install *.rpm -y   # install three RPM packages

Repeat on all nodes.

2. Edit the Configuration File

# vim /etc/mongod.conf
logpath=/var/log/mongodb/mongod.log
logappend=true
fork=true
dbpath=/mongodb/data   # data location
pidfilepath=/var/run/mongodb/mongod.pid
#bind_ip=127.0.0.1   # comment to listen on all interfaces
httpinterface=true   # enable web interface
rest=true
replSet=testSet      # replica set name
replIndexPrefetch=_id_only

3. Distribute the Config File

# scp /etc/mongod.conf node3:/etc
# scp /etc/mongod.conf node4:/etc

4. Create Data Directories

# mkdir -p /mongodb/data
# chown -R mongod:mongod /mongodb

5. Start the Service

# service mongod start
# ss -tnl | grep 27017   # verify listening on port 27017

Initialize the Replica Set

# mongo
> rs.initiate()
{ "info2" : "no configuration explicitly specified -- making one", "me" : "node1.scholar.com:27017", "info" : "Config now saved locally.  Should come online in about a minute.", "ok" : 1 }
> rs.status()
{ "set" : "testSet", "date" : ISODate("2015-07-13T12:33:27Z"), "myState" : 1, "members" : [ { "_id" : 0, "name" : "node1.scholar.com:27017", "health" : 1, "state" : 1, "stateStr" : "PRIMARY", "uptime" : 1111, "optime" : Timestamp(1436790736,1), "self" : true } ], "ok" : 1 }

The node becomes PRIMARY.

Add Secondary Nodes

> rs.add("172.16.10.125")
{ "ok" : 1 }
> rs.add("172.16.10.126")
{ "ok" : 1 }
> rs.status()
... (output shows PRIMARY and SECONDARY members) ...

Verify Data Synchronization

On the primary:

> use testdb
> db.students.insert({name:"ZhangSan",age:"21"})
WriteResult({ "nInserted" : 1 })

On a secondary (after enabling reads):

> rs.slaveOk()
> use testdb
> db.students.findOne()
{ "_id" : ObjectId("55a3b494ebcafd9edbdfce4d"), "name" : "ZhangSan", "age" : "21" }

Attempting writes on a secondary fails, confirming that only the primary accepts writes.

Conclusion

The replica set is now configured; if the primary fails, an automatic election will promote a secondary to primary.