MongoDB cheatsheet

mongosh — connect

Command	Description
`mongosh "mongodb://localhost:27017"`	Local connect
`mongosh "mongodb://user:pass@host:27017/mydb"`	Connect with auth
`mongosh "mongodb://h1,h2,h3/mydb?replicaSet=rs0"`	Replica set connection
`show dbs`	List databases
`use mydb`	Switch database
`show collections`	List collections

Service and health

Command	Description
`systemctl status mongod`	Service status
`ss -tlnp \| grep 27017`	Confirm listener
`db.serverStatus()`	Uptime, connections, opcounters
`db.stats()`	Database size and counts
`db.currentOp()`	Running operations

CRUD and queries

Command	Description
`db.users.findOne({ name: "alice" })`	Find one document
`db.users.find({ status: "active" }).limit(10)`	Filtered query
`db.users.insertOne({ name: "bob" })`	Insert document
`db.users.updateOne({ _id: id }, { $set: { x: 1 } })`	Update document
`db.users.deleteMany({ status: "inactive" })`	Delete matching docs
`db.users.countDocuments({ status: "active" })`	Count documents

Indexes and explain

Command	Description
`db.users.getIndexes()`	List indexes
`db.users.createIndex({ email: 1 }, { unique: true })`	Create index
`db.users.explain("executionStats").find({ email: "a@b.com" })`	Query plan with stats

Replica set

Command	Description
`rs.status()`	Member states, health, optime
`rs.conf()`	Replica set configuration
`rs.initiate()`	Initialize new replica set
`rs.add("host2:27017")`	Add member
`rs.stepDown()`	Primary steps down (election)
`rs.printSecondaryReplicationInfo()`	Lag per secondary
`db.hello()`	Am I primary? set name, me

Replication lag check

// On primary — oplog window
rs.printReplicationInfo()

// On primary — lag per secondary
rs.printSecondaryReplicationInfo()

// Member state: PRIMARY, SECONDARY, RECOVERING, etc.
rs.status().members.forEach(m => print(m.name, m.stateStr, m.health))

Users and auth

use admin
db.createUser({
  user: "appuser",
  pwd: "secret",
  roles: [{ role: "readWrite", db: "mydb" }]
})

db.getUsers()
db.auth("appuser", "secret")

Backup and restore

Command	Description
`mongodump --uri="mongodb://host/mydb" --out=/backup/`	Logical backup
`mongodump --oplog --out=/backup/`	Dump + oplog (replica set)
`mongorestore --uri="mongodb://host/mydb" /backup/mydb`	Restore database
`db.fsyncLock()` / `db.fsyncUnlock()`	Pause writes for snapshot (secondary)

Pro tips

Always use a replica set connection string in production — not a single host
Writes must target the PRIMARY — not primary errors mean failover or wrong read preference
Use explain("executionStats") — COLLSCAN on large collections needs an index
Three voting members minimum for reliable elections (or 2 data + 1 arbiter)
Secondaries are not backups — schedule mongodump; test restores

Practice scenarios

Hands-on MongoDB scenarios on live Linux VMs: mongodb