Change streams on databases in MongoDB - Step-by-Step Execution

Concept Flow - Change streams on databases

Start watching database

↓

Database change occurs

↓

Change stream detects change

↓

Change event sent to client

↓

Client processes event

↓

Continue listening or stop

Change streams let you watch for changes in a MongoDB database and get notified when something changes.

Execution Sample

MongoDB

const changeStream = db.watch();
changeStream.on('change', (change) => {
  console.log(change);
});

This code listens for any change in the database and prints the change details when it happens.

Execution Table

Step	Action	Change Detected?	Event Sent	Client Output
1	Start watching database with db.watch()	No	No	No output
2	Insert a document into a collection	Yes	Yes	{"operationType":"insert", "ns":{"db":"test", "coll":"users"}, "fullDocument":{...}}
3	Update a document in the collection	Yes	Yes	{"operationType":"update", "ns":{"db":"test", "coll":"users"}, "updateDescription":{...}}
4	Delete a document from the collection	Yes	Yes	{"operationType":"delete", "ns":{"db":"test", "coll":"users"}, "documentKey":{...}}
5	No further changes	No	No	No output

💡 Listening stops when client closes the change stream or application ends.

Variable Tracker

Variable	Start	After Insert	After Update	After Delete	Final
changeStream.isOpen	true	true	true	true	false
changeStream.latestEvent	null	{"operationType":"insert"}	{"operationType":"update"}	{"operationType":"delete"}	null

Key Moments - 3 Insights

Why does the change stream not output anything before any database change?

Because the change stream only sends events when a change happens, as shown in execution_table step 1 where no change means no output.

What happens if multiple changes occur quickly?

Each change triggers a separate event sent to the client, so the client receives multiple change events in order, as seen in steps 2, 3, and 4.

Does the change stream keep running forever?

No, it runs until the client closes it or the application stops, as noted in the exit_note.

Visual Quiz - 3 Questions

Look at the execution_table, what is the client output at step 3?

A No output

B {"operationType":"update", "ns":{"db":"test", "coll":"users"}, "updateDescription":{...}}

C {"operationType":"insert", "ns":{"db":"test", "coll":"users"}, "fullDocument":{...}}

D {"operationType":"delete", "ns":{"db":"test", "coll":"users"}, "documentKey":{...}}

💡 Hint

Check the 'Client Output' column at step 3 in the execution_table.

According to variable_tracker, what is the value of changeStream.isOpen after the delete operation?

A false

B null

C true

D undefined

💡 Hint

Look at the 'After Delete' column for changeStream.isOpen in variable_tracker.

If no changes happen after starting the change stream, what will the client see?

A No output

B Continuous change events

C An error message

D A snapshot of the database

💡 Hint

Refer to step 1 and 5 in execution_table where no changes mean no output.

Concept Snapshot

Change streams let you watch MongoDB databases for changes.
Use db.watch() to start listening.
Events like insert, update, delete trigger notifications.
Client receives change events in real time.
Stop listening by closing the stream.

Full Transcript

Change streams in MongoDB allow applications to listen for changes in the database. When you start a change stream using db.watch(), it waits for any changes like inserts, updates, or deletes. When a change happens, the stream sends an event to the client with details about the change. The client can then process this event, for example, by logging it or updating a UI. The stream keeps running until the client closes it or the application ends. If no changes happen, no events are sent. This lets applications react immediately to database changes without polling.

Practice

1. What is the main purpose of MongoDB change streams on databases?

easy

A. To listen for real-time changes in the database

B. To backup the database automatically

C. To create indexes on collections

D. To run queries faster

Solution

1. Step 1: Understand what change streams do

   Change streams allow applications to watch for changes like inserts, updates, or deletes in real-time.

2. Step 2: Compare options with this purpose

   Only To listen for real-time changes in the database describes listening to real-time changes, which matches the main purpose of change streams.

3. Final Answer:

   To listen for real-time changes in the database -> Option A

4. Quick Check:

   Change streams = real-time listening [OK]

Hint: Change streams = real-time database updates listener [OK]

Common Mistakes:

• Confusing change streams with backups
• Thinking change streams create indexes
• Assuming change streams speed up queries

2. Which of the following is the correct syntax to start watching changes on a MongoDB database named mydb?

easy

A. watch.db('mydb')

B. db.watch('mydb')

C. db.mydb.watch()

D. db.getSiblingDB('mydb').watch()

Solution

1. Step 1: Recall how to access a database in MongoDB shell

   To access a database other than the current one, use db.getSiblingDB('mydb').

2. Step 2: Apply the watch() method on the database object

   The watch() method is called on the database object to start change streams.

3. Final Answer:

   db.getSiblingDB('mydb').watch() -> Option D

4. Quick Check:

   Use getSiblingDB to access db, then watch() [OK]

Hint: Use db.getSiblingDB('name').watch() to watch a database [OK]

Common Mistakes:

• Trying to call watch() directly on db with database name
• Using incorrect method like db.watch('name')
• Confusing method order or syntax

3. Given the following code snippet, what will be the output if a document is inserted into any collection in the testdb database?

const changeStream = db.getSiblingDB('testdb').watch();
changeStream.forEach((change) => {
  printjson(change.operationType);
});

medium

A. "update"

B. "delete"

C. "insert"

D. No output

Solution

1. Step 1: Understand what triggers the change event

   The change stream listens to all changes in the database, including inserts, updates, and deletes.

2. Step 2: Identify the operationType for an insert

   When a document is inserted, the operationType is "insert".

3. Final Answer:

   "insert" -> Option C

4. Quick Check:

   Insert triggers operationType = "insert" [OK]

Hint: Insert triggers operationType 'insert' in change stream [OK]

Common Mistakes:

• Expecting 'update' or 'delete' for insert events
• Thinking no output occurs without filters
• Confusing event names with method names

4. Identify the error in this code snippet that tries to watch changes on the sales database:

const changeStream = db.sales.watch();
changeStream.forEach((change) => {
  printjson(change);
});

medium

A. The watch() method cannot be called on a database object

B. The database should be accessed using getSiblingDB('sales') before calling watch()

C. The on() method is not valid for change streams

D. Missing parentheses after watch

Solution

1. Step 1: Check how the database is accessed

   Using db.sales accesses a collection named 'sales' in the current database, not the database itself.

2. Step 2: Correct way to access a different database

   To watch changes on the 'sales' database, use db.getSiblingDB('sales').watch().

3. Final Answer:

   The database should be accessed using getSiblingDB('sales') before calling watch() -> Option B

4. Quick Check:

   Use getSiblingDB to access database, not db.sales [OK]

Hint: Use getSiblingDB('name') to access database, not db.name [OK]

Common Mistakes:

• Assuming db.sales is a database instead of a collection
• Thinking watch() can't be called on database objects
• Confusing event handling methods

5. You want to watch only update operations on the orders database using change streams. Which code snippet correctly applies a filter to achieve this?

hard

A. db.getSiblingDB('orders').watch([{ $match: { operationType: 'update' } }])

B. db.orders.watch({ operationType: 'update' })

C. db.getSiblingDB('orders').watch({ filter: { operationType: 'update' } })

D. db.getSiblingDB('orders').watch([{ $filter: { operationType: 'update' } }])

Solution

1. Step 1: Recall how to filter change streams

   Change streams accept an aggregation pipeline array to filter events, using stages like $match.

2. Step 2: Apply correct syntax for filtering updates

   The correct filter is [{ $match: { operationType: 'update' } }] passed as an array to watch().

3. Final Answer:

   db.getSiblingDB('orders').watch([{ $match: { operationType: 'update' } }]) -> Option A

4. Quick Check:

   Use $match in array to filter change streams [OK]

Hint: Use watch([{ $match: { operationType: 'update' } }]) to filter updates [OK]

Common Mistakes:

• Passing filter as object instead of pipeline array
• Using invalid keys like 'filter' or '$filter'
• Calling watch() on collection instead of database