Redis as message broker in Flask - Step-by-Step Execution

Concept Flow - Redis as message broker

Producer sends message

↓

Message stored in Redis list

↓

Consumer polls Redis list

↓

Consumer processes message

↓

Consumer waits for next message

This flow shows how a producer sends messages to Redis, which stores them in a list, and a consumer retrieves and processes them.

Execution Sample

Flask

from flask import Flask
import redis

r = redis.Redis()

# Producer
r.lpush('queue', 'hello')

# Consumer
msg = r.brpop('queue')

This code pushes a message 'hello' to a Redis list 'queue' and then a consumer blocks waiting to pop a message from the same list.

Execution Table

Step	Action	Redis Command	Result	State of 'queue'
1	Producer pushes 'hello'	LPUSH queue 'hello'	1 (list length)	['hello']
2	Consumer blocks waiting	BRPOP queue	Returns ('queue', 'hello')	[] (empty)
3	Consumer processes message	N/A	Message 'hello' processed	[]
4	Consumer waits for next message	BRPOP queue	Blocks waiting	[]

💡 Consumer blocks waiting because queue is empty after popping 'hello'

Variable Tracker

Variable	Start	After Step 1	After Step 2	After Step 3	After Step 4
queue	[]	['hello']	[]	[]	[]
msg	None	None	('queue', 'hello')	('queue', 'hello')	None (waiting)

Key Moments - 2 Insights

Why does the consumer block at step 4?

At step 4, the queue is empty (see execution_table row 4), so BRPOP blocks waiting for new messages to arrive.

What happens to the message after the consumer pops it?

After popping at step 2, the message 'hello' is removed from the queue (execution_table row 2), so the queue becomes empty.

Visual Quiz - 3 Questions

Look at the execution table, what is the state of 'queue' after step 1?

A ['hello']

B []

C ['world']

D None

💡 Hint

Check the 'State of queue' column in execution_table row 1

At which step does the consumer receive the message 'hello'?

A Step 1

B Step 2

C Step 3

D Step 4

💡 Hint

Look at the 'Result' column for when BRPOP returns 'hello' in execution_table

If the producer pushes another message after step 4, what happens to the consumer?

A Consumer crashes

B Consumer ignores the new message

C Consumer immediately receives the new message

D Consumer waits indefinitely

💡 Hint

BRPOP blocks waiting at step 4, so new messages unblock it (see concept_flow)

Concept Snapshot

Redis as message broker uses lists to queue messages.
Producer pushes messages with LPUSH.
Consumer blocks and pops messages with BRPOP.
BRPOP waits if queue is empty.
This enables simple, reliable message passing.

Full Transcript

This visual trace shows how Redis acts as a message broker in a Flask app. The producer uses LPUSH to add messages to a Redis list named 'queue'. The consumer uses BRPOP to block and wait for messages from the same list. When a message is pushed, the consumer unblocks, pops the message, and processes it. If the queue is empty, the consumer waits until a new message arrives. This pattern enables asynchronous communication between parts of an app using Redis as a simple message queue.

Practice

1. What is the main role of Redis when used as a message broker in a Flask application?

easy

A. To serve static files like images and CSS

B. To store user passwords securely

C. To send and receive messages between different parts of the app

D. To handle HTTP requests directly

Solution

1. Step 1: Understand Redis as a message broker

   Redis acts as a middleman to send messages between parts of an app, enabling communication.

2. Step 2: Identify Redis's role in Flask

   It is not used for storing passwords, serving files, or handling HTTP requests directly.

3. Final Answer:

   To send and receive messages between different parts of the app -> Option C

4. Quick Check:

   Redis message broker = send/receive messages [OK]

Hint: Redis in Flask is for messaging, not storage or HTTP [OK]

Common Mistakes:

• Confusing Redis with a database for user data
• Thinking Redis serves static files
• Assuming Redis handles HTTP requests

2. Which of the following is the correct way to publish a message to a Redis channel in Flask?

easy

A. redis_client.post('channel', 'message')

B. redis_client.send('channel', 'message')

C. redis_client.push('channel', 'message')

D. redis_client.publish('channel', 'message')

Solution

1. Step 1: Recall Redis publish method

   Redis uses the method publish to send messages to a channel.

2. Step 2: Check method names

   Only publish is the correct Redis command; others like send, push, or post are incorrect.

3. Final Answer:

   redis_client.publish('channel', 'message') -> Option D

4. Quick Check:

   Publish method = publish() [OK]

Hint: Use publish() to send messages in Redis [OK]

Common Mistakes:

• Using send() instead of publish()
• Confusing push() with publish()
• Trying to use post() which is not a Redis method

3. Given the following Flask code snippet using Redis, what will be printed when a message is published to 'test_channel'?

import redis
r = redis.Redis()

pubsub = r.pubsub()
pubsub.subscribe('test_channel')

for message in pubsub.listen():
    if message['type'] == 'message':
        print(message['data'].decode())
        break

medium

A. The published message content decoded as a string

B. The string 'test_channel'

C. The message type 'subscribe'

D. An error because decode() is not valid here

Solution

1. Step 1: Understand pubsub.listen() behavior

   The loop listens for messages on 'test_channel'. When a message of type 'message' arrives, it prints the decoded data.

2. Step 2: Decode message data

   The message data is bytes, so decode() converts it to a string, which is printed.

3. Final Answer:

   The published message content decoded as a string -> Option A

4. Quick Check:

   Message data decoded = printed string [OK]

Hint: listen() yields messages; decode bytes to string before print [OK]

Common Mistakes:

• Printing channel name instead of message data
• Confusing message type 'subscribe' with 'message'
• Forgetting to decode bytes to string

4. What is the main issue with this Flask Redis subscriber code snippet?

import redis
r = redis.Redis()
pubsub = r.pubsub()
pubsub.subscribe('channel')

for message in pubsub.listen():
    if message['type'] == 'message':
        print(message['data'].decode())

It causes the Flask app to freeze and not respond.

medium

A. The listen() loop blocks the main thread, freezing the app

B. The decode() method is called on a string, causing an error

C. The subscribe() method is missing required parameters

D. Redis client is not connected to the server

Solution

1. Step 1: Analyze listen() behavior

   The listen() method blocks execution, waiting for messages indefinitely.

2. Step 2: Understand Flask app impact

   Running listen() in the main thread stops Flask from handling other requests, freezing the app.

3. Final Answer:

   The listen() loop blocks the main thread, freezing the app -> Option A

4. Quick Check:

   Blocking listen() = app freeze [OK]

Hint: listen() blocks; run it in background thread [OK]

Common Mistakes:

• Thinking decode() causes error here
• Believing subscribe() needs extra parameters
• Assuming Redis connection is missing without error

5. You want to listen for Redis messages in a Flask app without freezing it. Which approach correctly solves this problem?

hard

A. Call pubsub.listen() directly in the main Flask route handler

B. Run the Redis pubsub.listen() loop inside a background thread

C. Use Redis publish() inside the listen() loop

D. Subscribe to Redis channels without calling listen()

Solution

1. Step 1: Identify blocking issue with listen()

   Calling listen() in the main thread blocks Flask, freezing the app.

2. Step 2: Use background thread for listening

   Running listen() in a separate thread allows Flask to stay responsive while listening for messages.

3. Final Answer:

   Run the Redis pubsub.listen() loop inside a background thread -> Option B

4. Quick Check:

   Background thread for listen() = no freeze [OK]

Hint: Use background thread to avoid blocking Flask [OK]

Common Mistakes:

• Running listen() in main route causing freeze
• Misusing publish() inside listen() loop
• Subscribing without listen() so no messages received