Disaster recovery planning in Kafka - Mini Project: Build & Apply

Disaster Recovery Planning with Kafka

📖 Scenario: You work for a company that uses Apache Kafka to handle real-time data streams. To keep the system safe from failures, you need to plan how to recover data if something goes wrong.This project will guide you through creating a simple Kafka topic setup with replication and configuring a consumer group to read messages safely, which are key parts of disaster recovery planning.

🎯 Goal: Build a basic Kafka setup with a topic that has replication enabled and a consumer group that reads messages reliably. This setup helps ensure data is not lost and can be recovered if a server fails.

📋 What You'll Learn

Create a Kafka topic named orders with 3 partitions and replication factor 2

Set a configuration variable consumer_group with the value order_processors

Write a Kafka consumer that subscribes to the orders topic using the consumer_group

Print the consumed messages to the console

💡 Why This Matters

🌍 Real World

Kafka is widely used in industries like finance, retail, and tech to process data streams reliably. Disaster recovery planning ensures data is safe and systems can recover quickly from failures.

💼 Career

Understanding Kafka topic replication and consumer groups is essential for roles like data engineer, site reliability engineer, and backend developer working with real-time data pipelines.

1

Create Kafka topic with replication

Create a Kafka topic named orders with 3 partitions and a replication factor of 2 using the kafka-topics.sh command.

Kafka

# Use kafka-topics.sh to create the topic
# Your code here

Hint

Use the kafka-topics.sh tool with the --create option and specify the topic name, partitions, and replication factor.

2

Set consumer group variable

Create a variable called consumer_group and set it to the string order_processors.

Kafka

kafka-topics.sh --create --topic orders --partitions 3 --replication-factor 2 --bootstrap-server localhost:9092

# Create a variable called consumer_group with value 'order_processors'
# Your code here

Hint

Assign the string 'order_processors' to the variable consumer_group.

3

Write Kafka consumer subscribing to topic

Write a Kafka consumer in Python that subscribes to the orders topic using the consumer_group variable. Use the KafkaConsumer class from the kafka library and set the group_id to consumer_group.

Kafka

kafka-topics.sh --create --topic orders --partitions 3 --replication-factor 2 --bootstrap-server localhost:9092

consumer_group = 'order_processors'

# Import KafkaConsumer and create a consumer subscribing to 'orders' with group_id consumer_group
# Your code here

Hint

Import KafkaConsumer and create a consumer subscribing to the orders topic with the group ID set to consumer_group.

4

Print consumed messages

Write a for loop to read messages from the consumer and print each message's value decoded as UTF-8.

Kafka

kafka-topics.sh --create --topic orders --partitions 3 --replication-factor 2 --bootstrap-server localhost:9092

consumer_group = 'order_processors'

from kafka import KafkaConsumer
consumer = KafkaConsumer('orders', group_id=consumer_group, bootstrap_servers=['localhost:9092'])

# Loop over consumer and print each message value decoded as UTF-8
# Your code here

Hint

Use a for loop to read messages from the consumer and print the decoded message values.

Practice

1. What is the main purpose of disaster recovery planning in Kafka?

easy

A. To keep Kafka data safe and services running after failures

B. To improve Kafka message throughput

C. To reduce Kafka cluster size

D. To speed up Kafka consumer processing

Solution

1. Step 1: Understand disaster recovery goal

   Disaster recovery planning focuses on protecting data and maintaining service availability during failures.

2. Step 2: Match goal to options

   Only To keep Kafka data safe and services running after failures mentions keeping data safe and services running after problems, which matches the goal.

3. Final Answer:

   To keep Kafka data safe and services running after failures -> Option A

4. Quick Check:

   Disaster recovery = data safety and uptime [OK]

Hint: Disaster recovery means data safety and uptime [OK]

Common Mistakes:

• Confusing disaster recovery with performance tuning
• Thinking it reduces cluster size
• Assuming it speeds up consumers

2. Which Kafka feature is used to replicate data to a backup cluster for disaster recovery?

easy

A. Kafka Connect

B. Kafka Streams

C. MirrorMaker

D. Kafka Producer

Solution

1. Step 1: Identify replication tool for backup clusters

   MirrorMaker is designed to copy data between Kafka clusters for backup and disaster recovery.

2. Step 2: Compare options

   Kafka Connect is for data integration, Kafka Streams for processing, and Producer for sending data, so only MirrorMaker fits replication to backup clusters.

3. Final Answer:

   MirrorMaker -> Option C

4. Quick Check:

   Replication to backup = MirrorMaker [OK]

Hint: MirrorMaker copies data between clusters [OK]

Common Mistakes:

• Choosing Kafka Connect which is for integration, not replication
• Confusing Kafka Streams with replication
• Thinking Kafka Producer replicates data

3. Given the following MirrorMaker command, what is its purpose?

bin/kafka-mirror-maker.sh --consumer.config consumer.properties --producer.config producer.properties --whitelist='topic1,topic2'

medium

A. Consume messages from topic1 and topic2 without replication

B. Delete topic1 and topic2 from the target cluster

C. Produce messages to topic1 and topic2 only on source cluster

D. Replicate only topic1 and topic2 from source to target cluster

Solution

1. Step 1: Understand MirrorMaker command components

   The command uses consumer and producer configs and a whitelist of topics to replicate.

2. Step 2: Interpret whitelist usage

   The whitelist='topic1,topic2' means only these topics are replicated from source to target cluster.

3. Final Answer:

   Replicate only topic1 and topic2 from source to target cluster -> Option D

4. Quick Check:

   Whitelist limits topics replicated [OK]

Hint: Whitelist limits topics MirrorMaker replicates [OK]

Common Mistakes:

• Thinking it only consumes without replication
• Assuming it produces only on source cluster
• Believing it deletes topics

4. You wrote a MirrorMaker config but replication fails. Which is a likely cause?

bootstrap.servers=source:9092
group.id=mirror-group
whitelist=topicA,topicB

What is wrong?

medium

A. The whitelist should be 'whitelist='topicA,topicB' with quotes

B. Missing producer bootstrap.servers config for target cluster

C. group.id should be 'producer-group' instead of 'mirror-group'

D. bootstrap.servers should point to target cluster, not source

Solution

1. Step 1: Check MirrorMaker config requirements

   MirrorMaker needs both consumer (source) and producer (target) bootstrap.servers configs to replicate data.

2. Step 2: Identify missing config

   The given config only shows source bootstrap.servers; missing producer config causes replication failure.

3. Final Answer:

   Missing producer bootstrap.servers config for target cluster -> Option B

4. Quick Check:

   Both consumer and producer configs needed [OK]

Hint: Both consumer and producer configs are required [OK]

Common Mistakes:

• Thinking quotes are mandatory around whitelist
• Confusing group.id with producer group
• Setting bootstrap.servers to target cluster only

5. You want to ensure minimal data loss during a disaster. Which combination is best for Kafka disaster recovery?

hard

A. Use topic replication factor 3 and MirrorMaker to backup cluster

B. Use only MirrorMaker without replication factor increase

C. Use replication factor 1 and rely on manual backups

D. Use Kafka Streams to replicate data to backup cluster

Solution

1. Step 1: Understand replication factor role

   A replication factor of 3 means data is copied to 3 brokers, reducing data loss risk.

2. Step 2: Combine with MirrorMaker backup

   MirrorMaker copies data to a separate cluster, adding disaster recovery protection.

3. Final Answer:

   Use topic replication factor 3 and MirrorMaker to backup cluster -> Option A

4. Quick Check:

   Replication + MirrorMaker = minimal data loss [OK]

Hint: Combine replication factor and MirrorMaker for best safety [OK]

Common Mistakes:

• Relying only on MirrorMaker without replication
• Using replication factor 1 which risks data loss
• Confusing Kafka Streams with replication tool