Replication factor in Kafka - Commands & Configuration

Introduction

When you run Kafka, you want your messages to be safe even if some servers fail. Replication factor tells Kafka how many copies of your data it should keep on different servers. This helps keep your data safe and available.

When you want to make sure your messages are not lost if one Kafka server crashes

When you run Kafka on multiple servers and want to spread copies of data for safety

When you want to balance between data safety and storage space used

When you want to improve Kafka availability during maintenance or unexpected failures

When you want to control how many servers share the same data for fault tolerance

Commands

This command creates a new Kafka topic named 'example-topic' with 3 partitions and a replication factor of 2. This means each partition will have 2 copies on different Kafka servers for safety.

Terminal

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

Expected Output

Created topic example-topic.

→

--replication-factor - Sets how many copies of each partition Kafka keeps on different servers

→

--partitions - Defines how many parts the topic is split into for parallel processing

This command shows details about the 'example-topic', including its partitions and replication factor. It helps verify that the topic was created with the right replication settings.

Terminal

kafka-topics.sh --describe --topic example-topic --bootstrap-server localhost:9092

Expected Output

Topic: example-topic	PartitionCount: 3	ReplicationFactor: 2	Configs: 

	Topic: example-topic	Partition: 0	Leader: 1	Replicas: 1,2	Isr: 1,2

	Topic: example-topic	Partition: 1	Leader: 2	Replicas: 2,3	Isr: 2,3

	Topic: example-topic	Partition: 2	Leader: 3	Replicas: 3,1	Isr: 3,1

Key Concept

If you remember nothing else from this pattern, remember: replication factor controls how many copies of your Kafka data exist to keep it safe from server failures.

Common Mistakes

Setting replication factor higher than the number of Kafka brokers

Kafka cannot create more copies than the number of servers available, so the topic creation will fail.

Set the replication factor to a number less than or equal to the number of Kafka brokers in your cluster.

Using replication factor of 1 in production

With only one copy, if the server holding the data fails, the data is lost and unavailable.

Use a replication factor of at least 2 or 3 to ensure data safety and availability.

Summary

Create a Kafka topic with the --replication-factor flag to set how many copies of data Kafka keeps.

Use kafka-topics.sh --describe to check the topic's replication factor and partitions.

Replication factor helps keep your data safe by storing copies on multiple servers.

Practice

1.

What does the replication factor in Kafka control?

easy

A. The size limit of each partition

B. The number of copies of each partition stored across brokers

C. The number of partitions in a topic

D. The number of consumers in a consumer group

Solution

1. Step 1: Understand replication factor meaning

   The replication factor defines how many copies of each partition Kafka keeps on different brokers.

2. Step 2: Differentiate from other Kafka settings

   Partitions count and consumer group size are unrelated to replication factor.

3. Final Answer:

   The number of copies of each partition stored across brokers -> Option B

4. Quick Check:

   Replication factor = number of copies [OK]

Hint: Replication factor = copies of data stored [OK]

Common Mistakes:

• Confusing replication factor with partition count
• Thinking replication factor controls consumer groups
• Assuming replication factor limits partition size

2.

Which of the following is the correct way to set the replication factor when creating a Kafka topic using the command line?

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

easy

A. --replication-factor 3

B. --replication-factor=3

C. --replication_factor 3

D. --replicationFactor 3

Solution

1. Step 1: Check Kafka CLI syntax for replication factor

   The correct flag is --replication-factor with a space before the value.

2. Step 2: Identify incorrect options

   Options with underscores or camelCase are invalid flags; equals sign is not used here.

3. Final Answer:

   --replication-factor 3 -> Option A

4. Quick Check:

   Use double dash and hyphen for replication factor [OK]

Hint: Use double dash and hyphen for flags in Kafka CLI [OK]

Common Mistakes:

• Using underscore instead of hyphen
• Using equals sign instead of space
• Using camelCase flag names

3.

Given a Kafka topic created with --partitions 2 and --replication-factor 3, how many total replicas will Kafka store across the cluster?

medium

A. 6

B. 3

C. 5

D. 2

Solution

1. Step 1: Calculate replicas per partition

   Each partition has 3 replicas because replication factor is 3.

2. Step 2: Multiply by number of partitions

   2 partitions x 3 replicas each = 6 total replicas.

3. Final Answer:

   6 -> Option A

4. Quick Check:

   Partitions x Replication factor = Total replicas [OK]

Hint: Multiply partitions by replication factor [OK]

Common Mistakes:

• Confusing replication factor with total partitions
• Adding instead of multiplying partitions and replicas
• Ignoring replication factor in calculation

4.

What is wrong with this Kafka topic creation command?

kafka-topics.sh --create --topic logs --partitions 3 --replication-factor 5 --bootstrap-server localhost:9092

Assuming the Kafka cluster has only 3 brokers.

medium

A. Partitions must be equal to replication factor

B. Topic name cannot be 'logs'

C. Bootstrap server address is incorrect

D. Replication factor cannot be greater than the number of brokers

Solution

1. Step 1: Understand replication factor limits

   Replication factor must be less than or equal to the number of brokers to ensure each replica is on a different broker.

2. Step 2: Check cluster size vs replication factor

   Replication factor 5 is greater than 3 brokers, which is invalid.

3. Final Answer:

   Replication factor cannot be greater than the number of brokers -> Option D

4. Quick Check:

   Replication factor ≤ brokers [OK]

Hint: Replication factor must not exceed broker count [OK]

Common Mistakes:

• Thinking partitions must equal replication factor
• Ignoring cluster size limits
• Assuming topic names have restrictions here

5.

You want to create a Kafka topic with 4 partitions and a replication factor of 3 for high availability. However, your cluster has only 2 brokers. What is the best solution?

hard

A. Create the topic with 1 partition and replication factor 3

B. Set replication factor to 4 to match partitions

C. Increase the number of brokers to at least 3

D. Create the topic with replication factor 2 instead

Solution

1. Step 1: Check replication factor vs brokers

   Replication factor 3 requires at least 3 brokers to store replicas on different machines.

2. Step 2: Evaluate options

   Reducing replication factor lowers availability; increasing brokers meets requirement.

3. Final Answer:

   Increase the number of brokers to at least 3 -> Option C

4. Quick Check:

   Replication factor ≤ brokers for high availability [OK]

Hint: Match replication factor to broker count for safety [OK]

Common Mistakes:

• Setting replication factor higher than brokers
• Reducing partitions instead of brokers
• Ignoring availability requirements