Acks and Min.insync.replicas

The video tutorial provides an in-depth explanation of Kafka's acknowledgement settings: X=0, X=1, and X=all. It covers how each setting affects data safety and performance, with X=0 offering no response, X=1 providing leader acknowledgement, and X=all ensuring full replica acknowledgement. The tutorial also discusses the importance of Min Insync Replicas for data safety and how it interacts with these settings to prevent data loss.

What is the main advantage of setting X=0 in Kafka?

• Provides leader acknowledgement

• Guarantees data replication

• Improves performance by not waiting for broker response

• Ensures no data loss

What is a potential risk when using X=1 in Kafka?

• No response from the broker

• High resource consumption

• Increased latency

• Data loss if the leader fails before replication

What is the default acknowledgement setting in Kafka 2.0?

• X=0

• X=1

• X=all

• No default setting

How does setting X=all affect data safety in Kafka?

• It allows data loss if the leader fails

• It ensures no data loss by requiring all replicas to acknowledge

• It reduces latency

• It only requires the leader to acknowledge

What is the role of Min Insync Replicas in Kafka?

• To decrease the replication factor

• To ensure a minimum number of replicas acknowledge data receipt

• To allow more brokers to go offline

• To increase the number of partitions

What happens if the number of Insync Replicas is less than required?

• Data is written without any issues

• The leader automatically retries

• The data is lost permanently

• An exception is thrown indicating not enough replicas

Which setting combination ensures maximum data safety in Kafka?

• X=0 with Min Insync Replicas of 1

• X=all with Min Insync Replicas of 1

• X=1 with Min Insync Replicas of 2

• X=all with Min Insync Replicas of 2