Which control ensures consumers do not read unreplicated data?

Which control ensures consumers do not read unreplicated data?

• A. High watermark
• B. Consumer group id
• C. Acks
• D. Retention policy

Answer: (A)

In Kafka, data safety for consumers is bounded by the high watermark. The high watermark marks the highest offset that has been replicated to all in-sync replicas, so that data up to this point is considered committed and safe to read. Consumers are allowed to read only up to this boundary, which means they don’t read messages that might be lost if a follower hasn’t yet caught up or if a broker fails. That’s why this control prevents reading unreplicated data. The other options don’t provide this protection: acks is about producer durability, retention policy controls how long data is kept, and the consumer group id just handles how consumers coordinate.

In Kafka, data safety for consumers is bounded by the high watermark. The high watermark marks the highest offset that has been replicated to all in-sync replicas, so that data up to this point is considered committed and safe to read. Consumers are allowed to read only up to this boundary, which means they don’t read messages that might be lost if a follower hasn’t yet caught up or if a broker fails. That’s why this control prevents reading unreplicated data. The other options don’t provide this protection: acks is about producer durability, retention policy controls how long data is kept, and the consumer group id just handles how consumers coordinate.