Which flag ensures the producer waits for all in-sync replicas to acknowledge messages?

• A. --bootstrap-servers
• B. --producer-property acks=all
• C. --producer-property acks=1
• D. --producer-property acks=leader

Answer: (B)

Durable delivery across the cluster is controlled by how the producer waits for acknowledgments. The flag that makes the producer wait for all in-sync replicas to acknowledge a message is the acks setting configured to all. When you set acks=all, the leader writes the record and waits for acknowledgments from every in-sync replica before signaling success to the producer. In-sync replicas are the replicas that are current with the leader; if a replica is lagging or down, it isn’t counted in this acknowledgment process. This provides the strongest durability guarantee because the data is replicated to all up-to-date copies before the write is considered complete, though it can add latency since the producer must wait for multiple replicas. The other options don’t require acknowledgments from all in-sync replicas: connecting to the cluster (bootstrap-servers) just establishes contact; waiting for only the leader (acks=1) or equivalent (acks=leader) means a write is confirmed even if some replicas haven’t persisted it yet.