Which combination uniquely identifies a consumed message in Kafka?

Which combination uniquely identifies a consumed message in Kafka?

• A. Topic and Offset
• B. Topic and Partition
• C. Topic, Partition, and Offset
• D. Partition and Offset

Answer: (C)

In Kafka, a message is uniquely identified by the topic it belongs to, the partition it sits in, and its offset within that partition. The reason is that partitions are the separate logs inside a topic, and each message within a partition gets its own increasing offset. Offsets are only meaningful inside a single partition, so the same offset value can appear in different partitions or even across different topics. To locate one exact record anywhere in the system, you need all three pieces: the topic to specify which log you’re looking at, the partition to choose the exact shard of that log, and the offset to pinpoint the exact message within that shard. If you omit the topic, you don’t know which log to search; if you omit the partition, you can’t identify which shard within the topic; if you omit the offset, you can’t specify the exact message within the chosen partition. That’s why the full combination of topic, partition, and offset uniquely identifies a consumed message.

In Kafka, a message is uniquely identified by the topic it belongs to, the partition it sits in, and its offset within that partition. The reason is that partitions are the separate logs inside a topic, and each message within a partition gets its own increasing offset. Offsets are only meaningful inside a single partition, so the same offset value can appear in different partitions or even across different topics. To locate one exact record anywhere in the system, you need all three pieces: the topic to specify which log you’re looking at, the partition to choose the exact shard of that log, and the offset to pinpoint the exact message within that shard. If you omit the topic, you don’t know which log to search; if you omit the partition, you can’t identify which shard within the topic; if you omit the offset, you can’t specify the exact message within the chosen partition. That’s why the full combination of topic, partition, and offset uniquely identifies a consumed message.