Kafka currently provides at-least-once messaging guarantees. Duplicates can arise due to either producer retries or consumer restarts after failure. One way to provide exactly-once messaging semantics is to implement an idempotent producer. This has been covered at length in the proposal for an Idempotent Producer. An alternative and more general approach is to support transactional messaging. This can enable use-cases such as replicated logging for transactional data services in addition to the classic idempotent producer use-cases.
What is transactional messaging?
Producers can explicitly initiate transactional sessions, send (transactional) messages within those sessions and either commit or abort the transaction. The guarantees that we aim to achieve for transactions are perhaps best understood by enumerating the functional requirements.
- Atomicity: A consumer's application should not be exposed to messages from uncommitted transactions.
- Durability: The broker cannot lose any committed transactions.
- Ordering: A transaction-aware consumer should see transactions in the original transaction-order within each partition.
- Interleaving: Each partition should be able to accept messages from both transactional and non-transactional producers
- There should be no duplicate messages within transactions.
If interleaving of transactional and non-transactional messages is allowed, then the relative ordering of non-transactional and transactional messages will be based on the relative order of append (for non-transactional messages) and final commit (for the transactional messages).
So in the above diagram, partitions p0 and p1 receive messages for transactions X1 and X2, and non-transactional messages as well. The time-line is the time of arrival of the messages to the broker. Since X2 is committed first, each partition will expose messages from X2 before X1. Since the non-transactional messages arrived before the commits for X2 and X2, those messages will be exposed before messages from either transaction.
Furthermore, we have the following implementation requirements:
- The implementation should be scalable. E.g., a dedicated log per transaction is unacceptable.
- Performance:
- The throughput of a transactional producer should be comparable to that of a non-transactional producer.
- Acceptable latency. E.g., avoid copying the transactional data as much as possible.
- Any implementation should not make the partition unavailable (say, due to locking) for an unreasonable period of time.
- Client simplicity: Favor a scheme that lend to a simpler client-side implementation (even if it adds more complexity to the broker). For example, it is acceptable (but not ideal) for a consumer implementation to (internally) buffer and subsequently discard messages from uncommitted transactions. i.e., if the chosen implementation allows the broker to materialize messages from uncommitted transactions in the data logs.
Finally, it is worth adding that any implementation should also provide the ability to associate each transaction's input state with the transaction itself. This is necessary to facilitate retries for transactions - i.e., if a transaction needs to be aborted and retried, then the entire input for that transaction needs to be replayed.
Each transaction is associated with a block of input that is processed and results in the output (transaction). When we commit the transaction we would need to also associate the next block of input with that transaction. In the event of a failure the processor would need to query (the downstream Kafka cluster) to determine the next block that needs to be processed. In our case, this would simply be an input offset vector (IOV) for the input partitions that are being processed for each transaction.
Implementation overview
In this implementation proposal, the producer sends transactional control messages that signal the begin/end/abort state of transactions to a highly-available transaction coordinator which manages transactions using a multi-phase protocol. The producer sends transaction control records (begin/end/abort) to the transaction coordinator, and sends the payload of the transaction directly to the destination data partitions. Consumers need to be transaction-aware and buffer each pending transaction until they reach its corresponding end (commit/abort) record.
- Transaction group
- Producer(s) in that transaction group
- Transaction coordinator for that transaction group
- Leader brokers (of the transaction payload's data partitions)
- Consumers of transactions
Transaction group
The transaction group is used to map to a specific transaction coordinator (say, based on a hash against the number of journal partitions). The producers in the group would need to be configured to use this group. Since all transactions from these producers go through this coordinator, we can achieve strict ordering across these transactional producers.
Producer IDs and state groups
In this section, I will go over the need to introduce two new parameters for transactional producers: producer ID and producer group. These don't necessarily need to be part of the producer configuration but may be specified as a parameter in the producer's transactional API.
The preceding overview describes the need to associate the input state of a producer (or in general a processor of some input) along with the last committed transaction. This enables the processor to redo a transaction (by recreating the input state for that transaction - which in our use cases is typically a vector of offsets).
We can utilize the consumer offset management feature to maintain this state. The consumer offset manager associates each key (consumergroup-topic-partition) to the last checkpointed offset and metadata for that partition. In the case of a transactional processor, we would want to save the offsets of its consumer that are associated with the commit point of the transaction. This offset commit record (in the __consumer_offsets topic) should be written as part of the transaction. i.e., the __consumer_offsets topic's partition that stores offsets for the consumer group will need to participate in the transaction. So (for example) suppose a producer fails in the middle of a transaction (which the transaction coordinator subsequently expires); when the producer recovers, it can issue an offset fetch request to recover the input offsets associated with the last committed transaction and resume transactional processing from that point.
There are a few enhancements that we need to make to the offset manager and the compacted __consumer_offsets topic in order to support this. First, the compacted topic will now contain transactional control records as well. We will need to come up with an eviction strategy for these control records. Second, the offset manager needs to become transaction-aware - specifically, an offset fetch request should return an error if the group is associated with a pending transaction.