RockteMQ-AI commented on issue #10489: URL: https://github.com/apache/rocketmq/issues/10489#issuecomment-4709877351
**Issue Evaluation** Category: `type/enhancement` | Status: **Evaluated** **Feasibility:** Feasible, but complex scope **Scope:** Proxy layer (gRPC handler), Broker layer (new request code + processor), potentially `rocketmq-clients` **Compatibility:** New request code — backward compatible. Older brokers will reject the batch request; Proxy should fall back to individual calls. This is a meaningful optimization for scenarios with bulk invisible time changes (client disconnect, write-back failure, periodic renewal). The proposal to follow the existing `BatchAck` pattern is sound. Key considerations: - **Batch size limit** — Need a configurable upper bound to prevent oversized requests. - **Partial failure semantics** — If some messages in the batch fail, should the whole batch rollback or return per-message results? - **RocksDB flush reduction** — The batch write should use a single `WriteBatch` to amortize flush cost. - **Client SDK coordination** — If the Java/Go/C++ clients need to call this directly, protocol changes in `rocketmq-clients` are needed. This is a larger feature that benefits from a design discussion before implementation. A RIP (RocketMQ Improvement Proposal) or detailed design doc would help align on the approach. --- *Automated evaluation by RockteMQ-AI* -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
