nzw921rx opened a new issue, #10607: URL: https://github.com/apache/seatunnel/issues/10607
### Search before asking - [x] I had searched in the [feature](https://github.com/apache/seatunnel/issues?q=is%3Aissue+label%3A%22Feature%22) and found no similar feature requirement. ### Description Currently, JDBC sink only supports batch-size-based flushing. Data is flushed to the database only when the batch reaches `batch_size`, or during `prepareCommit` / `close`. This causes a problem in low-throughput scenarios (e.g., CDC real-time sync with sparse updates): if data comes slowly and the batch never fills up, records stay in memory for a long time before being written to the target database, leading to high data latency. Apache Flink's JDBC connector already supports this feature with `sink.buffer-flush.interval` (default 1s). SeaTunnel's `JdbcOutputFormat` was originally derived from Flink's design but is missing this time-based flush capability. ### Proposed Solution Add a `batch_interval_ms` option to the JDBC sink connector. **Config example:** ```hocon sink { jdbc { url = "jdbc:mysql://localhost:3306/test" driver = "com.mysql.cj.jdbc.Driver" batch_size = 1000 batch_interval_ms = 1000 # flush every 1 second } } ``` **Implementation approach:** In `JdbcOutputFormat.open()`, start a `ScheduledExecutorService` that periodically calls `flush()`: ```java if (batchIntervalMs > 0 && batchSize != 1) { this.scheduler = Executors.newScheduledThreadPool(1); this.scheduledFuture = scheduler.scheduleWithFixedDelay( () -> { synchronized (JdbcOutputFormat.this) { if (!closed) { try { flush(); } catch (Exception e) { flushException = e; } } } }, batchIntervalMs, batchIntervalMs, TimeUnit.MILLISECONDS); } ``` Shut down the scheduler in `close()` before flushing remaining data. **Key design points:** - Default value: `0` (disabled, preserving current behavior for backward compatibility) - Thread safety: use `synchronized` on the existing `writeRecord` / `flush` lock - Exception handling: timer thread stores exception in `flushException`, checked on next `writeRecord` call (same pattern as Flink) - Shutdown order: cancel scheduler → flush remaining → close connection **Changes needed:** 1. `JdbcSinkOptions`: add `BATCH_INTERVAL_MS` option 2. `JdbcConnectionConfig`: add `batchIntervalMs` field 3. `JdbcOutputFormat`: add scheduler lifecycle (open/close) and periodic flush logic 4. Documentation: update JDBC sink docs (en & zh) 5. Tests: unit test for config parsing + E2E test verifying flush behavior ### Use Cases - CDC real-time sync with low write frequency - Any scenario where data latency matters more than throughput - Align with Flink JDBC connector behavior for users migrating from Flink ### Usage Scenario _No response_ ### Related issues _No response_ ### Are you willing to submit a PR? - [x] Yes I am willing to submit a PR! ### Code of Conduct - [x] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct) -- 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]
