Package org.apache.spark.sql.connector.write.streaming

Interface StreamingWrite

@Evolving public interface StreamingWrite
An interface that defines how to write the data to data source in streaming queries. The writing procedure is:
  1. Create a writer factory by createStreamingWriterFactory(PhysicalWriteInfo), serialize and send it to all the partitions of the input data(RDD).
  2. For each epoch in each partition, create the data writer, and write the data of the epoch in the partition with this writer. If all the data are written successfully, call DataWriter.commit(). If exception happens during the writing, call DataWriter.abort().
  3. If writers in all partitions of one epoch are successfully committed, call commit(long, WriterCommitMessage[]). If some writers are aborted, or the job failed with an unknown reason, call abort(long, WriterCommitMessage[]).

While Spark will retry failed writing tasks, Spark won't retry failed writing jobs. Users should do it manually in their Spark applications if they want to retry.

Please refer to the documentation of commit/abort methods for detailed specifications.

Since:
3.0.0

  • Method Details

    • createStreamingWriterFactory

      StreamingDataWriterFactory createStreamingWriterFactory(PhysicalWriteInfo info)
      Creates a writer factory which will be serialized and sent to executors.

      If this method fails (by throwing an exception), the action will fail and no Spark job will be submitted.

      Parameters:
      info - Information about the RDD that will be written to this data writer

    • useCommitCoordinator

      default boolean useCommitCoordinator()
      Returns whether Spark should use the commit coordinator to ensure that at most one task for each partition commits.
      Returns:
      true if commit coordinator should be used, false otherwise.

    • commit

      void commit(long epochId, WriterCommitMessage[] messages)
      Commits this writing job for the specified epoch with a list of commit messages. The commit messages are collected from successful data writers and are produced by DataWriter.commit().

      If this method fails (by throwing an exception), this writing job is considered to have been failed, and the execution engine will attempt to call abort(long, WriterCommitMessage[]).

      The execution engine may call commit multiple times for the same epoch in some circumstances. To support exactly-once data semantics, implementations must ensure that multiple commits for the same epoch are idempotent.

    • abort

      void abort(long epochId, WriterCommitMessage[] messages)
      Aborts this writing job because some data writers are failed and keep failing when retried, or the Spark job fails with some unknown reasons, or commit(long, WriterCommitMessage[]) fails.

      If this method fails (by throwing an exception), the underlying data source may require manual cleanup.

      Unless the abort is triggered by the failure of commit, the given messages will have some null slots, as there may be only a few data writers that were committed before the abort happens, or some data writers were committed but their commit messages haven't reached the driver when the abort is triggered. So this is just a "best effort" for data sources to clean up the data left by data writers.