damccorm opened a new pull request, #41022: URL: https://github.com/apache/beam/pull/41022
The Dataflow Unified Worker harness reads `streaming` (boolean) and `numberOfWorkerHarnessThreads` (number) from `sdk_pipeline_options.options` to configure streaming bundle processing parallelism (`max_bundle_processing_parallelism`) and sibling SDK worker sizing. Previously, the Go Dataflow runner only serialized `streaming` and `number_of_worker_harness_threads` as strings under `sdk_pipeline_options.go_options`, leaving `sdk_pipeline_options.options` without those fields. As a result, Go streaming pipelines on Dataflow Runner v2 always fell back to the default of 500 concurrent bundles per worker VM regardless of `--number_of_worker_harness_threads`. This change populates `streaming` and `numberOfWorkerHarnessThreads` in `dataflowOptions` (`sdk_pipeline_options.options`) when set. -- 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]
