andygrove commented on PR #2638: URL: https://github.com/apache/datafusion-ballista/pull/2638#issuecomment-6880483543
Thanks for bumping the pin. With #2721 in, a plain `UNION ALL` over an Iceberg table on a split stage now fails with a clear error instead of duplicating rows, which is a big improvement. I'd like to get this into the upcoming release as experimental. Could you mark the docs page and the crate docs as experimental, like `pipelined-execution.md` does? Could you also add a limitations bullet saying that a `UNION ALL` of Iceberg scans can fail when the adaptive planner splits its stage, for example with `ballista.scheduler.max_partitions_per_task` set or on a busy executor? The error points at `SchedulerConfig::partition_sliceable_resolver`, which standalone users can't set, so a note there would help. With those I'm happy to approve. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
