Vanzeren commented on issue #25701: URL: https://github.com/apache/datafusion/issues/25701#issuecomment-5946749706
This appears to be fixed on main already, by #24677 (same root cause as #24661: the fallback re-executed the build-side child, the second execution came back empty, so the spill file had 0 rows and the probe side was never read). I verified on current main with a scaled-down version of the repro (500 rows, 1MB FairSpillPool): the outer NLJ now spills properly (`spill_count=2, spilled_rows=250K`) and returns the correct count (124,500,500) instead of 0. Should be good in the next release. -- 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]
