jonkeane commented on pull request #11730: URL: https://github.com/apache/arrow/pull/11730#issuecomment-1010040486
Interesting! I've got one theory about the segfault. But still no clue what's going on in the other case (i.e. autodisconnect = FALSE / gc after each =TRUE). That is still very strange! We've narrowed it down that if the scan is _starting_ in DuckDB that seems to work. (have you tried setting `PRAGMA threads=N` to ensure parallelism in duckdb — this should be default now, but I don't remember which version that started in...). Have you tried running the duckdb -> arrow bit only multiple times checking the results (like you have here with the round trip)? -- 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]
