aIbrahiim commented on PR #39075: URL: https://github.com/apache/beam/pull/39075#issuecomment-5096905838
> What's weird is that the FnAPI FinalizeBundle path should have the same problem - it still relies on Windmill sending best effort finalize ids. Yes I think on Dataflow the Fn API FinalizeBundle path still depends on Windmill sending applied_finalize_ids, so it should have the same best-effort limitation and that part of my explanation was wrong So I’m less sure why portable isn’t showing the same flake. Maybe the SDF execution path is different enough or maybe V2 just doesn’t run the same tests as often (e.g. bounded SDF is excluded from the V2 streaming ValidatesRunner suite). Or maybe it is flaky there too and we just haven’t noticed as much -- 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]
