[GitHub] [hudi] danny0405 commented on issue #9101: [SUPPORT] Transaction and spark job final state inconsistency in batch processing

danny0405 commented on issue #9101: URL: https://github.com/apache/hudi/issues/9101#issuecomment-1619350001 inline archiving and cleaning may have this issue, do you try the async cleaning instead? Is there any spark param to control the failover behavior, seems not very easy to fix from Hu

[GitHub] [hudi] danny0405 commented on issue #9101: [SUPPORT] Transaction and spark job final state inconsistency in batch processing

danny0405 commented on issue #9101: URL: https://github.com/apache/hudi/issues/9101#issuecomment-1617582944 Is there anyway the cleaning and archive does not trigger recomputation of the whole job? -- This is an automated message from the Apache Git Service. To respond to the message, ple

[GitHub] [hudi] danny0405 commented on issue #9101: [SUPPORT] Transaction and spark job final state inconsistency in batch processing

danny0405 commented on issue #9101: URL: https://github.com/apache/hudi/issues/9101#issuecomment-1617520203 > when we trigger mayBeCleanAndArchive, it throw Exception, and make this job failed, it will retry in job level. But this commit has commit. So the final result is: instance commit