JulianJaffePinterest commented on pull request #12137: URL: https://github.com/apache/druid/pull/12137#issuecomment-1022907414
@loquisgon doesn't the same race condition exist though? Since tombstones will be loaded instantly but actual data segments will need to be read from deep storage, memory mapped, etc. if there's load on the historicals data will be dropped from a client perspective (e.g. replaced with a tombstone) before fresh data is available. If we take your logic to determine which data segments to tombstone and instead use it to identify which segments to mark as unused (leaving the segments that will be overshadowed marked as used), what race condition does it introduce compared to your PR? -- 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]
