pvary commented on PR #15042: URL: https://github.com/apache/iceberg/pull/15042#issuecomment-3783019981
> > Do we need the previously stored `TableChange`? Or we just acquire the new lock, and wait until it is removed by the LockRemover. After that we can start sending the proper payload > > 1. Does this mean we no longer need to store the previous state, such as the merged value after _n_ `TableChange` operations? My bad. You are right, we need to keep the `TableChange` in state. > 2. The current approach is that during recovery in `initializeState` we directly send a lock request with timestamp `Long.MaxValue` instead of using `recoverLock`, and only after `LockRemover` releases this lock do we start firing and sending triggers downstream. Is this logic aligned with what you had in mind? This is right. The approach seems reasonable to me. -- 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]
