netudima commented on PR #4825: URL: https://github.com/apache/cassandra/pull/4825#issuecomment-4534367564
ok, then the MR description seems incorrect and confusing from this point of view, it is not a new logic which we are trying to fix, it is an attempt to adjust the existing logic. > If you want to not make it functionally on par For me it looks as quite independent activity. I understand the danger of the current offline tools and agree that a protection is needed but it is not related to adding a new parameter.. Also I suppose the ID conflict is not only the issue and just adding the optimistic locking loop would not make the tool safe to use, you still have other issues like (but not limited to): - interference of the offiline process with the SSTable transactions are executed by DB process (like compaction) - visibility for Cassandra DB of the new SSTables So, for me it is a kind of separate independent story to check if DB process exists or to implement proper ability to execute the tool and DB process concurrently.., mixing it with an offline tool flag adding is a bit confusing (at least for me). I also want to avoid the illusion that only this change is enough to make the logic safe, as a result people may start to think that it is ok to avoid the warn, start to use the tool and get all kinds of issues hard to troubleshoot -- 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]
