CommitLog Recovery replay stop on first timestamp after restore-point-in-time

Hi, (Moved from user mailing list to here) I have found a statement in the Datastax documentation regarding CommitLog recovery that concerns me, namely: "*Restore stops when the first client-supplied timestamp is greater than the restore point timestamp. Because the order in which the database r

Re: CommitLog Recovery replay stop on first timestamp after restore-point-in-time

I don’t have any personal knowledge of the fix but out of interest I took a look in Jira and it looks sounds to me like the behaviour was fixed here (in 2.0.10): https://issues.apache.org/jira/browse/CASSANDRA-6905 --- *Ben Slater* *Chief Product Officer*

Re: CommitLog Recovery replay stop on first timestamp after restore-point-in-time

Thanks for the response Ben this was exactly what I was looking for! I think will write a test to ensure this condition holds when we do Cassandra upgrades, and the linked test in CASSANDRA-6905 seems to be a nice starting point. Best regards

Re: Question about PartitionUpdate.singleRowUpdate()

Thanks for the feedback. I've opened: https://issues.apache.org/jira/browse/CASSANDRA-14941 SK On 2018-12-19 17:03, Jeff Jirsa wrote: > Definitely worth a JIRA. Suspect it may be slow to get a response this > close to the holidays, but a JIRA will be a bit more durable than the > mailing list p