[jira] [Commented] (CASSANDRA-21252) CEP-45: BRR doesn't do tracked writes when needed during migration

["Caleb Rackliffe (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-21252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18076083#comment-18076083 ] 

Caleb Rackliffe commented on CASSANDRA-21252:
---------------------------------------------

+1 pending CI not regressing

Also, I think {{WriteCallback}} is dead code now

> CEP-45: BRR doesn't do tracked writes when needed during migration
> ------------------------------------------------------------------
>
>                 Key: CASSANDRA-21252
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-21252
>             Project: Apache Cassandra
>          Issue Type: Bug
>          Components: Consistency/Coordination
>            Reporter: Ariel Weisberg
>            Assignee: Ariel Weisberg
>            Priority: Normal
>          Time Spent: 2h 10m
>  Remaining Estimate: 0h
>
> I have some code in this direction and we have done similar for Accord migration so it's not a completely new invention, but we need to add this.
> During migration for a range we continue to do BRR because mutation tracking doesn't have enough state initialized to provide monotonic reads. Repair completes the migration and on completion TCM state is updated to reflect that it is safe for the range to use BRR.
> All the writes after the repair started needed me tracked writes so that it is safe to transition to using tracked reads.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org