[jira] [Updated] (CASSANDRA-18460) CEP-21 Ensure that ClusterMetadata::forceEpoch keeps component epochs consistent

["Sam Tunnicliffe (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CASSANDRA-18460?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sam Tunnicliffe updated CASSANDRA-18460:
----------------------------------------
    Change Category: Semantic
         Complexity: Normal
          Reviewers: Alex Petrov, Marcus Eriksson
             Status: Open  (was: Triage Needed)

https://github.com/beobal/cassandra/commit/4be5fb03122ed629f74bb121c7be48c6f0dabd27

> CEP-21 Ensure that ClusterMetadata::forceEpoch keeps component epochs consistent
> --------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-18460
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-18460
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Transactional Cluster Metadata
>            Reporter: Sam Tunnicliffe
>            Assignee: Sam Tunnicliffe
>            Priority: Normal
>
> {{forceEpoch}} is used to make the application of multiple transformations in series appear as a single atomic update. It's primary use is in {{UnsafeJoin}} (i.e. join without bootstrap) to apply the usual sequence of start/mid/finish join in a single transformation. In such cases, we must ensure that no component of {{ClusterMetadata}} which maintains its own last-modified epoch, ends up with an epoch greater than the one of the enclosing {{ClusterMetadata}}. 



--
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