[
https://issues.apache.org/jira/browse/HBASE-29350?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17953422#comment-17953422
]
Vinayak Hegde commented on HBASE-29350:
---------------------------------------
We'll do the following when the last remaining full backup is deleted:
* Remove all the entries from the continuous backup table set in the system
table.
* Delete all WAL files under the backup WAL directory (without deleting the
directory itself).
* Disable and the continuous backup replication peer.
> Ensure Cleanup of Continuous Backup WALs After Last Backup is Force Deleted
> ---------------------------------------------------------------------------
>
> Key: HBASE-29350
> URL: https://issues.apache.org/jira/browse/HBASE-29350
> Project: HBase
> Issue Type: Task
> Components: backup&restore
> Reporter: Vinayak Hegde
> Assignee: Vinayak Hegde
> Priority: Major
>
> When the last backup is deleted using the {{-fd}} (force delete) option,
> there are no remaining backups to recover from. In this state:
> * Continuous backup metadata in the system table becomes obsolete.
> * WAL files under the backup WAL directory are no longer needed.
> * The replication peer used for continuous WAL tracking serves no purpose.
> Since the entire backup chain is gone, we can safely clean up all related
> metadata and WAL artifacts to free up storage and avoid stale state in the
> system.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
