[
https://issues.apache.org/jira/browse/HBASE-30037?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ASF GitHub Bot updated HBASE-30037:
-----------------------------------
Labels: pull-request-available (was: )
> FullTableBackupClient should not delete bulk load tracking entries before
> backup completion
> -------------------------------------------------------------------------------------------
>
> Key: HBASE-30037
> URL: https://issues.apache.org/jira/browse/HBASE-30037
> Project: HBase
> Issue Type: Bug
> Affects Versions: 2.6.4
> Reporter: Mohammad Nazhiif Al-Ghoniy
> Priority: Major
> Labels: pull-request-available
>
> This bug is similar to HBASE-28697, but for FullTableBackupClient:execute().
> If completeBackup() failed after the delete, the full backup would end up
> failed but the tracking information would already be gone. That leaves the
> system in an inconsistent state where the full backup is not valid, but the
> next incremental backup may think those HFiles were already safely covered
> and skip them. The fix is to delete those tracking rows only after the full
> backup has been successfully completed.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
