You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Jingcheng Du (JIRA)" <ji...@apache.org> on 2016/10/14 11:35:20 UTC
[jira] [Updated] (HBASE-16841) Data loss in MOB files after cloning
a snapshot and deleting that snapshot
[ https://issues.apache.org/jira/browse/HBASE-16841?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jingcheng Du updated HBASE-16841:
---------------------------------
Description:
When taking a snapshot on a enabled mob table, each region flushes itself and takes a snapshot, and the mob snapshot is taken only if the current region is first region of the table. At that time, the flushing of some regions might not be finished, and some mob files are not flushed to disk yet. Eventually some mob files are not recorded in the snapshot manifest.
To solve this, we need to generate a mob manifest for each region server after the snapshot of all online regions in that server are finished, and merge them in master when all the snapshots of regions are finished at last.
was:
When taking a snapshot on a enabled mob table, a region flushes itself and takes a snapshot, and the mob snapshot is taken if it is first region of the table. At that time, the flushing of some regions might not be finished, and some mob files are not flushed to disk yet. Eventually some mob files are not existent in the snapshot manifest.
We need to have a mob manifest for each region server after the snapshot of all online regions in that server are finished, and merge them in master when all the snapshots of regions are finished at last.
> Data loss in MOB files after cloning a snapshot and deleting that snapshot
> --------------------------------------------------------------------------
>
> Key: HBASE-16841
> URL: https://issues.apache.org/jira/browse/HBASE-16841
> Project: HBase
> Issue Type: Bug
> Components: mob, snapshots
> Reporter: Jingcheng Du
> Assignee: Jingcheng Du
>
> When taking a snapshot on a enabled mob table, each region flushes itself and takes a snapshot, and the mob snapshot is taken only if the current region is first region of the table. At that time, the flushing of some regions might not be finished, and some mob files are not flushed to disk yet. Eventually some mob files are not recorded in the snapshot manifest.
> To solve this, we need to generate a mob manifest for each region server after the snapshot of all online regions in that server are finished, and merge them in master when all the snapshots of regions are finished at last.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)