[
https://issues.apache.org/jira/browse/HDFS-9440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16845675#comment-16845675
]
Fred Peng commented on HDFS-9440:
---------------------------------
HI [~xiaochen],do you still work on this? The fix HDFS-9314 can not handle
\{SSD(rack 1), DISK(rack 3), DISK(rack 3), DISK(rack 3)}. Thus, the SSD replica
is always left in cluster.
> Improve BlockPlacementPolicyDefault's to be able to rebalance when picking of
> excess replicas
> ---------------------------------------------------------------------------------------------
>
> Key: HDFS-9440
> URL: https://issues.apache.org/jira/browse/HDFS-9440
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Xiao Chen
> Priority: Major
>
> The test case used in HDFS-9313 and HDFS-9314 identified the limitation of
> excess replica picking. If the current replicas are on
> {SSD(rack 1), DISK(rack 3), DISK(rack 3), DISK(rack 3)}
> and the storage policy changes to HOT_STORAGE_POLICY_ID,
> BlockPlacementPolicyDefault's won't be able to delete SSD replica, because
> deleting the SSD on rack 1 will violate the block placement policy.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
