[
https://issues.apache.org/jira/browse/HDFS-9876?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15174488#comment-15174488
]
Tsz Wo Nicholas Sze commented on HDFS-9876:
-------------------------------------------
excessTypes should be computed if it is needed. Let's move it inside
{code}
if (candidates.size() > 1) {
..
} {code}
> shouldProcessOverReplicated should not count number of pending replicas
> -----------------------------------------------------------------------
>
> Key: HDFS-9876
> URL: https://issues.apache.org/jira/browse/HDFS-9876
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Takuya Fukudome
> Assignee: Jing Zhao
> Attachments: HDFS-9876.000.patch
>
>
> Currently when checking if we should process over-replicated block in
> {{addStoredBlock}}, we count both the number of reported replicas and pending
> replicas. However, {{processOverReplicatedBlock}} chooses excess replicas
> only among all the reported storages of the block. So in a situation where we
> have over-replicated replica/internal blocks which only reside in the pending
> queue, we will not be able to choose any extra replica to delete.
> For contiguous blocks, this causes {{chooseExcessReplicasContiguous}} to do
> nothing. But for striped blocks, this may cause endless loop in
> {{chooseExcessReplicasStriped}} in the following while loop:
> {code}
> while (candidates.size() > 1) {
> List<DatanodeStorageInfo> replicasToDelete = placementPolicy
> .chooseReplicasToDelete(nonExcess, candidates, (short) 1,
> excessTypes, null, null);
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
