hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mingliang Liu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-8938) Refactor BlockManager in blockmanagement
Date Thu, 27 Aug 2015 22:17:47 GMT

    [ https://issues.apache.org/jira/browse/HDFS-8938?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14717651#comment-14717651
] 

Mingliang Liu commented on HDFS-8938:
-------------------------------------

The failing tests
{code}
hadoop.hdfs.server.blockmanagement.TestBlockManager
hadoop.hdfs.TestRenameWhileOpen
{code}

can pass with patch v7 on my local dev box.

> Refactor BlockManager in blockmanagement
> ----------------------------------------
>
>                 Key: HDFS-8938
>                 URL: https://issues.apache.org/jira/browse/HDFS-8938
>             Project: Hadoop HDFS
>          Issue Type: Task
>          Components: build
>            Reporter: Mingliang Liu
>            Assignee: Mingliang Liu
>         Attachments: HDFS-8938.000.patch, HDFS-8938.001.patch, HDFS-8938.002.patch, HDFS-8938.003.patch,
HDFS-8938.004.patch, HDFS-8938.005.patch, HDFS-8938.006.patch, HDFS-8938.007.patch
>
>
> This jira tracks the effort of refactoring inner classes {{BlockManager$BlockToMarkCorrupt}}
and {{BlockManager$ReplicationWork}} in {{hdfs.server.blockmanagement}} package. As the line
number of {{BlockManager}} is getting larger than 2000, we can move those two inner classes
out of the it.
> Meanwhile, the logic in method {{computeReplicationWorkForBlocks}} can be simplified
if we extract code sections to _schedule replication_ and to _validate replication work_ to
private helper methods respectively.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message