You are viewing a plain text version of this content. The canonical link for it is here.

Posted to hdfs-issues@hadoop.apache.org by "Rakesh R (JIRA)" <ji...@apache.org> on 2015/10/26 07:57:27 UTC

[jira] [Commented] (HDFS-9261) Erasure Coding: Skip encoding the data cells if all the parity data streamers are failed for the current block group

    [ https://issues.apache.org/jira/browse/HDFS-9261?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14973827#comment-14973827 ] 

Rakesh R commented on HDFS-9261:
--------------------------------

Attached simple patch to check the health of parity streamers before writeParityCells. Please review the changes!

> Erasure Coding: Skip encoding the data cells if all the parity data streamers are failed for the current block group
> --------------------------------------------------------------------------------------------------------------------
>
>                 Key: HDFS-9261
>                 URL: https://issues.apache.org/jira/browse/HDFS-9261
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Rakesh R
>            Assignee: Rakesh R
>            Priority: Minor
>         Attachments: HDFS-9261-00.patch
>
>
> {{DFSStripedOutputStream}} will continue writing with minimum number (dataBlockNum) of live datanodes. It won't replace the failed datanodes immediately for the current block group. Consider a case where all the parity data streamers are failed, now it is unnecessary to encode the data block cells and generate the parity data. This is a corner case where it can skip {{writeParityCells()}} step.



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