hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Suresh Srinivas (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-1673) Federation: Datanode changes to track block token secret per namenode.
Date Mon, 28 Feb 2011 20:11:36 GMT

    [ https://issues.apache.org/jira/browse/HDFS-1673?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13000494#comment-13000494

Suresh Srinivas commented on HDFS-1673:

Making the following changes to address it:
# Create a BlockPoolTokenSecretManager that has a map of block pool to corresponding BlockTokenSecretManager.
# Routes the calls based on input block pool Id to corresponding BlockTokenSecretManager.
# Datanode adds block pool and corresponding BlockTokenSecretManager to BlockPoolTokenSecretManager
when registration
for corresponding block pool completes.
# BlockPoolTokenSecretManager is used by Datanode RPC server and DataXceiver.
# BlockTokenIdentifier now has an additional field blockPoolId.

> Federation: Datanode changes to track block token secret per namenode. 
> -----------------------------------------------------------------------
>                 Key: HDFS-1673
>                 URL: https://issues.apache.org/jira/browse/HDFS-1673
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: data-node, name-node
>    Affects Versions: Federation Branch
>            Reporter: Suresh Srinivas
>            Assignee: Suresh Srinivas
>             Fix For: Federation Branch
> Datanode currently keeps track of a single block token secret. Given datanode talks to
multiple namenodes, it needs to track it per namenode

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message