[
https://issues.apache.org/jira/browse/HDFS-4963?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Arpit Agarwal updated HDFS-4963:
--------------------------------
Attachment: HDFS-4963.branch-1.patch
Add a new setting {{dfs.namenode.master.name}} to allow choosing the target
hostname/IP address to bind to on a multi-homed NameNode. Without this setting
the result is unpredictable and depends on the DNS configuration.
Add documentation for the setting to {{hdfs-default.xml}}.
> Improve multihoming support in namenode
> ---------------------------------------
>
> Key: HDFS-4963
> URL: https://issues.apache.org/jira/browse/HDFS-4963
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode
> Affects Versions: 1.3.0
> Reporter: Arpit Agarwal
> Assignee: Arpit Agarwal
> Attachments: HDFS-4963.branch-1.patch
>
>
> HDFS does not work very well on multi-homed machines. A few open Jiras refer
> to this:
> # HDFS-1379
> # HADOOP-8198
> There are multiple issues involved here and some of them can be worked around
> by using alternate DNS names and configuring {{slave.host.name}} on Datanodes
> and task trackers.
> However namenode issues cannot be worked around because it does not respect
> the {{fs.default.name}} configuration. e.g. {{Namenode#initialize}} performs
> a gratuitous reverse DNS lookup to regenerate the hostname. Similar issues
> exist elsewhere.
> This Jira is being filed to fix some of the more egregious problems. To avoid
> affecting existing users a new config setting may be introduced.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
