[
https://issues.apache.org/jira/browse/HADOOP-993?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12471443
]
Raghu Angadi commented on HADOOP-993:
-------------------------------------
I should have specified that this was mainly a place holder. Details are
deliberately unclear.
I am not sure of all the motivations and associated issues yet. I think this is
more of a desired policy that needs to be thought about. Broadly, the idea is
Namenode does not need any peristent info about the datanodes. When it starts
it just learns about the datanodes as they register.
> Namenode does not need to store any data node info persistently.
> ----------------------------------------------------------------
>
> Key: HADOOP-993
> URL: https://issues.apache.org/jira/browse/HADOOP-993
> Project: Hadoop
> Issue Type: Improvement
> Reporter: Raghu Angadi
> Assigned To: Sameer Paranjpye
>
> Namenode does not need to serialize datanode info. It will map datanode to
> storageID when datanode register.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
