hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "dhruba borthakur (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-993) Namenode does not need to store any data node info persistently.
Date Fri, 16 Mar 2007 22:02:09 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-993?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12481771

dhruba borthakur commented on HADOOP-993:

>From my understanding, if you have a include file then the namenode reads in that list
at startuptime. Thus, all the nodes listed in the include files will be shown in the UI. Those
datanodes that have not yet sent registration/heartbeats will be marked as dead.

Currently, the storageid is a string that is generated by the namenode by using a random number.
And you are right in saying that if the storageid is not persisted, then there is a possibility
of the namenode incorrectly re-issuing a previously issued storageid to a new datanode. Maybe
we can append the currenttime with a random number to generate the storageid.

> 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
>          Components: dfs
>            Reporter: Raghu Angadi
>         Assigned To: Sameer Paranjpye
>         Attachments: noDatanodesInFsimage.patch
> 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.

View raw message