Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Message-ID: <9838106.196471284528096300.JavaMail.jira@thor>
Date: Wed, 15 Sep 2010 01:21:36 -0400 (EDT)
From: "dhruba borthakur (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Subject: [jira] Commented: (HDFS-779) Automatic move to safe-mode when
 cluster size drops
In-Reply-To: <492528515.1258656579663.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HDFS-779?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12909598#action_12909598 ] 

dhruba borthakur commented on HDFS-779:
---------------------------------------

@Joy: I agree that not losing heartbeats is the best scenario. In the case I mentioned in my above comment, we would not lose datanode heartbeats if we implement HDFS-1392.

On the other hand, if there is a true network partition, then NN will "lose heartbeats" from datanodes, because the datanodes cannot send messages to the NN. In that case, the NN should delay creating a replication-storm in the hope that the network partition gets resolved soon. So, the heuristic that I listed above to detect a network partition should still be applicable, isn't it?

> Automatic move to safe-mode when cluster size drops
> ---------------------------------------------------
>
>                 Key: HDFS-779
>                 URL: https://issues.apache.org/jira/browse/HDFS-779
>             Project: Hadoop HDFS
>          Issue Type: New Feature
>          Components: name-node
>            Reporter: Owen O'Malley
>            Assignee: dhruba borthakur
>
> As part of looking at using Kerberos, we want to avoid the case where both the primary (and optional secondary) KDC go offline causing a replication storm as the DataNodes' service tickets time out and they lose the ability to connect to the NameNode. However, this is a specific case of a more general problem of loosing too many nodes too quickly. I think we should have an option to go into safe mode if the cluster size goes down more than N% in terms of DataNodes.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.