Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Wed, 7 Mar 2012 22:56:57 +0000 (UTC)
From: "Eli Collins (Commented) (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: 
 <958700930.36759.1331161017300.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <1259796423.14836.1314301470826.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] [Commented] (HDFS-2288) Replicas awaiting recovery should
 return a full visible length
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HDFS-2288?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13224813#comment-13224813 ] 

Eli Collins commented on HDFS-2288:
-----------------------------------

Hey Todd,

I think you're right, there's a bug in the way visible length is calculated. My understanding of visible length is "the length that all datanodes in the pipeline contain at least such amount of data." (HDFS-814), and given that RBW#getVisibleLength returns getBytesAcked() and not all DNs necessarily have the same # bytes Ack'd, they could disagree on the visible length of the replica. That being said, aside from this bug (and potentially others) the intent / goal should be to maintain this property right?

What's the case in HBase that can hit this, it's reading a file that's recovering its pipeline that is also under lease recovery? 

Thanks,
Eli
                
> Replicas awaiting recovery should return a full visible length
> --------------------------------------------------------------
>
>                 Key: HDFS-2288
>                 URL: https://issues.apache.org/jira/browse/HDFS-2288
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: data-node
>    Affects Versions: 0.23.0
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>            Priority: Critical
>             Fix For: 0.24.0
>
>         Attachments: hdfs-2288.txt
>
>
> Currently, if the client calls getReplicaVisibleLength for a RWR, it returns a visible length of 0. This causes one of HBase's tests to fail, and I believe it's incorrect behavior.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira