Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Wed, 10 Oct 2012 20:47:03 +0000 (UTC)
From: "Christopher Conner (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <455057428.22328.1349902023437.JavaMail.jiratomcat@arcas>
In-Reply-To: <1566823474.10942.1349731203157.JavaMail.jiratomcat@arcas>
Subject: [jira] [Updated] (HDFS-4021) Misleading error message when
 resources are low on the NameNode
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/HDFS-4021?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Christopher Conner updated HDFS-4021:
-------------------------------------

    Attachment: HDFS-4021.patch

New patch implementing Aaron's recommended changes.
                
> Misleading error message when resources are low on the NameNode
> ---------------------------------------------------------------
>
>                 Key: HDFS-4021
>                 URL: https://issues.apache.org/jira/browse/HDFS-4021
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: name-node
>    Affects Versions: 2.0.3-alpha
>            Reporter: Colin Patrick McCabe
>            Assignee: Christopher Conner
>            Priority: Minor
>              Labels: newbie
>         Attachments: HDFS-4021.patch, HDFS-4021.patch
>
>
> When resources are low on the namenode, it enters SafeMode with a message like this:
> {code}
> 14:00:10,666  WARN NameNodeResourceChecker:89 - Space available on
> volume '/dev/hda1' is 104357888, which is below the configured
> reserved amount 104857600
> 14:00:10,670  WARN FSNamesystem:3190 - NameNode low on available disk
> space. Entering safe mode.
> 14:00:10,670  INFO StateChange:3836 - STATE* Safe mode is ON.
> Resources are low on NN. Safe mode must be turned off manually.
> {code}
> However, turning off safe mode manually has no effect, since it immediately puts itself into Safe Mode again with a log message like this:
> {code}
>  14:00:10,666  WARN NameNodeResourceChecker:89 - Space available on
> volume '/dev/hda1' is 104357888, which is below the configured
> reserved amount 104857600
> 14:00:10,670  WARN FSNamesystem:3190 - NameNode low on available disk
> space. Entering safe mode.
> 14:00:10,670  INFO StateChange:3836 - STATE* Safe mode is ON.
> Resources are low on NN. Safe mode must be turned off manually.
> {code}
> From the shell, it looks like this:
> {code}
>  [cmccabe@vm1 h]$ ./bin/hdfs dfsadmin -safemode get
> Safe mode is ON
> [cmccabe@vm1 h]$ ./bin/hdfs dfsadmin -safemode leave
> Safe mode is OFF
> [cmccabe@vm1 h]$ ./bin/hdfs dfsadmin -safemode get
> Safe mode is ON
> {code}
> It seems like we should change the message about turning off safe mode manually, if turning off safe mode manually does not actually work for the case where resources are low.  Probably we need to explain that safe mode should be turned off manually *after* adding more resources.  As it is, the error message seems misleading.

--
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