Mailing-List: contact common-user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: common-user@hadoop.apache.org
Received-SPF: pass (nike.apache.org: domain of guxiaobo1982@gmail.com
 designates 209.85.218.48 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:date:message-id:subject:from:to:content-type;
        b=pl4iEzAi67fbhtfeMYPiWQus8KFgaV3ngskYh87aBnPVpG6aHJ3gSkt7wBw/rWuCze
         1GkeliEXDHkMbDDYbiPS81XBnc/6cuBxs+/VaZtecpcS9VhOR2QScpAALWlV4pcWQn14
         qig2Oc7/D8w3VNfL596ZjNEEXXLERINXeh9S4=
MIME-Version: 1.0
Date: Mon, 25 Apr 2011 23:04:11 +0800
Message-ID: <BANLkTikKYNG0AuFCUF+b0AuyacVP6vwnFg@mail.gmail.com>
Subject: Does it mean that single disk failure causes the whole datanode to
 fail?
From: Xiaobo Gu <guxiaobo1982@gmail.com>
To: common-user@hadoop.apache.org
Content-Type: text/plain; charset=ISO-8859-1

Hi,
I heard from so many people saying we should using JBOD instead of
RAID, that is we should format each local disk(used for data storage)
into an individual file system and define the mount point in
dfs.data.dir. In this kind of configuration, disk failure will cause
the coresponding file system fail, can datanode continue running with
file systems defined in dfs.data.dir not accessable? If the answer is
no, does it mean single disk failure will cause the whole data node to
fail?

Regards,

Xiaobo Gu