Return-Path: X-Original-To: apmail-hadoop-hdfs-user-archive@minotaur.apache.org Delivered-To: apmail-hadoop-hdfs-user-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id C6473955A for ; Wed, 1 Feb 2012 03:39:41 +0000 (UTC) Received: (qmail 30685 invoked by uid 500); 1 Feb 2012 03:37:49 -0000 Delivered-To: apmail-hadoop-hdfs-user-archive@hadoop.apache.org Received: (qmail 29982 invoked by uid 500); 1 Feb 2012 03:37:32 -0000 Mailing-List: contact hdfs-user-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: hdfs-user@hadoop.apache.org Delivered-To: mailing list hdfs-user@hadoop.apache.org Received: (qmail 29461 invoked by uid 99); 1 Feb 2012 03:31:37 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 01 Feb 2012 03:31:37 +0000 X-ASF-Spam-Status: No, hits=-0.7 required=5.0 tests=RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of harsh@cloudera.com designates 209.85.160.48 as permitted sender) Received: from [209.85.160.48] (HELO mail-pw0-f48.google.com) (209.85.160.48) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 01 Feb 2012 03:31:31 +0000 Received: by pbcc11 with SMTP id c11so936763pbc.35 for ; Tue, 31 Jan 2012 19:31:10 -0800 (PST) Received: by 10.68.74.167 with SMTP id u7mr54963205pbv.103.1328067070218; Tue, 31 Jan 2012 19:31:10 -0800 (PST) MIME-Version: 1.0 Received: by 10.142.72.17 with HTTP; Tue, 31 Jan 2012 19:30:50 -0800 (PST) In-Reply-To: References: From: Harsh J Date: Wed, 1 Feb 2012 09:00:50 +0530 Message-ID: Subject: Re: Adding dfs.name.dir To: hdfs-user@hadoop.apache.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Virus-Checked: Checked by ClamAV on apache.org Hey Prem, Inline. On Wed, Feb 1, 2012 at 6:58 AM, Jain, Prem wrot= e: > Thanks Joey, > > So this means, I edit the hdfs-site.xml file to reflect the empty directo= ry, restart the namenode. That is correct. Just append the new dir to the list of dfs.name.dir and restart. > Will Namenode populate =A0all the metadata to the =A0newly added =A0direc= tory ? Yes, just ensure the dir is writable by the 'hdfs' user, or the user that runs the NameNode. > Just wanted to make sure I do everything right. =A0While =A0googling, I a= lso noticed issues when one of the directories becomes unavailable (NFS mou= nt), HDFS hangs, is this still true ? > > https://groups.google.com/a/cloudera.org/group/cdh-user/msg/aa014cf61dc2d= aae? First, that was not a HDFS issue, it was a lockd deadlock condition when the NFS server went unreachable. Second, that is no longer true if you are on an updated version of CentOS. IIRC it was a lockd bug in an earlier version of CentOS 5 (5.3 perhaps, the exact version eludes me presently), that went away with a fix later. So just make sure you are using one of the latest stable kernels and the unlock call won't hang. --=20 Harsh J Customer Ops. Engineer Cloudera | http://tiny.cloudera.com/about