Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Wed, 29 Feb 2012 00:36:54 +0000 (UTC)
From: "Todd Lipcon (Commented) (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: 
 <1205470459.435.1330475814726.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <267347033.422.1330475695515.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] [Commented] (HDFS-3028) DFS rollback broken
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HDFS-3028?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13218767#comment-13218767 ] 

Todd Lipcon commented on HDFS-3028:
-----------------------------------

When I run {{start-dfs.sh -rollback}} the DNs do properly get the {{-rollback}} flag, but then it doesn't seem to take effect. I am seeing the following, after which the DN exits:

{code}
2012-02-28 16:31:32,409 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Block pool <registering> (storage id unknown) connecting to c0804.hal.cloudera.com/172.29.97.195:9020 starting to offer service
2012-02-28 16:31:32,411 INFO org.apache.hadoop.ipc.Server: IPC Server Responder: starting
2012-02-28 16:31:32,412 INFO org.apache.hadoop.ipc.Server: IPC Server listener on 43280: starting
2012-02-28 16:31:32,589 FATAL org.apache.hadoop.hdfs.server.datanode.DataNode: Initialization failed for block pool Block pool BP-252729080-172.29.97.195-1330388141883 (storage id ) registered with c0804.hal.cloudera.com/172.29.97.195:9020
org.apache.hadoop.hdfs.server.common.IncorrectVersionException: Unexpected version of storage directory /data/1/todd/dn. Reported: -40. Expecting = -39.
        at org.apache.hadoop.hdfs.server.common.Storage.setLayoutVersion(Storage.java:952)
        at org.apache.hadoop.hdfs.server.datanode.DataStorage.setFieldsFromProperties(DataStorage.java:305)
        at org.apache.hadoop.hdfs.server.common.Storage.readProperties(Storage.java:791)
        at org.apache.hadoop.hdfs.server.datanode.DataStorage.doTransition(DataStorage.java:375)
        at org.apache.hadoop.hdfs.server.datanode.DataStorage.recoverTransitionRead(DataStorage.java:194)
        at org.apache.hadoop.hdfs.server.datanode.DataStorage.recoverTransitionRead(DataStorage.java:222)
        at org.apache.hadoop.hdfs.server.datanode.DataNode.initStorage(DataNode.java:897)
        at org.apache.hadoop.hdfs.server.datanode.DataNode.initBlockPool(DataNode.java:856)
        at org.apache.hadoop.hdfs.server.datanode.BPOfferService.connectToNNAndHandshake(BPOfferService.java:241)
        at org.apache.hadoop.hdfs.server.datanode.BPOfferService.run(BPOfferService.java:622)
        at java.lang.Thread.run(Thread.java:662)
2012-02-28 16:31:32,590 WARN org.apache.hadoop.hdfs.server.datanode.DataNode: Ending block pool service for: Block pool BP-252729080-172.29.97.195-1330388141883 (storage id ) registered with c0804.hal.cloudera.com/172.29.97.195:9020
{code}

This is on trunk - haven't tried the same on branch-0.23.
                
> DFS rollback broken
> -------------------
>
>                 Key: HDFS-3028
>                 URL: https://issues.apache.org/jira/browse/HDFS-3028
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: data-node
>    Affects Versions: 0.24.0
>            Reporter: Todd Lipcon
>
> I ran an upgrade to test a patch which updates the data format, but when I then tried to rollback, the DNs don't seem to respond properly.

--
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