You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-user@hadoop.apache.org by Dennis <ar...@yahoo.com.cn> on 2010/05/08 10:14:57 UTC

hadoop start error - could only be replicated to 0 nodes, instead of 1

Hi,I am trying to set up a Hadoop claster. I have a server machine, and I installed XenServer on it. I installed 3 Debian Lenny VMs on the XenServer. And on every VM, I installed sun-java6-jre, openssh-server, rsync and hadoop. VM jack is the namenode, VM kim and VM lynne are datanodes. Everything was cool until I run bin/start-all.sh. I attached three configuration files and the hadoop-cs-namenode-jack.log file who printed a lot errors.
Two important errors are:java.lang.IllegalArgumentException: Duplicate metricsName:getProtocolVersionjava.io.IOException: File /home/cs/HadoopInstall/tmp/mapred/system/jobtracker.info could only be replicated to 0 nodes, instead of 1
Thanks.Dennis