Hi, Do we have to run multiple task trackers when running multiple data nodes on a single computer?
Regards, Xiaobo Gu > -----Original Message----- > From: Xiaobo Gu [mailto:guxiaobo1...@gmail.com] > Sent: Friday, July 01, 2011 11:35 PM > To: hdfs-user@hadoop.apache.org > Subject: How to run multiple data nodes and multiple task trackers on single > server. > > Hi: > I am following the guides in > http://hadoop-karma.blogspot.com/2010/05/hadoop-cookbook-4-how-to-run-multiple.html > with Hadoop 0.20.203.0 on Solaris 10u9 X64, but I have revised the > hadoop-daemon.sh call to > > hadoop-daemon.sh $1 datanode $DN_CONF_OPTS > > But I failed with error messages as following: > > > -bash-3.00$ run-additionalDN.sh start 1 > starting datanode, logging to > /data/hdp/additionalDN/1/logs/hadoop-gpadmin-datanode-solsvr2.out > Usage: java DataNode > [-rollback] > > Regards, > > Xiaobo Gu
