1.启动hadoop, start-all.sh 启动hadoop2. copy local example data to hdfs bin/hadoop jar contrib/streaming/hadoop-*streaming*.jar -file /home/hduser/reducer.py -reducer /home/hduser/reducer.py bin/hadoop dfs -cat /user/hduser/gutenberg-output/part-00000【阅读全文】
hadoop namenode -formate后,上传文件报错 0 datanode(s) running and no node(s) are excluded 1.报错内容: ?put: File /user/username/testData/matrixdata.txt._COPYING_ could only be replicated to 0 nodes instead of minReplication (=1). There are 0 datanode(s) running and no node(s) are excluded in this operation. 2.错误分析与解决 ?打开hdfs-site.xml里配置的datanode和namenode对应的目录,分别打开current文件夹里的VERSION,可以看到clusterID项正如日志里记录的一样,确实不一致,修改datanode里VERSION文件的clusterID 与namenode里的一致,再重新启动dfs(执行start-dfs.sh)再执行jps命令可以看【阅读全文】
现象:在访问Master:50070之后,点击browse the filesystem后,该页无法显示。<br /><br />原因:点击browse the filesystem后,网页转向的地址用的是hadoop集群的某一个datanode的主机名,由于客户端的浏览器无法解析这个主机名,因此该页无法显示。<br /><br />解决:需要在客户端的hosts文件里加入hadoop集群的ip地...【阅读全文】