After the Hadoop environment is deployed, how can I solve the problem of running jps on the slave machine without seeing the Datanode process ?, Hadoopjps
Problem description: After the Hadoop environment is deployed, you cannot see the Datanode process when running jps on the slave machine.
Work und: delete all contents in the Datanode folder configured in the hdfs-site.xml on all slave machines (dfs. data. dir parameter), initialize Namenode, run
Hdfs namenode-format
Reinitialize.
Cause: the Namenode is initialized multiple times, but the master does not clear the initialization data in all the Datanode folders, causing the IDS generated in the two folders to be inconsistent, delete the initialization data in the Datanode folder and then initialize Namenode. (Usually the/current file under/dfs/data)
Start Hadoop again and use jps to see the Datanode process.