Hadoop datanode using JPS to view the solution that can’t be started

Problem Description: after the Hadoop environment is deployed, the process of datanode cannot be seen when running JPS on the slave machine.

Solution: delete all HDFS on slave- site.xml All contents in the datanode folder configured in( dfs.data.dir Parameter), and then initialize the namenode to run

hadoop namenode -format

Reinitialize.

Reason: the namenode has been initialized many times, but the master has not cleared all the initialization data in the datanode folder, which makes the ID generated in the two folders inconsistent. After deleting the initialization data in the datanode folder, the namenode initialization will take effect. Start Hadoop again and use JPS to see the datanode process.


Read More: