Ubuntu fails to log on properly due to an incorrect Ubuntu environment variable when Hadoop2.7.5 is installed,

Ubuntu fails to log on properly due to an incorrect Ubuntu environment variable when Hadoop2.7.5 is installed,

When configuring java environment variables, it should be noted that if other environment variables are damaged, many commands may be unavailable, and desktop access may fail. A solution is provided to solve the problem of abnormal login caused by incorrect Ubuntu environment variables.

The pseudo-distribution configuration contains the following two segments: usr/local/hadoop/tmp/dfs/data and usr/local/hadoop/tmp/dfs/name, after extracting the Hadoop tar package, there is no dfs folder in usr/local/hadoop, and there is no data or name folder. Therefore, you must first create these empty folders and do not create any errors in the location. The permissions of the data and name folders are very important. You need to grant the write permission because some files will be automatically written to these folders when namenode-format is executed later, if the folder does not have the write permission, an error is reported. The content is probably cannot create directory. For the solution, see Hadoop formatting namenode error: java. io. IOException: Cannot create directory. Pay special attention to the data and name files !!!

If you find that the namenode or datanode is not displayed after jps execution, first make sure that your configuration file is correct and then you need to reformat the namenode, then first execute the stop-dfs.sh, delete all the files in the logs folder. Delete the entire tmp folder created earlier, and then re-execute the entire step.

The above are all my problems and solutions, which are valid for test.