There is no namenode problem with master jps

Execute the command / opt/modules/app/hadoop-2.7.4/sbin/start-all.sh on the primary node machine (equivalent to / opt/modules/app/hadoop-2.5.0/sbin/start-dfs.sh and / opt/modules/app/hadoop-2.5.0/sbin/start-yarn.sh)

Then jps found that there was no namenode and checked the logs log.

Found that namenode was not formatted

So execute the command to format the namenode node, and note that all the files under the original NameNode and DataNode must be deleted each time, otherwise an error will be reported. The directories where NameNode and DataNode are located are configured with hadoop.tmp.dir, dfs.namenode.name.dir and dfs.datanode.data.dir attributes in core-site.xml.

/ opt/modules/app/hadoop-2.7.4/sbin/start-all.sh

Then check to see if there are any new files in the tmp directory (my namenode has been reformatted several times before the new file appears), as shown in the figure

. . . Once upon a time, horses and carriages were slow, letters were far away, and there was only one person I could love all my life.