solve:

1. Look at Hadoop logs:

Check the namenode log: tail -n 200 hadoop-xinjie-namenode-VM-0-9-centos.log (location of file directory: Hadoop installation location logs file)

2. It is found that the port is occupied

3. Command to check the port occupancy: netstat -anp|grep 9866

4. Kill process: kill -9 9866

5. Restart the cluster after killing all the occupied ports. The problem is solved