Solution: set livenodes to 0 for hadoop under centos6.6

The most important thing is to check whether the host name and ip address under/etc/hosts match correctly. if this rule is excluded, follow these steps: first, disable the firewall (each machine sub ): serviceiptablesstop: disable the safemode mode: hadoopdfsadmin-safemodeleave. then, delete the haoop folder first.

The most important thing is to check whether the host name and ip address under/etc/hosts match correctly. if this rule is excluded, follow these steps:

Firewall shutdown (sub-computers): service iptables stop

Second, disable the safemode mode: hadoop dfsadmin-safemode leave

Then, delete the formatted files such as logs in the haoop folder and format them again: hdfs namenode-format

Finally, check whether the memory is sufficient, df-hl. if the memory is insufficient, increase the memory (knowledge about Baidu and Google linux memory ).