我是hadoop新手，最近部署了2个分布式hadoop系统，由opentsdb运行。主机a和从机一起是主节点，主机b是从节点。在计算机a中编辑/etc/hosts：

127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6
10.94.212.144   node1
10.48.22.52     node2

我没有机器b的根权限，不能修改b的/etc/hosts，所以在b的core-site.xml中，　我写道：

<property>
     <name>fs.default.name</name>
     <value>hdfs://10.94.212.144:49000</value>
</property>

在a fs.default.name is node1:49000中，a是主节点，a，b都是从节点，当我运行start dfs.sh，namenode时，datanode1正常启动，但是datanode2启动失败，出现错误：

2016-11-15 12:42:43,103 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 0 time(s).
2016-11-15 12:42:44,106 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 1 time(s).
2016-11-15 12:42:45,109 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 2 time(s).
2016-11-15 12:42:46,112 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 3 time(s).
2016-11-15 12:42:47,115 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 4 time(s).
2016-11-15 12:42:48,117 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 5 time(s).
2016-11-15 12:42:49,120 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 6 time(s).
2016-11-15 12:42:50,123 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 7 time(s).
2016-11-15 12:42:51,125 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 8 time(s).
2016-11-15 12:42:52,128 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: /10.94.212.144:49000. Already tried 9 time(s).

是网络问题吗？但是当我用ssh连接a和b的时候就没关系了。还有什么原因呢？已经困扰我好几天了。。。