[高分]Hadoop集群配置问题 Server at hadoop-master/192.168.83.128:49000 not available yet, Zzzzz...

Posted 2023-03-29

一台master，两台slave

只能看到一个datanode（master机器上的）两台slave datanode日志一直报错：
2014-07-06 10:11:23,652 INFO org.apache.hadoop.ipc.RPC: Server at hadoop-master/192.168.83.128:49000 not available yet, Zzzzz...
2014-07-06 10:11:25,658 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: hadoop-master/192.168.83.128:49000. Already tried 0 time(s).
2014-07-06 10:11:26,661 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: hadoop-master/192.168.83.128:49000. Already tried 1 time(s).
2014-07-06 10:11:27,665 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: hadoop-master/192.168.83.128:49000. Already tried 2 time(s).
2014-07-06 10:11:28,668 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: hadoop-master/192.168.83.128:49000. Already tried 3 time(s).
2014-07-06 10:11:29,672 INFO org.apache.hadoop.ipc.Client: Retrying connect to server: hadoop-

在slave上手动 telnet master的49000端口也打不开，
master机器上telne自己的49000端口是通的（也是为什么本机的datanode可用）

所有iptables 都关了
求大神指导
127.0.0.1 hadoop-master
192.168.83.129 hadoop-slave1
192.168.83.130 hadoop-slave2
127.0.0.1 hadoop-master localhost.localdomain localhost
::1 localhost6.localdomain6 localhost6
贴上hosts文件（master）

其中一台slave的 hosts

# that require network functionality will fail.
127.0.0.1 hadoop-slave1
192.168.83.128 hadoop-master
192.168.83.130 hadoop-slave2

参考技术A 错误有可能是多方面引起的，首先从服务器ping主 ping不通，查看下slave网络状态是否正常，其次检查下主从master，slave的配置情况吧追问

谢谢热心回答啊，

网络应该没有问题，telnet 其他端口比如22 是通的，ping肯定是通的

追答

晕，你这配置，主从配置要求是一样的。

你到底哪台是master？
一会127.0.0.1为master
一会又192.168.83.128为master。
看你的意思是128为主，129,130为从，host配置文件：
192.168.83.128 hadoop-master
192.168.83.129 hadoop-slave1
192.168.83.130 hadoop-slave2
三台机器配置文件都这么写即可。
注意127.0.0.1是文件默认的，可删除，放那不动也行。