关于zookeeper集群搭建，无法选举领导错误

Cannot open channel to 2 at election address zookeeper2.example.com/192.168.3.162:3888
java.net.ConnectException: Connection refused (Connection refused)
	at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:350)
	at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:206)
	at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:188)
	at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
	at java.net.Socket.connect(Socket.java:589)
	at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:558)
	at org.apache.zookeeper.server.quorum.QuorumCnxManager.handleConnection(QuorumCnxManager.java:486)
	at org.apache.zookeeper.server.quorum.QuorumCnxManager.receiveConnection(QuorumCnxManager.java:386)
	at org.apache.zookeeper.server.quorum.QuorumCnxManager$Listener.run(QuorumCnxManager.java:754)

因为要搭建fabric1.4的，orderer节点，kafka共识，zookeeper注册，出现问题考虑几个原因

1.防火墙问题

service iptable  status

service firewalld status

若开启直接关闭

2.host问题

若是多机器物理环境，记得配置host

根据设置的docker网络模式修改

docker网络模式为host，直接物理机修改  vi   /etc/hosts

docker桥接网络，新增docker容器配置 

docker run -d --name test1 \
    --add-host test1.a:1.2.3.4 \
    local/test

docker-compose配置

test2:
  build: local/test
  extra_hosts:
    test1.a: 1.2.3.4
    test1.b: 4.3.2.1

比如这里采用的新增docker-compose配置

3.本机服务0.0.0.0绑定问题

本示例三台物理机zookeeper，每个本机的服务地址都修改为0.0.0.0即可，网络环境复杂，写成域名可能会导致接收不到消息，zookeeper1的配置对应修改为0.0.0.0，2的把域名修改为0.0.0.0，3的修改3的域名为0.0.0.0