五 flink 分布式集群搭建

1、准备3台服务器

  分别重命名服务器名称:

hostnamectl set-hostname node180
hostnamectl set-hostname node181
hostnamectl set-hostname node182

  

2、安装jdk-1.8.X

  文件解压路径:文件解压:

tar zxvf  jdk-8u161-linux-x64.gz  -C   /opt/module

  

  配置环境变量:

vi /etc/profile

JAVA_HOME=/opt/module/jdk1.8.0_161
JRE_HOME=$JAVA_HOME/jre
PATH=$PATH:$JAVA_HOME/bin:$JRE_HOME/bin
CLASSPATH=:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar:$JRE_HOME/lib/dt.jar
export JAVA_HOME JRE_HOME PATH CLASSPATH

  

 

 

 

3、linux局域网中,通过主机名进行通信

vi /etc/hosts

192.168.0.180 node180
192.168.0.181 node181
192.168.0.182 node182

  

利用ifconfig和iwconfig完成每台主机的IP配置,在终端即可通过ping 主机名的方式来验证相互之间是否联通。

 

5、flink

flink 下载:http://flink.apache.org/downloads.html 

 

 

 

源文件存放目录:/usr/software

  文件解压:

tar zxvf  flink-1.10.0-bin-scala_2.12.tgz  -C   /opt/module

  

  设置主从:

vi masters

node180:8081

  

vi slaves

node181

node182

  

vi flink-conf.yaml

  修改flink配置

#jobmanager.rpc.address: node180
high-availability: zookeeper
high-availability.zookeeper.quorum: node180:2181,node181:2181,node182:2181
high-availability.storageDir: hdfs:///flink/ha/
#high-availability.zookeeper.path.root: /flink
#high-availability.cluster-id: /flinkCluster
state.backend: filesystem
state.checkpoints.dir: hdfs:///flink/checkpoints
state.savepoints.dir: hdfs:///flink/checkpoints
state.savepoints.dir: hdfs:///flink/checkpoint

  

cd /opt/module

  拷贝安装节点包

scp -r  flink-1.10.0/ root@192.168.0.181:/opt/module

scp -r  flink-1.10.0/ root@192.168.0.182:/opt/module

  

配置环境变量

#flink
export FLINK_HOME=/opt/module/flink-1.10.0
export PATH=$PATH:$FLINK_HOME/bin

  

 

 

执行刷新配置命令:

source /etc/profile

   

启动flink 

./bin/start-cluster.sh

  

 

  

jps查看进程

 

 

 

 

  webui查看

http://192.168.0.180:8081

 

 

 

 

flink同步配置文件

scp -r flink-1.10.0/conf/. root@192.168.0.181:/opt/module/flink-1.10.0/conf
scp -r flink-1.10.0/conf/. root@192.168.0.182:/opt/module/flink-1.10.0/conf

  

 问题

1、org.apache.flink.core.fs.UnsupportedFileSystemSchemeException: Could not find a file system implementation for scheme 'hdfs'. The scheme is not directly supported by Flink and no Hadoop file system to support this scheme could be loaded.

 

 

flink 下lib 文件缺失jar(flink-shaded-hadoop-2-uber-2.6.5-10.0.jar)包导致,下载指定包上传到lib文件夹重新启动flink即可

 

posted @ 2020-03-20 18:16  qk523  阅读(441)  评论(0编辑  收藏  举报