Scala + Thrift+ Zookeeper+Flume+Kafka配置笔记

1. 开发环境

1.1. 软件包下载

1.1.1. JDK下载地址

http://www.oracle.com/technetwork/java/javase/downloads/jdk8-downloads-2133151.html

安装到 D:GreenSoftwareJavaJava8X64jdk1.8.0_91 目录

1.1.2. Maven下载地址

https://maven.apache.org/download.cgi

解压到 D:GreenSoftwareapache-maven-3.3.9 目录

1.1.3. Scala下载地址

https://www.scala-lang.org/download/

解压到 D:GreenSoftwareJavascala-2.12.6 目录

1.1.4. Thrift下载地址

http://thrift.apache.org/download

将下载的thrift-0.11.0.exe文件放到 D:ProjectServiceMiddleWare hrift目录下，并重命名为thrift.exe

1.1.5. Zookeeper

https://mirrors.tuna.tsinghua.edu.cn/apache/zookeeper/current/

解压到 D:ProjectServiceMiddleWarezookeeper-3.4.10目录

1.1.6. Flume下载地址

http://flume.apache.org/download.html

解压到 D:ProjectServiceMiddleWareflume-1.8.0目录

1.1.7. Kafka下载地址

http://kafka.apache.org/downloads

解压到 D:ProjectServiceMiddleWarekafka_2.12-1.1.0 目录

1.2. JDK+Maven+Scala+Thrift

1.2.1. 环境变量

JAVA_HOME D:GreenSoftwareJavaJava8X64jdk1.8.0_91

CLASSPATH .;%JAVA_HOME%libdt.jar;%JAVA_HOME%lib ools.jar

MAVEN_HOME D:GreenSoftwareapache-maven-3.3.9

SCALA_HOME D:GreenSoftwareJavascala-2.12.6

PATH 中加入

%JAVA_HOME%in;

%MAVEN_HOME%in;

%SCALA_HOME%in;

D:ProjectServiceMiddleWare hrift;

1.3. 安装、配置及验证

1.3.1. JDK

C:Userszyx>java -version

java version "1.8.0_91"

Java(TM) SE Runtime Environment (build 1.8.0_91-b14)

Java HotSpot(TM) 64-Bit Server VM (build 25.91-b14, mixed mode)

C:Userszyx>

1.3.2. Maven

C:Userszyx>mvn -version

Apache Maven 3.3.9 (bb52d8502b132ec0a5a3f4c09453c07478323dc5; 2015-11-11T00:41:47+08:00)

Maven home: D:GreenSoftwareapache-maven-3.3.9in..

Java version: 1.8.0_91, vendor: Oracle Corporation

Java home: D:GreenSoftwareJavaJava8X64jdk1.8.0_91jre

Default locale: zh_CN, platform encoding: GBK

OS name: "windows 10", version: "10.0", arch: "amd64", family: "dos"

1.3.3. Scala

C:Userszyx>scala -version

Scala code runner version 2.12.6 -- Copyright 2002-2018, LAMP/EPFL and Lightbend, Inc.

C:Userszyx>scala

Welcome to Scala 2.12.6 (Java HotSpot(TM) 64-Bit Server VM, Java 1.8.0_91).

Type in expressions for evaluation. Or try :help.

scala> :quit

C:Userszyx>

1.3.4. Thrift

C:Userszyx>thrift -version

Thrift version 0.11.0

1.3.5. Zookeeper

1.3.5.1. 配置

在D:ProjectServiceMiddleWarezookeeper-3.4.10conf目录下创建一个zoo.cfg文件，内容如下

tickTime=2000

dataDir=D:/Project/ServiceMiddleWare/zookeeper-3.4.10/data/db

dataLogDir=D:/Project/ServiceMiddleWare/zookeeper-3.4.10/data/log

clientPort=2181

# Zookeeper Cluster

# server.1=127.0.0.1:12888:1388

# server.2=127.0.0.1:12889:1389

# server.3=127.0.0.1:12887:1387

1.3.5.2. 启动Zookeeper

D:ProjectServiceMiddleWarezookeeper-3.4.10inzkServer.cmd

1.3.5.3. 启动客户端

D:ProjectServiceMiddleWarezookeeper-3.4.10inzkCli.cmd -server 127.0.0.1:2181

1.3.5.4. 基本命令

查看目录

ls /

1.3.5.5. 创建节点

create /config 0

1.3.5.6. 删除节点

delete /config

1.3.5.7. 退出客户端

quit

1.3.6. Kafka

1.3.6.1. 配置

1.3.6.1.1. Kafka的配置文件

修改D:ProjectServiceMiddleWarekafka_2.12-1.1.0config目录下的server.properties文件，内容如下

broker.id=0

port=9092

host.name=127.0.0.1

# listeners=PLAINTEXT://127.0.0.1:9092

# register zookeeper’s node data

# advertised.listeners=PLAINTEXT://127.0.0.1:9092

# log.dirs=/tmp/kafka-logs

log.dirs=D:/Project/ServiceMiddleWare/kafka_2.12-1.1.0/data/log

log.dir = D:/Project/ServiceMiddleWare/kafka_2.12-1.1.0/data/log

# zookeeper.connect=localhost:2181

zookeeper.connect=127.0.0.1:2181

# Zookeeper Cluster

# zookeeper.connect=127.0.0.1:2181,127.0.0.1:2182,127.0.0.1:2183

1.3.6.1.2. 日志的配置

修改D:ProjectServiceMiddleWarekafka_2.12-1.1.0config目录下的log4j.properties文件，内容如下

kafka.logs.dir=D:/Project/ServiceMiddleWare/kafka_2.12-1.1.0/data/log

log.dir = /tmp/kafka-logs

1.3.6.1.3. 端口配置（如果同一台服务器配置多个Kafka，则需要）

修改D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindows目录下的kafka-run-class.bat文件，在文件开头增加内容如下

set JMX_PORT=19093

set JAVA_DEBUG_PORT =5005

1.3.6.2. 启动Kafka

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-server-start.bat D:ProjectServiceMiddleWarekafka_2.12-1.1.0configserver.properties

1.3.6.3. 创建主题(Topic)

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-topics.bat --create --zookeeper 127.0.0.1:2181 --replication-factor 1 --partitions 1 --topic testTopic

1.3.6.4. 创建生产者(Producer)

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-console-producer.bat --broker-list 127.0.0.1:9092 --topic testTopic

1.3.6.5. 创建消费者(Consumer)

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-console-consumer.bat --zookeeper 127.0.0.1:2181 --topic testTopic

1.3.6.6. 生成者发送消息

Hello world

你好

1.3.6.7. 消费者收到消息

1.3.6.8. 查看主题(Topic)状态

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-topics.bat --describe --zookeeper 127.0.0.1:2181 --topic testTopic

1.3.6.9. 彻底删除topic

https://blog.csdn.net/odailidong/article/details/61615554

1.3.7. Flume

1.3.7.1. 配置

1.3.7.1.1. Flume的配置文件

在D:ProjectServiceMiddleWareflume-1.8.0conf目录下创建一个flume.agent1.conf文件，内容如下：

# flume.agent1.conf: A single-node Flume listen netcat configuration

# Name the components on this agent

agent1.sources = sources1

agent1.sinks = sinks1

agent1.channels = channels1

# Describe/configure the source

# agent1.sources.sources1.type = Avro, Exec, HTTP, JMS, Netcat, Sequence generator, Spooling directory, Syslog, Thrift, Twitter

agent1.sources.sources1.type = netcat

agent1.sources.sources1.bind = 127.0.0.1

agent1.sources.sources1.port = 44444

agent1.sources.source1.interceptors = interceptor1

# agent1.sources.source1.interceptors.interceptor1.type = Host, Morphline, Regex extractor, Regex filtering, Static, Timestamp, UUID

agent1.sources.source1.interceptors.interceptor1.type = timestamp

# Describe the sink

# agent1.sinks.sinks1.type = Avro, Elasticsearch, File roll, HBase, HDFS, IRC, Logger, Morphline(Solor), Null, Thrift

agent1.sinks.sinks1.type = logger

# Use a channel which buffers events in memory

# agent1.channels.channels1.type = File, JDBC, Memory

agent1.channels.channels1.type = memory

agent1.channels.channels1.capacity = 1000

agent1.channels.channels1.transactionCapacity = 100

# Bind the source and sink to the channel

agent1.sources.sources1.channels = channels1

agent1.sinks.sinks1.channel = channels1

1.3.7.1.2. 日志的配置

修改D:ProjectServiceMiddleWareflume-1.8.0conf目录下的log4j.properties文件，内容如下

# flume.log.dir=./logs

flume.log.dir=D:/Project/ServiceMiddleWare/flume-1.8.0/data/log

1.3.7.2. 启动Flume

D:ProjectServiceMiddleWareflume-1.8.0influme-ng.cmd agent -conf D:/Project/ServiceMiddleWare/flume-1.8.0/conf -conf-file D:/Project/ServiceMiddleWare/flume-1.8.0/conf/flume.agent1.conf -name agent1 -property flume.root.logger=INFO,console

或

cd D:ProjectServiceMiddleWareflume-1.8.0in

flume-ng.cmd agent -conf ../conf -conf-file ../conf/flume.agent1.conf -name agent1 -property flume.root.logger=INFO,console

1.3.7.3. 访问44444端口，模拟数据输入

telnet 127.0.0.1 44444

然后输入任意内容

查看Flume接收结果：

1.3.8. Flume与Kafka连接

1.3.8.1. Flume的配置

在D:ProjectServiceMiddleWareflume-1.8.0conf目录下修改flume.agent1.conf文件，内容如下：

# flume.agent1.conf: A single-node Flume listen netcat configuration

# Name the components on this agent

agent1.sources = sources1

agent1.sinks = sinks1

agent1.channels = channels1

# Describe/configure the source

# agent1.sources.sources1.type = Avro, Exec, HTTP, JMS, Netcat, Sequence generator, Spooling directory, Syslog, Thrift, Twitter

agent1.sources.sources1.type = netcat

agent1.sources.sources1.bind = 127.0.0.1

agent1.sources.sources1.port = 44444

agent1.sources.source1.interceptors = interceptor1

# agent1.sources.source1.interceptors.interceptor1.type = Host, Morphline, Regex extractor, Regex filtering, Static, Timestamp, UUID

agent1.sources.source1.interceptors.interceptor1.type = timestamp

# Describe the sink

# agent1.sinks.sinks1.type = Avro, Elasticsearch, File roll, HBase, HDFS, IRC, Logger, Morphline(Solor), Null, Thrift

# agent1.sinks.sinks1.type = logger

agent1.sinks.sinks1.type = org.apache.flume.sink.kafka.KafkaSink

agent1.sinks.sinks1.kafka.topic = testTopic

agent1.sinks.sinks1.kafka.bootstrap.servers = 127.0.0.1:9092

agent1.sinks.sinks1.kafka.flumeBatchSize = 20

agent1.sinks.sinks1.kafka.producer.acks = 1

agent1.sinks.sinks1.kafka.producer.linger.ms = 1

agent1.sinks.sinks1.kafka.producer.compression.type = snappy

# Use a channel which buffers events in memory

# agent1.channels.channels1.type = File, JDBC, Memory

agent1.channels.channels1.type = memory

agent1.channels.channels1.capacity = 1000

agent1.channels.channels1.transactionCapacity = 100

# Bind the source and sink to the channel

agent1.sources.sources1.channels = channels1

agent1.sinks.sinks1.channel = channels1

1.3.8.2. 启动Zookeeper

D:ProjectServiceMiddleWarezookeeper-3.4.10inzkServer.cmd

1.3.8.3. 启动Kafka

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-server-start.bat D:ProjectServiceMiddleWarekafka_2.12-1.1.0configserver.properties

1.3.8.4. 创建消费者(Consumer)

D:ProjectServiceMiddleWarekafka_2.12-1.1.0inwindowskafka-console-consumer.bat --zookeeper 127.0.0.1:2181 --topic testTopic

1.3.8.5. 启动Flume

D:ProjectServiceMiddleWareflume-1.8.0influme-ng.cmd agent -conf D:/Project/ServiceMiddleWare/flume-1.8.0/conf -conf-file D:/Project/ServiceMiddleWare/flume-1.8.0/conf/flume.agent1.conf -name agent1 -property flume.root.logger=INFO,console

1.3.8.6. 启动Flume的数据生产者（访问44444端口，模拟数据输入）

telnet 127.0.0.1 44444

然后输入任意内容

1.3.8.7. 查看kafka中消费者收到消息结果