@zhangyy
2021-05-26T16:35:42.000000Z
字数 4851
阅读 117
协作框架
Flume是Cloudera提供的一个高可用的,高可靠的,分布式的海量日志采集、聚合和传输的系统,Flume支持在日志系统中定制各类数据发送方,用于收集数据;同时,Flume提供对数据进行简单处理,并写到各种数据接受方(可定制)的能力。
当前Flume有两个版本Flume 0.9X版本的统称Flume-og,Flume1.X版本的统称Flume-ng。由于Flume-ng经过重大重构,与Flume-og有很大不同,使用时请注意区分。
系统:Centos7.9x64
主机名:
cat /etc/hosts
----
192.168.100.11 node01.flyfish.cn
192.168.100.12 node02.flyfish.cn
192.168.100.13 node03.flyfish.cn
192.168.100.14 node04.flyfish.cn
192.168.100.15 node05.flyfish.cn
192.168.100.16 node06.flyfish.cn
192.168.100.17 node07.flyfish.cn
192.168.100.18 node08.flyfish.cn
----
在node01.flyfish.cn 节点上面执行:
上传 apache-flume-1.9.0-bin.tar.gz 的 包到 /opt/bigdata 解压
tar -zxvf apache-flume-1.9.0-bin.tar.gz
mv apache-flume-1.9.0-bin /opt/bigdata/flume
cd /opt/bigdata/flume/conf
cp -p flume-env.sh.template flume-env.sh
echo "JAVA_HOME=/opt/bigdata/jdk" >> flume-env.sh
vim /etc/profile
-----
加上flume的环境变量
#### flume
export FLUME_HOME=/opt/bigdata/flume
PATH=$PATH:$HOME/bin:$FLUME_HOME/bin:$FLUME_HOME/sbin
----
source /etc/profile
Flume-ng version
cd /opt/bigdata/flume/conf
vim test-flume.properties
---
# example.conf: A single-node Flume configuration
# Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the source
a1.sources.r1.type = netcat
a1.sources.r1.bind = localhost
a1.sources.r1.port = 44444
# Describe the sink
a1.sinks.k1.type = logger
# Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100
# Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
----
yum install -y telnet-* netcat-*
运行一个agent 实例
cd /opt/bigdata/flume/
bin/flume-ng agent --conf conf --conf-file conf/test-flume.properties --name a1 -Dflume.root.logger=INFO,console
测试:
telnet localhost 44444
验证:
Flume-ng最大的改动就是不再有分工角色设置,所有的都是agent,可以彼此之间相连,多个agent连到一个agent,此agent也就相当于collector了,NG也支持负载均衡.
由node02.flyfish.cn和node03.flyfish.cn收集日志信息,传给node01.flyfish.cn,再由node01.flyfish.cn上传到hdfs上
打包 node01.flyfish 节点的flume
cd /opt/bigdata/
tar -zcvf flume.tar.gz flume
scp flume.tar.gz root@node02.flyfish.cn:/opt/bigdata/
scp flume.tar.gz root@node03.flyfish.cn:/opt/bigdata/
node02.flyfish.cn与node03.flyfish.cn 上面配置
cd /opt/bigdata/
tar -zxvf flume.tar.gz
cd /opt/bigdata/flume/conf
vim slave.conf
-----
# 主要作用是监听目录中的新增数据,采集到数据之后,输出到avro (输出到agent)
# 注意:Flume agent的运行,主要就是配置source channel sink
# 下面的a1就是agent的代号,source叫r1 channel叫c1 sink叫k1
a1.sources = r1
a1.sinks = k1
a1.channels = c1
#具体定义source
a1.sources.r1.type = spooldir
#先创建此目录,保证里面空的
a1.sources.r1.spoolDir = /opt/bigdata/flume/logs
#对于sink的配置描述 使用avro日志做数据的消费
a1.sinks.k1.type = avro
# hostname是最终传给的主机名称或者ip地址
a1.sinks.k1.hostname = node01.flyfish.cn
a1.sinks.k1.port = 44444
#对于channel的配置描述 使用文件做数据的临时缓存 这种的安全性要高
a1.channels.c1.type = file
a1.channels.c1.checkpointDir = /opt/bigdata/flume/checkpoint
a1.channels.c1.dataDirs = /opt/bigdata/flume/data
#通过channel c1将source r1和sink k1关联起来
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
配置flume 的master 端:node01.flyfish.cn
cd /opt/bigdata/flume/conf
vim master.conf
----
# 获取slave1,2上的数据,聚合起来,传到hdfs上面
# 注意:Flume agent的运行,主要就是配置source channel sink
# 下面的a1就是agent的代号,source叫r1 channel叫c1 sink叫k1
a1.sources = r1
a1.sinks = k1
a1.channels = c1
#对于source的配置描述 监听avro
a1.sources.r1.type = avro
# hostname是最终传给的主机名称或者ip地址
a1.sources.r1.bind = node01.flyfish.cn
a1.sources.r1.port = 44444
#定义拦截器,为消息添加时间戳
a1.sources.r1.interceptors = i1
a1.sources.r1.interceptors.i1.type = org.apache.flume.interceptor.TimestampInterceptor$Builder
#对于sink的配置描述 传递到hdfs上面
a1.sinks.k1.type = hdfs
#集群的nameservers名字
#单节点的直接写:hdfs://192.168.100.11:8020
#ns是hadoop集群名称 [这个地方前提已经搭好了hadoop2.7.7]
a1.sinks.k1.hdfs.path = hdfs://192.168.100.11:8020/flume-test/%Y%m%d
a1.sinks.k1.hdfs.filePrefix = events-
a1.sinks.k1.hdfs.fileType = DataStream
#不按照条数生成文件
a1.sinks.k1.hdfs.rollCount = 0
#HDFS上的文件达到128M时生成一个文件
a1.sinks.k1.hdfs.rollSize = 134217728
#HDFS上的文件达到60秒生成一个文件
a1.sinks.k1.hdfs.rollInterval = 60
#对于channel的配置描述 使用内存缓冲区域做数据的临时缓存
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100
#通过channel c1将source r1和sink k1关联起来
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
----
node01.flyfish.cn:
cd /opt/bigdata/flume/
mkdir logs
nohup bin/flume-ng agent -n a1 -c conf -f conf/master.conf -Dflume.root.logger=INFO,console >> flume.logs &
node02.flyfish.cn与node03.flyfish.cn
cd /opt/bigdata/flume/
mkdir logs
nohup bin/flume-ng agent -n a1 -c conf -f conf/slave.conf -Dflume.root.logger=INFO,console >> flume.logs &
node01.flyfish.cn:
hdfs dfs -mkdir /flume-test/
hdfs dfs -chmod 777 /flume-test/
node02.flyfish.cn:
vim test-flume.txt
-----
11111
22222
33333
44444
55555
-----
cp -p test-flume.txt /opt/bigdata/flume/logs
去hdfs的页面上查看
将这个数据download下来
node01.flyfish.cn:
hdfs dfs -get /flume-test/20210525
cat events-.1621928807491