Flume：本地文件到HDFS

Flume下载地址

apache-flume-1.6.0-bin.tar.gz
http://pan.baidu.com/s/1o81nR8e s832

apache-flume-1.5.2-bin.tar.gz
http://pan.baidu.com/s/1bp6tXVL 4n4z

官网
https://flume.apache.org/download.html

配置文件
cd /usr/app/flume1.6/conf
vi flume-dirTohdfs.properties

#agent1 nameagent1.sources=source1agent1.sinks=sink1agent1.channels=channel1#Spooling Directory#set source1agent1.sources.source1.type=spooldiragent1.sources.source1.spoolDir=/usr/app/flumelog/dir/logdfsagent1.sources.source1.channels=channel1agent1.sources.source1.fileHeader = falseagent1.sources.source1.interceptors = i1agent1.sources.source1.interceptors.i1.type = timestamp#set sink1agent1.sinks.sink1.type=hdfsagent1.sinks.sink1.hdfs.path=/user/yuhui/flumeagent1.sinks.sink1.hdfs.fileType=DataStreamagent1.sinks.sink1.hdfs.writeFormat=TEXTagent1.sinks.sink1.hdfs.rollInterval=1agent1.sinks.sink1.channel=channel1agent1.sinks.sink1.hdfs.filePrefix=%Y-%m-%d#set channel1agent1.channels.channel1.type=fileagent1.channels.channel1.checkpointDir=/usr/app/flumelog/dir/logdfstmp/pointagent1.channels.channel1.dataDirs=/usr/app/flumelog/dir/logdfstmp

建立Linux目录

[root@hadoop11 app]# mkdir /usr/app/flumelog/dir
[root@hadoop11 app]# mkdir /usr/app/flumelog/dir/logdfs
[root@hadoop11 app]# mkdir /usr/app/flumelog/dir/logdfstmp
[root@hadoop11 app]# mkdir /usr/app/flumelog/dir/logdfstmp/point

建立Hadoop目录

[root@hadoop11 app]#hadoop fs -mkdir /user/yuhui/flume

启动配置文件

flume-ng agent -n agent1 -c conf -f ./flume-dirTohdfs.properties-Dflume.root.logger=DEBUG,console >./flume1.log 2>&1 &

查看启动进程：

建立数据测试

查看hdfs数据测试路径没有数据

将测试数据放入监控路径，Flume读完数据之后，文件会自动更名

查看hdfs数据测试路径，Flume输出的数据

备注
日志监控路径，只要日志放入则被Flume监控
[root@hadoop11 app]# mkdir /usr/app/flumelog/dir/logdfs

日志读取完毕存储路径，日志在这里则一直存储在Channel中（最多只有两个log-number日志，且默认达到1.6G之后删除前面一个log，建立新的log）
[root@hadoop11 app]# mkdir /usr/app/flumelog/dir/logdfstmp

日志监控路径中文件的路径存放点
[root@hadoop11 app]# mkdir /usr/app/flumelog/dir/logdfstmp/point