Flume组件安装配置

1、下载和解压 Flume

# 传Flume安装包
[root@master ~]# cd /opt/software/
[root@master software]# ls
apache-flume-1.6.0-bin.tar.gz  hadoop-2.7.1.tar.gz     jdk-8u152-linux-x64.tar.gz  mysql-5.7.18.zip                 sqoop-1.4.7.bin__hadoop-2.6.0.tar.gz
apache-hive-2.0.0-bin.tar.gz   hbase-1.2.1-bin.tar.gz  mysql-5.7.18                mysql-connector-java-5.1.46.jar  zookeeper-3.4.8.tar.gz

# 使用root用户解压Flume到“/usr/local/src”路径
[root@master software]# tar xf /opt/software/apache-flume-1.6.0-bin.tar.gz -C /usr/local/src/

# 修改Flume安装路径文件夹名称
[root@master software]# cd /usr/local/src
[root@master src]# mv apache-flume-1.6.0-bin flume

# 修改文件夹归属用户和归属组为hadoop用户和hadoop组
[root@master src]# chown -R hadoop.hadoop /usr/local/src/

2、Flume 组件部署

# 编辑系统环境变量配置文件
[root@master src]# vi /etc/profile.d/flume.sh
添加：
export FLUME_HOME=/usr/local/src/flume
export PATH=${FLUME_HOME}/bin:$PATH

# 切换hadoop用户
[root@master src]# su - hadoop

# 查看是否成功
[hadoop@master ~]$ echo $PATH
/usr/local/src/zookeeper/bin:/usr/local/src/sqoop/bin:/usr/local/src/hbase/bin:/usr/local/src/jdk/bin:/usr/local/src/hadoop/bin:/usr/local/src/hadoop/sbin:/usr/local/src/flume/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/usr/local/src/hive/bin:/home/hadoop/.local/bin:/home/hadoop/bin

# 修改hbase-env.sh文件
[hadoop@master ~]$ vim /usr/local/src/hbase/conf/hbase-env.sh
#export HBASE_CLASSPATH=/usr/local/src/hadoop/etc/hadoop/ 注释掉这一行的内容

# 拷贝 flume-env.sh.template 文件
[hadoop@master ~]$ cd /usr/local/src/flume/conf
[hadoop@master conf]$ cp flume-env.sh.template flume-env.sh

# 修改并配置 flume-env.sh 文件
[hadoop@master conf]$ vi flume-env.sh
修改：
export JAVA_HOME=/usr/local/src/jdk

# 启动hadoop
[hadoop@master conf]$ start-all.sh
This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh
Starting namenodes on [master]
hadoop@master's password: 
master: namenode running as process 50448. Stop it first.
192.168.100.30: datanode running as process 43460. Stop it first.
192.168.100.20: datanode running as process 46094. Stop it first.
Starting secondary namenodes [0.0.0.0]
hadoop@0.0.0.0's password: 
0.0.0.0: secondarynamenode running as process 50670. Stop it first.
starting yarn daemons
resourcemanager running as process 50836. Stop it first.
192.168.100.30: nodemanager running as process 43584. Stop it first.
192.168.100.20: nodemanager running as process 46228. Stop it first.

# 执行以上命令后要确保master上有NameNode、SecondaryNameNode、ResourceManager进程，在slave节点上要能看到DataNode、NodeManager进程

# master节点查看
[hadoop@master conf]$ jps
50448 NameNode
50836 ResourceManager
47502 QuorumPeerMain
50670 SecondaryNameNode
55855 Jps

# slave1节点查看
[root@slave1 ~]# su - hadoop
[hadoop@slave1 ~]$ jps
2070 DataNode
2364 Jps
2191 NodeManager
[hadoop@slave1 ~]$ 

# slave2节点查看
[root@slave2 ~]# su - hadoop
[hadoop@slave2 ~]$ jps
2166 NodeManager
2055 DataNode
2345 Jps
[hadoop@slave2 ~]$ 

# 验证安装是否成功
[hadoop@master conf]$ flume-ng version
Flume 1.6.0
Source code repository: https://git-wip-us.apache.org/repos/asf/flume.git
Revision: 2561a23240a71ba20bf288c7c2cda88f443c2080
Compiled by hshreedharan on Mon May 11 11:15:44 PDT 2015
From source with checksum b29e416802ce9ece3269d34233baf43f
# 若能够正常查询 Flume 组件版本为1.6.0，则表示安装成功。

3、使用 Flume 发送和接受信息

# 在 Flume 安装目录中创建 simple-hdfs-flume.conf 文件
[hadoop@master conf]$ cd /usr/local/src/flume
[hadoop@master flume]$ vi simple-hdfs-flume.conf
a1.sources=r1
a1.sinks=k1
a1.channels=c1
a1.sources.r1.type=spooldir
a1.sources.r1.spoolDir=/usr/local/src/hadoop/logs
a1.sources.r1.fileHeader=true
a1.sources.r1.deserializer.maxLineLength=30000
a1.sinks.k1.type=hdfs
a1.sinks.k1.hdfs.path=hdfs://master:9000/tmp/flume
a1.sinks.k1.hdfs.rollsize=1024000
a1.sinks.k1.hdfs.rollCount=0
a1.sinks.k1.hdfs.rollInterval=900
a1.sinks.k1.hdfs.useLocalTimeStamp=true
a1.channels.c1.type=file
a1.channels.c1.capacity=10000
a1.channels.c1.transactionCapacity=1000
a1.sources.r1.channels=c1
a1.sinks.k1.channel=c1

# 删除/tmp
[hadoop@master flume]$ hdfs dfs -rm -r /tmp
22/05/08 22:16:44 INFO fs.TrashPolicyDefault: Namenode trash configuration: Deletion interval = 0 minutes, Emptier interval = 0 minutes.
Deleted /tmp

# 创建/tmp/flume
[hadoop@master flume]$ hdfs dfs -mkdir -p /tmp/flume

# 查看文件
[hadoop@master flume]$ hdfs dfs -ls /
Found 5 items
drwxr-xr-x   - hadoop supergroup          0 2022-04-29 15:26 /hbase
drwxr-xr-x   - hadoop supergroup          0 2022-04-29 11:59 /input
drwxr-xr-x   - hadoop supergroup          0 2022-04-29 12:00 /output
drwxr-xr-x   - hadoop supergroup          0 2022-05-08 22:16 /tmp
drwxr-xr-x   - hadoop supergroup          0 2022-04-29 16:49 /user

# 使用 flume-ng agent 命令加载 simple-hdfs-flume.conf 配置信息，启动 flume 传输数据
[hadoop@master flume]$ flume-ng agent --conf-file simple-hdfs-flume.conf --name a1
22/05/08 22:17:56 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471497.tmp
22/05/08 22:17:56 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471497.tmp
22/05/08 22:17:56 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471497.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471497
22/05/08 22:17:56 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471498.tmp
22/05/08 22:17:56 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471498.tmp
22/05/08 22:17:56 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471498.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471498
22/05/08 22:17:56 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471499.tmp
22/05/08 22:17:56 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471499.tmp
22/05/08 22:17:56 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471499.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471499
22/05/08 22:17:56 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471500.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471500.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471500.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471500
22/05/08 22:17:57 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471501.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471501.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471501.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471501
22/05/08 22:17:57 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471502.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471502.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471502.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471502
22/05/08 22:17:57 INFO hdfs.BucketWriter: Creating hdfs://master:9000/tmp/flume/FlumeData.1652019471503.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Closing hdfs://master:9000/tmp/flume/FlumeData.1652019471503.tmp
22/05/08 22:17:57 INFO hdfs.BucketWriter: Renaming hdfs://master:9000/tmp/flume/FlumeData.1652019471503.tmp to hdfs://master:9000/tmp/flume/FlumeData.1652019471503

# 查看 Flume 传输到 HDFS 的文件
[hadoop@master flume]$ hdfs dfs -ls /tmp/flume
-rw-r--r--   2 hadoop supergroup       1329 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471536
-rw-r--r--   2 hadoop supergroup       1479 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471537
-rw-r--r--   2 hadoop supergroup       1360 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471538
-rw-r--r--   2 hadoop supergroup       1249 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471539
-rw-r--r--   2 hadoop supergroup       1349 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471540
-rw-r--r--   2 hadoop supergroup       1550 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471541
-rw-r--r--   2 hadoop supergroup       1241 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471542
-rw-r--r--   2 hadoop supergroup       1372 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471543
-rw-r--r--   2 hadoop supergroup       1362 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471544
-rw-r--r--   2 hadoop supergroup       1485 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471545
-rw-r--r--   2 hadoop supergroup      17253 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471546
-rw-r--r--   2 hadoop supergroup       1296 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471547
-rw-r--r--   2 hadoop supergroup       1285 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471548
-rw-r--r--   2 hadoop supergroup       1447 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471549
-rw-r--r--   2 hadoop supergroup       1363 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471550
-rw-r--r--   2 hadoop supergroup       1246 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471551
-rw-r--r--   2 hadoop supergroup       1366 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471552
-rw-r--r--   2 hadoop supergroup       1630 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471553
-rw-r--r--   2 hadoop supergroup       1250 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471554
-rw-r--r--   2 hadoop supergroup       1425 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471555
-rw-r--r--   2 hadoop supergroup       1329 2022-05-08 22:17 /tmp/flume/FlumeData.1652019471556
……
# 若能查看到 HDFS 上/tmp/flume 目录有传输的数据文件，则表示数据传输成功。

使用 flume-ng agent 命令加载 simple-hdfs-flume.conf 配置信息