一、什麼是Flume?
flume 作爲 cloudera 開發的實時日誌收集系統，受到了業界的認可與廣泛應用。Flume 初始的發行版本目前被統稱爲 Flume OG（original generation），屬於 cloudera。但隨着 FLume 功能的擴展，Flume OG 代碼工程臃腫、核心組件設計不合理、核心配置不標準等缺點暴露出來，尤其是在 Flume OG 的最後一個發行版本 0.94.0 中，日誌傳輸不穩定的現象尤爲嚴重，爲了解決這些問題，2011 年
10 月 22 號，cloudera 完成了 Flume-728，對 Flume 進行了里程碑式的改動：重構核心組件、核心配置以及代碼架構，重構後的版本統稱爲 Flume NG（next generation）；改動的另一原因是將 Flume 納入 apache 旗下，cloudera Flume 改名爲 Apache Flume。
flume的特點：
flume是一個分佈式、可靠、和高可用的海量日誌採集、聚合和傳輸的系統。支持在日誌系統中定製各類數據發送方，用於收集數據;同時，Flume提供對數據進行簡單處理，並寫到各種數據接受方(比如文本、HDFS、Hbase等)的能力 。
flume的數據流由事件(Event)貫穿始終。事件是Flume的基本數據單位，它攜帶日誌數據(字節數組形式)並且攜帶有頭信息，這些Event由Agent外部的Source生成，當Source捕獲事件後會進行特定的格式化，然後Source會把事件推入(單個或多個)Channel中。你可以把Channel看作是一個緩衝區，它將保存事件直到Sink處理完該事件。Sink負責持久化日誌或者把事件推向另一個Source。
flume的可靠性
當節點出現故障時，日誌能夠被傳送到其他節點上而不會丟失。Flume提供了三種級別的可靠性保障，從強到弱依次分別爲：end-to-end（收到數據agent首先將event寫到磁盤上，當數據傳送成功後，再刪除；如果數據發送失敗，可以重新發送。），Store on failure（這也是scribe採用的策略，當數據接收方crash時，將數據寫到本地，待恢復後，繼續發送），Besteffort（數據發送到接收方後，不會進行確認）。
flume的可恢復性：
還是靠Channel。推薦使用FileChannel，事件持久化在本地文件系統裏(性能較差)。
flume的一些核心概念：
Agent使用JVM 運行Flume。每臺機器運行一個agent，但是可以在一個agent中包含多個sources和sinks。
Client生產數據，運行在一個獨立的線程。
Source從Client收集數據，傳遞給Channel。
Sink從Channel收集數據，運行在一個獨立線程。
Channel連接 sources 和 sinks ，這個有點像一個隊列。
Events可以是日誌記錄、 avro 對象等。
Flume以agent爲最小的獨立運行單位。一個agent就是一個JVM。單agent由Source、Sink和Channel三大組件構成，如下圖：
值得注意的是，Flume提供了大量內置的Source、Channel和Sink類型。不同類型的Source,Channel和Sink可以自由組合。組合方式基於用戶設置的配置文件，非常靈活。比如：Channel可以把事件暫存在內存裏，也可以持久化到本地硬盤上。Sink可以把日誌寫入HDFS, HBase，甚至是另外一個Source等等。Flume支持用戶建立多級流，也就是說，多個agent可以協同工作，並且支持Fan-in、Fan-out、Contextual Routing、Backup Routes，這也正是NB之處。如下圖所示:
二、flume的官方網站在哪裏？
三、在哪裏下載？
http://www.apache.org/dyn/closer.cgi/flume/1.5.0/apache-flume-1.5.0-bin.tar.gz
四、如何安裝？
1)將下載的flume包，解壓到/home/hadoop目錄中，你就已經完成了50%：）簡單吧
2)修改 flume-env.sh 配置文件,主要是JAVA_HOME變量設置
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop/flume-1.5.0-bin# cp conf/flume-env.sh.template conf/flume-env.sh</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop/flume-1.5.0-bin# vi conf/flume-env.sh</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># or more contributor license agreements. See the NOTICE file</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Licensed to the Apache Software Foundation (ASF) under one</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># distributed with this work for additional information</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># "License"); you may not use this file except in compliance</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># regarding copyright ownership. The ASF licenses this file</span></div># to you under the Apache License, Version 2.0 (the
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">#</span></div># with the License. You may obtain a copy of the License at
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># distributed under the License is distributed on an "AS IS" BASIS,</span></div># http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Enviroment variables can be set here.</span></div># See the License for the specific language governing permissions and
# limitations under the License.
# If this file is placed at FLUME_CONF_DIR/flume-env.sh, it will be sourced
# during Flume startup.
JAVA_HOME=/usr/lib/jvm/java-7-oracle
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">#FLUME_CLASSPATH=""</span></div># Give Flume more memory and pre-allocate, enable remote monitoring via JMX
#JAVA_OPTS="-Xms100m -Xmx200m -Dcom.sun.management.jmxremote"
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Note that the Flume conf directory is always included in the classpath.</span></div>
3)驗證是否安裝成功
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng version</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Flume 1.5.0</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Source code repository: https://git-wip-us.apache.org/repos/asf/flume.git</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Revision: 8633220df808c4cd0c13d1cf0320454a94f1ea97</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">From sourcewith checksum a01fe726e4380ba0c9f7a7d222db961f</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Compiled by hshreedharan on Wed May 7 14:49:18 PDT 2014</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop#</span></div>出現上面的信息，表示安裝成功了
五、flume的案例
1)案例1：Avro
Avro可以發送一個給定的文件給Flume，Avro 源使用AVRO RPC機制。
a)創建agent配置文件
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop#vi /home/hadoop/flume-1.5.0-bin/conf/avro.conf</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"> </span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources = r1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks = k1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"> </span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Describe/configure the source</span></div>a1.sources.r1.type= avro
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Use a channel which buffers events in memory</span></div>a1.sources.r1.bind = 0.0.0.0
a1.sources.r1.port = 4141
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks.k1.type= logger</span></div>
# Describe the sink
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Bind the source and sink to the channel</span></div>a1.channels.c1.type= memory
a1.channels.c1.capacity = 1000
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels.c1.transactionCapacity = 100</span></div>
a1.sources.r1.channels = c1
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks.k1.channel = c1</span></div>b)啓動flume agent a1
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng avro-client -c . -H m1 -p 4141 -F /home/hadoop/flume-1.5.0-bin/log.00f)在m1的控制檯，可以看到以下信息，注意最後一行：
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop/flume-1.5.0-bin/conf# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/avro.conf -n a1 -Dflume.root.logger=INFO,console</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Info: Sourcing environment configuration script/home/hadoop/flume-1.5.0-bin/conf/flume-env.sh</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Info: Excluding /home/hadoop/hadoop-2.2.0/share/hadoop/common/lib/slf4j-api-1.7.5.jar from classpath</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Info: Including Hadoop libraries found via (/home/hadoop/hadoop-2.2.0/bin/hadoop)for HDFS access</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">-08-10 10:43:25,112 (New I/Oworker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x92464c4f, /192.168.1.50:59850 :> /192.168.1.50:4141] UNBOUND</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">Info: Excluding /home/hadoop/hadoop-2.2.0/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar from classpath</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">...</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">-08-10 10:43:25,112 (New I/Oworker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x92464c4f, /192.168.1.50:59850 :> /192.168.1.50:4141] CLOSED</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">-08-10 10:43:25,112 (New I/Oworker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.channelClosed(NettyServer.java:209)] Connection to /192.168.1.50:59850 disconnected.</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">-08-10 10:43:26,718 (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.flume.sink.LoggerSink.process(LoggerSink.java:70)] Event: { headers:{} body: 68 65 6C 6C 6F 20 77 6F 72 6C 64 hello world }</span></div>2)案例2：Spool
Spool監測配置的目錄下新增的文件，並將文件中的數據讀取出來。需要注意兩點：
1) 拷貝到spool目錄下的文件不可以再打開編輯。
2) spool目錄下不可包含相應的子目錄
a)創建agent配置文件
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/spool.conf</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources = r1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks = k1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Describe/configure the source</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.type= spooldir</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.fileHeader =true</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.spoolDir =/home/hadoop/flume-1.5.0-bin/logs</span></div># Describe the sink
a1.sinks.k1.type= logger
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels.c1.transactionCapacity = 100</span></div># Use a channel which buffers events in memory
a1.channels.c1.type= memory
a1.channels.c1.capacity = 1000
# Bind the source and sink to the channel
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks.k1.channel = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div>b)啓動flume agent a1
EXEC執行一個給定的命令獲得輸出的源,如果要使用tail命令，必選使得file足夠大才能看到輸出內容
a)創建agent配置文件
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/exec_tail.conf</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources = r1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks = k1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Describe/configure the source</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.type= exec</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Describe the sink</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.command= tail -F /home/hadoop/flume-1.5.0-bin/log_exec_tail</span></div>a1.sinks.k1.type= logger
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels.c1.transactionCapacity = 100</span></div># Use a channel which buffers events in memory
a1.channels.c1.type= memory
a1.channels.c1.capacity = 1000
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks.k1.channel = c1</span></div># Bind the source and sink to the channel
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div>
b)啓動flume agent a1
Syslogtcp監聽TCP的端口做爲數據源
a)創建agent配置文件
a)創建agent配置文件
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/post_json.conf</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources = r1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks = k1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.type= org.apache.flume.source.http.HTTPSource</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Describe/configure the source</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.port = 8888</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;"># Use a channel which buffers events in memory</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div># Describe the sink
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels.c1.transactionCapacity = 100</span></div>a1.sinks.k1.type= logger
a1.channels.c1.type= memory
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.channels.c1.capacity = 1000</span></div># Bind the source and sink to the channel
<div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sinks.k1.channel = c1</span></div><div style="text-align: left;"><span style="font-family: Arial, Helvetica, sans-serif;">a1.sources.r1.channels = c1</span></div>
b)啓動flume agent a1
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/post_json.conf -n a1 -Dflume.root.logger=INFO,console c)生成JSON 格式的POST request
/
|
1
2
3
4
5
6
7
8
9
10
11
|
08/10 11:49:59 INFO node.Application: Starting Channel c1/08/10 11:49:59 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: CHANNEL, name: c1: Successfully registered new MBean./08/10 11:49:59 INFO instrumentation.MonitoredCounterGroup: Component type: CHANNEL, name: c1 started/08/10 11:49:59 INFO node.Application: Starting Sink k1/08/10 11:49:59 INFO node.Application: Starting Source r1/08/10 11:49:59 INFO mortbay.log: Logging to org.slf4j.impl.Log4jLoggerAdapter(org.mortbay.log) via org.mortbay.log.Slf4jLog/08/10 11:49:59 INFO mortbay.log: jetty-6.1.26/08/10 11:50:00 INFO mortbay.log: Started [email protected]:8888/08/10 11:50:00 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: SOURCE, name: r1: Successfully registered new MBean./08/10 11:50:00 INFO instrumentation.MonitoredCounterGroup: Component type: SOURCE, name: r1 started/08/10 12:14:32 INFO sink.LoggerSink: Event: { headers:{b=b1, a=a1} body: 69 64 6F 61 6C 6C 2E 6F 72 67 5F 62 6F 64 79 idoall.org_body } |
6)案例6：Hadoop sink
其中關於hadoop2.2.0部分的安裝部署，請參考文章《ubuntu12.04+hadoop2.2.0+zookeeper3.4.5+hbase0.96.2+hive0.13.1分佈式環境部署》
a)創建agent配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/hdfs_sink.confa1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= syslogtcp
a1.sources.r1.port = 5140a1.sources.r1.host = localhosta1.sources.r1.channels = c1# Describe the sinka1.sinks.k1.type= hdfs
a1.sinks.k1.channel = c1a1.sinks.k1.hdfs.path = hdfs://m1:9000/user/flume/syslogtcpa1.sinks.k1.hdfs.filePrefix = Sysloga1.sinks.k1.hdfs.round =truea1.sinks.k1.hdfs.roundValue = 10a1.sinks.k1.hdfs.roundUnit = minute# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100# Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
b)啓動flume agent a1
|
1
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/hdfs_sink.conf
-n a1 -Dflume.root.logger=INFO,console |
c)測試產生syslog
|
1
|
root@m1:/home/hadoop# echo "hello idoall flume -> hadoop testing one" | nc localhost 5140 |
d)在m1的控制檯，可以看到以下信息：
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
/08/10 12:20:39 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: CHANNEL, name: c1: Successfully registered new MBean./08/10 12:20:39 INFO instrumentation.MonitoredCounterGroup: Component type: CHANNEL, name: c1 started/08/10 12:20:39 INFO node.Application: Starting Sink k1/08/10 12:20:39 INFO node.Application: Starting Source r1/08/10 12:20:39 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: SINK, name: k1: Successfully registered new MBean./08/10 12:20:39 INFO instrumentation.MonitoredCounterGroup: Component type: SINK, name: k1 started/08/10 12:20:39 INFO source.SyslogTcpSource: Syslog TCP Source starting.../08/10 12:21:46 WARN source.SyslogUtils: Event created from Invalid Syslog data./08/10 12:21:49 INFO hdfs.HDFSSequenceFile: writeFormat = Writable, UseRawLocalFileSystem = false/08/10 12:21:49 INFO hdfs.BucketWriter: Creatinghdfs://m1:9000/user/flume/syslogtcp//Syslog.1407644509504.tmp/08/10 12:22:20 INFO hdfs.BucketWriter: Closinghdfs://m1:9000/user/flume/syslogtcp//Syslog.1407644509504.tmp/08/10 12:22:20 INFO hdfs.BucketWriter: Close tries incremented/08/10 12:22:20 INFO hdfs.BucketWriter: Renaminghdfs://m1:9000/user/flume/syslogtcp/Syslog.1407644509504.tmp tohdfs://m1:9000/user/flume/syslogtcp/Syslog.1407644509504/08/10 12:22:20 INFO hdfs.HDFSEventSink: Writer callback called. |
e)在m1上再打開一個窗口，去hadoop上檢查文件是否生成
|
1
2
3
4
5
|
root@m1:/home/hadoop# /home/hadoop/hadoop-2.2.0/bin/hadoop fs -ls /user/flume/syslogtcpFound 1 items
-rw-r--r-- 3 root supergroup 155 2014-08-10 12:22/user/flume/syslogtcp/Syslog.1407644509504root@m1:/home/hadoop# /home/hadoop/hadoop-2.2.0/bin/hadoop fs -cat /user/flume/syslogtcp/Syslog.1407644509504SEQ!org.apache.hadoop.io.LongWritable"org.apache.hadoop.io.BytesWritable^;>Gv$hello idoall flume -> hadoop testing one |
7)案例7：File Roll Sink
a)創建agent配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/file_roll.confa1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= syslogtcp
a1.sources.r1.port = 5555a1.sources.r1.host = localhosta1.sources.r1.channels = c1# Describe the sinka1.sinks.k1.type= file_roll
a1.sinks.k1.sink.directory =/home/hadoop/flume-1.5.0-bin/logs# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100# Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
b)啓動flume agent a1
|
1
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/file_roll.conf
-n a1 -Dflume.root.logger=INFO,console |
c)測試產生log
|
1
2
|
root@m1:/home/hadoop# echo "hello idoall.org syslog" | nc localhost 5555root@m1:/home/hadoop# echo "hello idoall.org syslog 2" | nc localhost 5555 |
d)查看/home/hadoop/flume-1.5.0-bin/logs下是否生成文件,默認每30秒生成一個新文件
|
1
2
3
4
5
6
7
8
9
10
|
root@m1:/home/hadoop# ll /home/hadoop/flume-1.5.0-bin/logs總用量 272 drwxr-xr-x 3 root root 4096 Aug 10 12:50 ./drwxr-xr-x 9 root root 4096 Aug 10 10:59 ../-rw-r--r-- 1 root root 50 Aug 10 12:49 1407646164782-1-rw-r--r-- 1 root root 0 Aug 10 12:49 1407646164782-2-rw-r--r-- 1 root root 0 Aug 10 12:50 1407646164782-3root@m1:/home/hadoop# cat /home/hadoop/flume-1.5.0-bin/logs/1407646164782-1 /home/hadoop/flume-1.5.0-bin/logs/1407646164782-2hello idoall.org sysloghello idoall.org syslog 2 |
8)案例8：Replicating Channel Selector
Flume支持Fan out流從一個源到多個通道。有兩種模式的Fan out，分別是複製和複用。在複製的情況下，流的事件被髮送到所有的配置通道。在複用的情況下，事件被髮送到可用的渠道中的一個子集。Fan out流需要指定源和Fan out通道的規則。
這次我們需要用到m1,m2兩臺機器
a)在m1創建replicating_Channel_Selector配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector.confa1.sources = r1
a1.sinks = k1 k2
a1.channels = c1 c2# Describe/configure the sourcea1.sources.r1.type= syslogtcp
a1.sources.r1.port = 5140a1.sources.r1.host = localhosta1.sources.r1.channels = c1 c2a1.sources.r1.selector.type= replicating
# Describe the sinka1.sinks.k1.type= avro
a1.sinks.k1.channel = c1a1.sinks.k1.hostname= m1
a1.sinks.k1.port = 5555a1.sinks.k2.type= avro
a1.sinks.k2.channel = c2a1.sinks.k2.hostname= m2
a1.sinks.k2.port = 5555# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100a1.channels.c2.type= memory
a1.channels.c2.capacity = 1000a1.channels.c2.transactionCapacity = 100 |
b)在m1創建replicating_Channel_Selector_avro配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector_avro.confa1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= avro
a1.sources.r1.channels = c1a1.sources.r1.bind = 0.0.0.0a1.sources.r1.port = 5555# Describe the sinka1.sinks.k1.type= logger
# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100# Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
c)在m1上將2個配置文件複製到m2上一份
|
1
2
|
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector.conf
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector_avro.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector_avro.conf<br> |
d)打開4個窗口，在m1和m2上同時啓動兩個flume agent
|
1
2
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector_avro.conf
-n a1 -Dflume.root.logger=INFO,console root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/replicating_Channel_Selector.conf
-n a1 -Dflume.root.logger=INFO,console |
e)然後在m1或m2的任意一臺機器上，測試產生syslog
|
1
|
root@m1:/home/hadoop# echo "hello idoall.org syslog" | nc localhost 5140 |
f)在m1和m2的sink窗口，分別可以看到以下信息,這說明信息得到了同步：
|
1
2
3
4
5
6
7
8
|
/08/10 14:08:18 INFO ipc.NettyServer: Connection to /192.168.1.51:46844 disconnected./08/10 14:08:52 INFO ipc.NettyServer: [id: 0x90f8fe1f, /192.168.1.50:35873 => /192.168.1.50:5555] OPEN/08/10 14:08:52 INFO ipc.NettyServer: [id: 0x90f8fe1f, /192.168.1.50:35873 => /192.168.1.50:5555] BOUND: /192.168.1.50:5555/08/10 14:08:52 INFO ipc.NettyServer: [id: 0x90f8fe1f, /192.168.1.50:35873 => /192.168.1.50:5555] CONNECTED: /192.168.1.50:35873/08/10 14:08:59 INFO ipc.NettyServer: [id: 0xd6318635, /192.168.1.51:46858 => /192.168.1.50:5555] OPEN/08/10 14:08:59 INFO ipc.NettyServer: [id: 0xd6318635, /192.168.1.51:46858 => /192.168.1.50:5555] BOUND: /192.168.1.50:5555/08/10 14:08:59 INFO ipc.NettyServer: [id: 0xd6318635, /192.168.1.51:46858 => /192.168.1.50:5555] CONNECTED: /192.168.1.51:46858/08/10 14:09:20 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 68 65 6C 6C 6F 20 69 64 6F 61 6C 6C 2E 6F 72 67 hello idoall.org } |
9)案例9：Multiplexing Channel Selector
a)在m1創建Multiplexing_Channel_Selector配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector.confa1.sources = r1
a1.sinks = k1 k2
a1.channels = c1 c2# Describe/configure the sourcea1.sources.r1.type= org.apache.flume.source.http.HTTPSourcea1.sources.r1.port = 5140a1.sources.r1.channels = c1 c2a1.sources.r1.selector.type= multiplexing
a1.sources.r1.selector.header =type#映射允許每個值通道可以重疊。默認值可以包含任意數量的通道。a1.sources.r1.selector.mapping.baidu = c1a1.sources.r1.selector.mapping.ali = c2a1.sources.r1.selector.default = c1# Describe the sinka1.sinks.k1.type= avro
a1.sinks.k1.channel = c1a1.sinks.k1.hostname= m1
a1.sinks.k1.port = 5555a1.sinks.k2.type= avro
a1.sinks.k2.channel = c2a1.sinks.k2.hostname= m2
a1.sinks.k2.port = 5555# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100a1.channels.c2.type= memory
a1.channels.c2.capacity = 1000a1.channels.c2.transactionCapacity = 100 |
b)在m1創建Multiplexing_Channel_Selector_avro配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector_avro.confa1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= avro
a1.sources.r1.channels = c1a1.sources.r1.bind = 0.0.0.0a1.sources.r1.port = 5555# Describe the sinka1.sinks.k1.type= logger
# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100# Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
c)將2個配置文件複製到m2上一份
|
1
2
|
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector.conf
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector_avro.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector_avro.conf |
d)打開4個窗口，在m1和m2上同時啓動兩個flume agent
|
1
2
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector_avro.conf
-n a1 -Dflume.root.logger=INFO,console root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Multiplexing_Channel_Selector.conf
-n a1 -Dflume.root.logger=INFO,console |
e)然後在m1或m2的任意一臺機器上，測試產生syslog
|
1
|
root@m1:/home/hadoop# curl -X POST -d '[{ "headers" :{"type" : "baidu"},"body" : "idoall_TEST1"}]'http://localhost:5140
&& curl -X POST -d '[{ "headers" :{"type" : "ali"},"body" : "idoall_TEST2"}]'http://localhost:5140 && curl -X POST -d '[{ "headers" :{"type" : "qq"},"body" : "idoall_TEST3"}]'http://localhost:5140 |
f)在m1的sink窗口，可以看到以下信息：
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
14/08/10 14:32:21 INFO node.Application: Starting Sink k114/08/10 14:32:21 INFO node.Application: Starting Source r114/08/10 14:32:21 INFO source.AvroSource: Starting Avro source r1: { bindAddress: 0.0.0.0, port: 5555 }...14/08/10 14:32:21 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: SOURCE, name: r1: Successfully registered new MBean.14/08/10 14:32:21 INFO instrumentation.MonitoredCounterGroup: Component type: SOURCE, name: r1 started14/08/10 14:32:21 INFO source.AvroSource: Avro source r1 started.14/08/10 14:32:36 INFO ipc.NettyServer: [id: 0xcf00eea6, /192.168.1.50:35916 => /192.168.1.50:5555] OPEN14/08/10 14:32:36 INFO ipc.NettyServer: [id: 0xcf00eea6, /192.168.1.50:35916 => /192.168.1.50:5555] BOUND: /192.168.1.50:555514/08/10 14:32:36 INFO ipc.NettyServer: [id: 0xcf00eea6, /192.168.1.50:35916 => /192.168.1.50:5555] CONNECTED: /192.168.1.50:3591614/08/10 14:32:44 INFO ipc.NettyServer: [id: 0x432f5468, /192.168.1.51:46945 => /192.168.1.50:5555] OPEN14/08/10 14:32:44 INFO ipc.NettyServer: [id: 0x432f5468, /192.168.1.51:46945 => /192.168.1.50:5555] BOUND: /192.168.1.50:555514/08/10 14:32:44 INFO ipc.NettyServer: [id: 0x432f5468, /192.168.1.51:46945 => /192.168.1.50:5555] CONNECTED: /192.168.1.51:4694514/08/10 14:34:11 INFO sink.LoggerSink: Event: { headers:{type=baidu} body: 69 64 6F 61 6C 6C 5F 54 45 53 54 31 idoall_TEST1 }14/08/10 14:34:57 INFO sink.LoggerSink: Event: { headers:{type=qq} body: 69 64 6F 61 6C 6C 5F 54 45 53 54 33 idoall_TEST3 } |
g)在m2的sink窗口，可以看到以下信息：
|
1
2
3
4
5
6
7
8
9
10
11
12
13
|
14/08/10 14:32:27 INFO node.Application: Starting Sink k114/08/10 14:32:27 INFO node.Application: Starting Source r114/08/10 14:32:27 INFO source.AvroSource: Starting Avro source r1: { bindAddress: 0.0.0.0, port: 5555 }...14/08/10 14:32:27 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: SOURCE, name: r1: Successfully registered new MBean.14/08/10 14:32:27 INFO instrumentation.MonitoredCounterGroup: Component type: SOURCE, name: r1 started14/08/10 14:32:27 INFO source.AvroSource: Avro source r1 started.14/08/10 14:32:36 INFO ipc.NettyServer: [id: 0x7c2f0aec, /192.168.1.50:38104 => /192.168.1.51:5555] OPEN14/08/10 14:32:36 INFO ipc.NettyServer: [id: 0x7c2f0aec, /192.168.1.50:38104 => /192.168.1.51:5555] BOUND: /192.168.1.51:555514/08/10 14:32:36 INFO ipc.NettyServer: [id: 0x7c2f0aec, /192.168.1.50:38104 => /192.168.1.51:5555] CONNECTED: /192.168.1.50:3810414/08/10 14:32:44 INFO ipc.NettyServer: [id: 0x3d36f553, /192.168.1.51:48599 => /192.168.1.51:5555] OPEN14/08/10 14:32:44 INFO ipc.NettyServer: [id: 0x3d36f553, /192.168.1.51:48599 => /192.168.1.51:5555] BOUND: /192.168.1.51:555514/08/10 14:32:44 INFO ipc.NettyServer: [id: 0x3d36f553, /192.168.1.51:48599 => /192.168.1.51:5555] CONNECTED: /192.168.1.51:4859914/08/10 14:34:33 INFO sink.LoggerSink: Event: { headers:{type=ali} body: 69 64 6F 61 6C 6C 5F 54 45 53 54 32 idoall_TEST2 } |
可以看到，根據header中不同的條件分佈到不同的channel上
10)案例10：Flume Sink Processors
failover的機器是一直髮送給其中一個sink，當這個sink不可用的時候，自動發送到下一個sink。
a)在m1創建Flume_Sink_Processors配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors.conf a1.sources = r1
a1.sinks = k1 k2
a1.channels = c1 c2 #這個是配置failover的關鍵，需要有一個sink groupa1.sinkgroups = g1a1.sinkgroups.g1.sinks = k1 k2#處理的類型是failovera1.sinkgroups.g1.processor.type= failover
#優先級，數字越大優先級越高，每個sink的優先級必須不相同a1.sinkgroups.g1.processor.priority.k1 = 5a1.sinkgroups.g1.processor.priority.k2 = 10#設置爲10秒，當然可以根據你的實際狀況更改成更快或者很慢a1.sinkgroups.g1.processor.maxpenalty = 10000 # Describe/configure the sourcea1.sources.r1.type= syslogtcp
a1.sources.r1.port = 5140a1.sources.r1.channels = c1 c2a1.sources.r1.selector.type= replicating
# Describe the sinka1.sinks.k1.type= avro
a1.sinks.k1.channel = c1a1.sinks.k1.hostname= m1
a1.sinks.k1.port = 5555 a1.sinks.k2.type= avro
a1.sinks.k2.channel = c2a1.sinks.k2.hostname= m2
a1.sinks.k2.port = 5555 # Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100 a1.channels.c2.type= memory
a1.channels.c2.capacity = 1000a1.channels.c2.transactionCapacity = 100 |
b)在m1創建Flume_Sink_Processors_avro配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors_avro.conf a1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= avro
a1.sources.r1.channels = c1a1.sources.r1.bind = 0.0.0.0a1.sources.r1.port = 5555 # Describe the sinka1.sinks.k1.type= logger
# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100 # Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
c)將2個配置文件複製到m2上一份
|
1
2
|
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors.conf root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors_avro.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors_avro.conf |
d)打開4個窗口，在m1和m2上同時啓動兩個flume agent
|
1
2
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors_avro.conf
-n a1 -Dflume.root.logger=INFO,console root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors.conf
-n a1 -Dflume.root.logger=INFO,console |
e)然後在m1或m2的任意一臺機器上，測試產生log
|
1
|
root@m1:/home/hadoop# echo "idoall.org test1 failover" | nc localhost 5140 |
f)因爲m2的優先級高，所以在m2的sink窗口，可以看到以下信息，而m1沒有：
|
1
2
3
4
5
|
14/08/10 15:02:46 INFO ipc.NettyServer: Connection to /192.168.1.51:48692 disconnected.14/08/10 15:03:12 INFO ipc.NettyServer: [id: 0x09a14036, /192.168.1.51:48704 => /192.168.1.51:5555] OPEN14/08/10 15:03:12 INFO ipc.NettyServer: [id: 0x09a14036, /192.168.1.51:48704 => /192.168.1.51:5555] BOUND: /192.168.1.51:555514/08/10 15:03:12 INFO ipc.NettyServer: [id: 0x09a14036, /192.168.1.51:48704 => /192.168.1.51:5555] CONNECTED: /192.168.1.51:4870414/08/10 15:03:26 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 31 idoall.org test1 } |
g)這時我們停止掉m2機器上的sink(ctrl+c)，再次輸出測試數據：
|
1
|
root@m1:/home/hadoop# echo "idoall.org test2 failover" | nc localhost 5140 |
h)可以在m1的sink窗口，看到讀取到了剛纔發送的兩條測試數據：
|
1
2
3
4
5
6
|
14/08/10 15:02:46 INFO ipc.NettyServer: Connection to /192.168.1.51:47036 disconnected.14/08/10 15:03:12 INFO ipc.NettyServer: [id: 0xbcf79851, /192.168.1.51:47048 => /192.168.1.50:5555] OPEN14/08/10 15:03:12 INFO ipc.NettyServer: [id: 0xbcf79851, /192.168.1.51:47048 => /192.168.1.50:5555] BOUND: /192.168.1.50:555514/08/10 15:03:12 INFO ipc.NettyServer: [id: 0xbcf79851, /192.168.1.51:47048 => /192.168.1.50:5555] CONNECTED: /192.168.1.51:4704814/08/10 15:07:56 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 31 idoall.org test1 }14/08/10 15:07:56 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 32 idoall.org test2 } |
i)我們再在m2的sink窗口中，啓動sink：
|
1
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Flume_Sink_Processors_avro.conf
-n a1 -Dflume.root.logger=INFO,console |
j)輸入兩批測試數據：
|
1
|
root@m1:/home/hadoop# echo "idoall.org test3 failover" | nc localhost 5140 && echo "idoall.org test4 failover" | nc localhost 5140 |
k)在m2的sink窗口，我們可以看到以下信息，因爲優先級的關係，log消息會再次落到m2上：
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
|
14/08/10 15:09:47 INFO node.Application: Starting Sink k114/08/10 15:09:47 INFO node.Application: Starting Source r114/08/10 15:09:47 INFO source.AvroSource: Starting Avro source r1: { bindAddress: 0.0.0.0, port: 5555 }...14/08/10 15:09:47 INFO instrumentation.MonitoredCounterGroup: Monitored counter group for type: SOURCE, name: r1: Successfully registered new MBean.14/08/10 15:09:47 INFO instrumentation.MonitoredCounterGroup: Component type: SOURCE, name: r1 started14/08/10 15:09:47 INFO source.AvroSource: Avro source r1 started.14/08/10 15:09:54 INFO ipc.NettyServer: [id: 0x96615732, /192.168.1.51:48741 => /192.168.1.51:5555] OPEN14/08/10 15:09:54 INFO ipc.NettyServer: [id: 0x96615732, /192.168.1.51:48741 => /192.168.1.51:5555] BOUND: /192.168.1.51:555514/08/10 15:09:54 INFO ipc.NettyServer: [id: 0x96615732, /192.168.1.51:48741 => /192.168.1.51:5555] CONNECTED: /192.168.1.51:4874114/08/10 15:09:57 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 32 idoall.org test2 }14/08/10 15:10:43 INFO ipc.NettyServer: [id: 0x12621f9a, /192.168.1.50:38166 => /192.168.1.51:5555] OPEN14/08/10 15:10:43 INFO ipc.NettyServer: [id: 0x12621f9a, /192.168.1.50:38166 => /192.168.1.51:5555] BOUND: /192.168.1.51:555514/08/10 15:10:43 INFO ipc.NettyServer: [id: 0x12621f9a, /192.168.1.50:38166 => /192.168.1.51:5555] CONNECTED: /192.168.1.50:3816614/08/10 15:10:43 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 33 idoall.org test3 }14/08/10 15:10:43 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 34 idoall.org test4 } |
11)案例11：Load balancing Sink Processor
load balance type和failover不同的地方是，load balance有兩個配置，一個是輪詢，一個是隨機。兩種情況下如果被選擇的sink不可用，就會自動嘗試發送到下一個可用的sink上面。
a)在m1創建Load_balancing_Sink_Processors配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors.conf a1.sources = r1
a1.sinks = k1 k2
a1.channels = c1
#這個是配置Load balancing的關鍵，需要有一個sink groupa1.sinkgroups = g1a1.sinkgroups.g1.sinks = k1 k2a1.sinkgroups.g1.processor.type= load_balance
a1.sinkgroups.g1.processor.backoff =truea1.sinkgroups.g1.processor.selector = round_robin # Describe/configure the sourcea1.sources.r1.type= syslogtcp
a1.sources.r1.port = 5140a1.sources.r1.channels = c1 # Describe the sinka1.sinks.k1.type= avro
a1.sinks.k1.channel = c1a1.sinks.k1.hostname= m1
a1.sinks.k1.port = 5555 a1.sinks.k2.type= avro
a1.sinks.k2.channel = c1a1.sinks.k2.hostname= m2
a1.sinks.k2.port = 5555 # Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100 |
b)在m1創建Load_balancing_Sink_Processors_avro配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors_avro.conf a1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= avro
a1.sources.r1.channels = c1a1.sources.r1.bind = 0.0.0.0a1.sources.r1.port = 5555 # Describe the sinka1.sinks.k1.type= logger
# Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100 # Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
c)將2個配置文件複製到m2上一份
|
1
2
|
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors.conf
root@m1:/home/hadoop/flume-1.5.0-bin# scp -r /home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors_avro.conf
root@m2:/home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors_avro.conf |
d)打開4個窗口，在m1和m2上同時啓動兩個flume agent
|
1
2
|
root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors_avro.conf
-n a1 -Dflume.root.logger=INFO,console root@m1:/home/hadoop# /home/hadoop/flume-1.5.0-bin/bin/flume-ng agent -c . -f /home/hadoop/flume-1.5.0-bin/conf/Load_balancing_Sink_Processors.conf
-n a1 -Dflume.root.logger=INFO,console |
e)然後在m1或m2的任意一臺機器上，測試產生log，一行一行輸入，輸入太快，容易落到一臺機器上
|
1
2
3
4
|
root@m1:/home/hadoop# echo "idoall.org test1" | nc localhost 5140root@m1:/home/hadoop# echo "idoall.org test2" | nc localhost 5140root@m1:/home/hadoop# echo "idoall.org test3" | nc localhost 5140root@m1:/home/hadoop# echo "idoall.org test4" | nc localhost 5140 |
f)在m1的sink窗口，可以看到以下信息：
|
1
2
|
14/08/10 15:35:29 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 32 idoall.org test2 }14/08/10 15:35:33 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 34 idoall.org test4 } |
g)在m2的sink窗口，可以看到以下信息：
|
1
2
|
14/08/10 15:35:27 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 31 idoall.org test1 }14/08/10 15:35:29 INFO sink.LoggerSink: Event: { headers:{Severity=0, flume.syslog.status=Invalid, Facility=0} body: 69 64 6F 61 6C 6C 2E 6F 72 67 20 74 65 73 74 33 idoall.org test3 } |
說明輪詢模式起到了作用。
12)案例12：Hbase sink
a)在測試之前，請先參考《ubuntu12.04+hadoop2.2.0+zookeeper3.4.5+hbase0.96.2+hive0.13.1分佈式環境部署》將hbase啓動
b)然後將以下文件複製到flume中：
|
1
2
3
4
5
6
7
8
|
cp
/home/hadoop/hbase-0.96.2-hadoop2/lib/protobuf-java-2.5.0.jar/home/hadoop/flume-1.5.0-bin/libcp
/home/hadoop/hbase-0.96.2-hadoop2/lib/hbase-client-0.96.2-hadoop2.jar/home/hadoop/flume-1.5.0-bin/libcp
/home/hadoop/hbase-0.96.2-hadoop2/lib/hbase-common-0.96.2-hadoop2.jar/home/hadoop/flume-1.5.0-bin/libcp
/home/hadoop/hbase-0.96.2-hadoop2/lib/hbase-protocol-0.96.2-hadoop2.jar/home/hadoop/flume-1.5.0-bin/libcp
/home/hadoop/hbase-0.96.2-hadoop2/lib/hbase-server-0.96.2-hadoop2.jar/home/hadoop/flume-1.5.0-bin/libcp
/home/hadoop/hbase-0.96.2-hadoop2/lib/hbase-hadoop2-compat-0.96.2-hadoop2.jar/home/hadoop/flume-1.5.0-bin/libcp
/home/hadoop/hbase-0.96.2-hadoop2/lib/hbase-hadoop-compat-0.96.2-hadoop2.jar/home/hadoop/flume-1.5.0-bin/lib@@@cp
/home/hadoop/hbase-0.96.2-hadoop2/lib/htrace-core-2.04.jar/home/hadoop/flume-1.5.0-bin/lib |
c)確保test_idoall_org表在hbase中已經存在，test_idoall_org表的格式以及字段請參考《ubuntu12.04+hadoop2.2.0+zookeeper3.4.5+hbase0.96.2+hive0.13.1分佈式環境部署》中關於hbase部分的建表代碼。
d)在m1創建hbase_simple配置文件
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
|
root@m1:/home/hadoop# vi /home/hadoop/flume-1.5.0-bin/conf/hbase_simple.conf a1.sources = r1
a1.sinks = k1
a1.channels = c1
# Describe/configure the sourcea1.sources.r1.type= syslogtcp
a1.sources.r1.port = 5140a1.sources.r1.host = localhosta1.sources.r1.channels = c1 # Describe the sinka1.sinks.k1.type= logger
a1.sinks.k1.type= hbase
a1.sinks.k1.table = test_idoall_orga1.sinks.k1.columnFamily = namea1.sinks.k1.column = idoalla1.sinks.k1.serializer = org.apache.flume.sink.hbase.RegexHbaseEventSerializera1.sinks.k1.channel = memoryChannel # Use a channel which buffers events in memorya1.channels.c1.type= memory
a1.channels.c1.capacity = 1000a1.channels.c1.transactionCapacity = 100 # Bind the source and sink to the channela1.sources.r1.channels = c1a1.sinks.k1.channel = c1 |
e)啓動flume agent
|
1
|
/home/hadoop/flume-1.5.0-bin/bin/flume-ngagent -c . -f
/home/hadoop/flume-1.5.0-bin/conf/hbase_simple.conf -n a1 -Dflume.root.logger=INFO,console |
f)測試產生syslog
|
1
|
root@m1:/home/hadoop# echo "hello idoall.org from flume" | nc localhost 5140 |
g)這時登錄到hbase中，可以發現新數據已經插入
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
|
root@m1:/home/hadoop# /home/hadoop/hbase-0.96.2-hadoop2/bin/hbase shell2014-08-10 16:09:48,984 INFO [main] Configuration.deprecation: hadoop.native.lib is deprecated. Instead, use io.native.lib.availableHBase Shell; enter'help<RETURN>'
for list of supported commands.
Type "exit<RETURN>"to leave the HBase Shell
Version 0.96.2-hadoop2, r1581096, Mon Mar 24 16:03:18 PDT 2014 hbase(main):001:0> listTABLE SLF4J: Class path contains multiple SLF4J bindings.SLF4J: Found bindingin
[jar:file:/home/hadoop/hbase-0.96.2-hadoop2/lib/slf4j-log4j12-1.6.4.jar!/org/slf4j/impl/StaticLoggerBinder.class]SLF4J: Found bindingin
[jar:file:/home/hadoop/hadoop-2.2.0/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings
for an explanation. hbase2hive_idoall hive2hbase_idoall test_idoall_org 3 row(s)
in 2.6880 seconds
=> ["hbase2hive_idoall","hive2hbase_idoall","test_idoall_org"]hbase(main):002:0> scan"test_idoall_org"ROW COLUMN+CELL 10086 column=name:idoall, timestamp=1406424831473, value=idoallvalue 1 row(s)
in 0.0550 seconds
hbase(main):003:0> scan"test_idoall_org"ROW COLUMN+CELL 10086 column=name:idoall, timestamp=1406424831473, value=idoallvalue 1407658495588-XbQCOZrKK8-0 column=name:payload, timestamp=1407658498203, value=hello idoall.org from flume 2 row(s)
in 0.0200 seconds
hbase(main):004:0> quit |
經過這麼多flume的例子測試，如果你全部做完後，會發現flume的功能真的很強大，可以進行各種搭配來完成你想要的工作，俗話說師傅領進門，修行在個人，如何能夠結合你的產品業務，將flume更好的應用起來，快去動手實踐吧。
