flume-ng负载均衡load-balance、failover集群搭建
转自：http://blog.csdn.net/morning_pig/article/details/9093149集群采用3台机器：
viewplaincopy
[*]host1 load-balance设置
[*]host2 机器1
[*]host3 机器2
其中，host1 机器配置：
viewplaincopy
[*]#Define a memory channel called c1 on a1
[*]a1.channels = c1
[*]a1.sources = r1
[*]a1.sinks = k1 k2
[*]a1.sinkgroups = g1
[*]a1.sinkgroups.g1.sinks = k1 k2
[*]a1.sinkgroups.g1.processor.type = load_balance
[*]a1.sinkgroups.g1.processor.selector = round_robin
[*]a1.sinkgroups.g1.processor.backoff = true
[*]
[*]a1.channels.c1.type = file
[*]a1.channels.c1.checkpointDir = /tmp/flume/loadcheckpoint
[*]a1.channels.c1.dataDirs = /tmp/flume/loaddata
[*]
[*]a1.sources.r1.channels = c1
[*]a1.sources.r1.type = avro
[*]a1.sources.r1.bind = 0.0.0.0
[*]a1.sources.r1.port = 41415
[*]
[*]a1.sinks.k1.channel = c1
[*]a1.sinks.k1.type = avro
[*]a1.sinks.k1.hostname = host2
[*]a1.sinks.k1.port = 41414
[*]a1.sinks.k2.channel = c1
[*]a1.sinks.k2.type = AVRO
[*]a1.sinks.k2.hostname = host3
[*]a1.sinks.k2.port = 41414
host2 机器配置：
viewplaincopy
[*]a2.channels = c1
[*]a2.sources = r1
[*]a2.sinks = k1
[*]
[*]a2.channels.c1.type = FILE
[*]a1.channels.c1.checkpointDir = /tmp/flume/checkpoint
[*]a1.channels.c1.dataDirs = /tmp/flume/data
[*]
[*]a2.sources.r1.channels = c1
[*]a2.sources.r1.type = AVRO
[*]a2.sources.r1.bind = 0.0.0.0
[*]a2.sources.r1.port = 41414
[*]
[*]a2.sinks.k1.channel = c1
[*]a2.sinks.k1.type = file_roll
[*]a2.sinks.k1.sink.directory = /tmp/load/
[*]a2.sinks.k1.sink.rollInterval = 0
host3 机器配置：
viewplaincopy
[*]a2.channels = c1
[*]a2.sources = r1
[*]a2.sinks = k1
[*]
[*]a2.channels.c1.type = FILE
[*]a1.channels.c1.checkpointDir = /tmp/flume/checkpoint
[*]a1.channels.c1.dataDirs = /tmp/flume/data
[*]
[*]a2.sources.r1.channels = c1
[*]a2.sources.r1.type = AVRO
[*]a2.sources.r1.bind = 0.0.0.0
[*]a2.sources.r1.port = 41414
[*]
[*]a2.sinks.k1.channel = c1
[*]a2.sinks.k1.type = file_roll
[*]a2.sinks.k1.sink.directory = /tmp/load/
[*]a2.sinks.k1.sink.rollInterval = 0
客户端机器暂时使用flume-ng的agent发送，配置如下：
viewplaincopy
[*]# Define a memory channel called c1 on a1
[*]a1.channels.c1.type = file
[*]a1.channels.c1.checkpointDir = /tmp/flume/checkpoint
[*]a1.channels.c1.dataDirs = /tmp/flume/data
[*]
[*]
[*]a1.sources.r1.channels = c1
[*]a1.sources.r1.type = exec
[*]a1.sources.r1.command = cat /tmp/linux.log
[*]
[*]a1.sinks.k1.type = avro
[*]a1.sinks.k1.channel = c1
[*]a1.sinks.k1.hostname = host1
[*]a1.sinks.k1.port = 41415
[*]
[*]a1.channels = c1
[*]a1.sources = r1
[*]a1.sinks = k1
客户端/tmp/linux.log文件3G左右，发送给host1。
启动：
host2：
viewplaincopy
[*]bin/flume-ng agent -c conf -f conf/load-sink1.conf -n a2
host3：
viewplaincopy
[*]bin/flume-ng agent -c conf -f conf/load-sink2.conf -n a2
host1：
viewplaincopy
[*]bin/flume-ng agent -c conf -f conf/load-balance.conf -n a1
客户端：
viewplaincopy
[*]bin/flume-ng agent -c conf -f conf/client.conf -n a1
注意：启动方式最好是从下往上启动。即：先启动host2和host3，然后启动host1，最后启动client。
测试过程中，可以随时将host2或host3停止，过一段时间再启动。
这样，就测试了flume-ng的load-balance和failover功能。
页:
[1]