[TOC] # failover故障转移 在完成单点的Flume NG搭建后，下面我们搭建一个高可用的Flume NG集群，架构图如下所示：  ## 节点分配 Flume的Agent和Collector分布如下表所示 | 名称 | ip地址 | Host | 角色 | | --- | --- | --- | --- | | Agent1 | 1992.168.200.101 | it1 | webserver | | collector1 | 192.168.200.102 | it2 | AgentMstr1 | | collector2 | 192.168.200.103 | it3 | AgentMstr2 | Agent1数据分别流入到Collector1和Collector2，Flume NG本身提供了Failover机制，可以自动切换和恢复。下面我们开发配置Flume NG集群 ## 配置 在下面单点Flume中，基本配置都完成了，我们只需要新添加两个配置文件，它们是flume-client.conf和flume-server.conf，其配置内容如下所示： 这边it1,it2,it3是不同机器的hostname 1. it1上的flume-client.conf配置 ~~~ #agent1 name agent1.channels = c1 agent1.sources = r1 # sinks可以配置多个 agent1.sinks = k1 k2 #设置组别为g1 agent1.sinkgroups = g1 #把g1和k1,k2这2个组装起来 agent1.sinkgroups.g1.sinks = k1 k2 #设置channel agent1.channels.c1.type = memory agent1.channels.c1.capacity = 1000 agent1.channels.c1.transactionCapacity = 100 agent1.sources.r1.channels = c1 agent1.sources.r1.type = exec agent1.sources.r1.command = tail -F /root/log/test.log agent1.sources.r1.interceptors = i1 i2 agent1.sources.r1.interceptors.i1.type = static agent1.sources.r1.interceptors.i1.key = Type agent1.sources.r1.interceptors.i1.value = LOGIN agent1.sources.r1.interceptors.i2.type = timestamp # set sink1 agent1.sinks.k1.channel = c1 agent1.sinks.k1.type = avro agent1.sinks.k1.hostname = it2 agent1.sinks.k1.port = 52020 # set sink2 agent1.sinks.k2.channel = c1 agent1.sinks.k2.type = avro agent1.sinks.k2.hostname = it3 agent1.sinks.k2.port = 52020 #设置failover故障转移 agent1.sinkgroups.g1.processor.type = failover # 配置优先级,优先传输到k1 agent1.sinkgroups.g1.processor.priority.k1 = 10 agent1.sinkgroups.g1.processor.priority.k2 = 5 # 将故障的sink剔除的时间,出了故障不会立刻剔除 agent1.sinkgroups.g1.processor.maxpenalty = 10000 #这里首先要申明一个sinkgroups,然后再设置2个sink ,k1与k2,其中2个优先级是10和5， #而processor的maxpenalty被设置为10秒，默认是30秒 ~~~ 启动命令： ~~~ flume-ng agent -n agent1 -c conf -f /root/flume/conf/flume-client.conf -Dflume.root.logger=DEBUG,console ~~~ 2. it2和it3上的flume-server.conf配置 ~~~ #set Agent name a1.sources = r1 a1.channels = c1 a1.sinks = k1 #set channel a1.channels.c1.type = memory a1.channels.c1.capacity = 1000 a1.channels.c1.transactionCapacity = 100 # other node,nna to nns a1.sources.r1.type = avro # 任意一台服务器都可以连接 a1.sources.r1.bind = 0.0.0.0 a1.sources.r1.port = 52020 a1.sources.r1.channels = c1 # 配置拦截器 a1.sources.r1.interceptors = i1 i2 a1.sources.r1.interceptors.i1.type = timestamp a1.sources.r1.interceptors.i2.type = host a1.sources.r1.interceptors.i2.hostHeader=hostname #set sink to hdfs a1.sinks.k1.type=hdfs a1.sinks.k1.hdfs.path=/data/flume/logs/%{hostname} a1.sinks.k1.hdfs.filePrefix=%Y-%m-%d a1.sinks.k1.hdfs.fileType=DataStream a1.sinks.k1.hdfs.writeFormat=TEXT a1.sinks.k1.hdfs.rollInterval=10 a1.sinks.k1.channel=c1 ~~~ 启动命令： ~~~ flume-ng agent -n a1 -c conf -f /root/flume/conf/flume-server.conf -Dflume.root.logger=DEBUG,console ~~~ ## 测试failover 1. 先在it2和it3上启动脚本 ~~~ flume-ng agent -n a1 -c conf -f conf/flume-server.conf -Dflume.root.logger=DEBUG,console ~~~ 2. 然后启动it1上的脚本 ~~~ flume-ng agent -n agent1 -c conf -f conf/flume-client.conf -Dflume.root.logger=DEBUG,console ~~~ 3. Shell脚本生成数据 ~~~ while true;do date >> test.log; sleep 1s ;done ~~~ # load balance负载均衡 1. 节点分配 如failover故障转移的节点分配 2. 配置 在failover故障转移的配置上稍作修改 it1上的flume-client-loadbalance.conf配置 ~~~ #agent1 name agent1.channels = c1 agent1.sources = r1 agent1.sinks = k1 k2 #set gruop agent1.sinkgroups = g1 #set channel agent1.channels.c1.type = memory agent1.channels.c1.capacity = 1000 agent1.channels.c1.transactionCapacity = 100 agent1.sources.r1.channels = c1 agent1.sources.r1.type = exec agent1.sources.r1.command = tail -F /root/log/test.log # set sink1 agent1.sinks.k1.channel = c1 agent1.sinks.k1.type = avro agent1.sinks.k1.hostname = it2 agent1.sinks.k1.port = 52020 # set sink2 agent1.sinks.k2.channel = c1 agent1.sinks.k2.type = avro agent1.sinks.k2.hostname = it3 agent1.sinks.k2.port = 52020 #set sink group agent1.sinkgroups.g1.sinks = k1 k2 #设置负载均衡 agent1.sinkgroups.g1.processor.type = load_balance # 默认是round_robin轮询，还可以选择random随机 agent1.sinkgroups.g1.processor.selector = round_robin #如果backoff被开启，则 sink processor会屏蔽故障的sink # 开启,表示如果有某个slink发生故障,他就不会再继续发送,会把他剔除 agent1.sinkgroups.g1.processor.backoff = true ~~~ it2和it3上的flume-server-loadbalance.conf配置 ~~~ #set Agent name a1.sources = r1 a1.channels = c1 a1.sinks = k1 #set channel a1.channels.c1.type = memory a1.channels.c1.capacity = 1000 a1.channels.c1.transactionCapacity = 100 # other node,nna to nns a1.sources.r1.type = avro a1.sources.r1.bind = 0.0.0.0 a1.sources.r1.port = 52020 a1.sources.r1.channels = c1 a1.sources.r1.interceptors = i1 i2 a1.sources.r1.interceptors.i1.type = timestamp a1.sources.r1.interceptors.i2.type = host a1.sources.r1.interceptors.i2.hostHeader=hostname a1.sources.r1.interceptors.i2.useIP=false #set sink to hdfs a1.sinks.k1.type=hdfs a1.sinks.k1.hdfs.path=/data/flume/loadbalance/%{hostname} a1.sinks.k1.hdfs.fileType=DataStream a1.sinks.k1.hdfs.writeFormat=TEXT a1.sinks.k1.hdfs.rollInterval=10 a1.sinks.k1.channel=c1 a1.sinks.k1.hdfs.filePrefix=%Y-%m-%d ~~~ ## 测试load balance 1. 先在it2和it3上启动脚本 ~~~ bin/flume-ng agent -n a1 -c conf -f conf/flume-server-loadbalance.conf -Dflume.root.logger=DEBUG,console ~~~ 2. 然后启动itcast01上的脚本 ~~~ bin/flume-ng agent -n agent1 -c conf -f conf/flume-client-loadbalance.conf -Dflume.root.logger=DEBUG,console ~~~ 3. Shell脚本生成数据 ~~~ while true;do date >> test.log; sleep 1s ;done ~~~ 4. 观察HDFS上生成的数据目录，由于轮训机制都会收集到数据 5. it2上的agent被干掉之后，it2上不在产生数据 6. it2上的agent重新启动后，两者都可以接受到数据