设为首页 加入收藏

TOP

Flume使用大全之kafka source-kafka channel-hdfs(kerberos认证)
2018-11-28 17:43:55 】 浏览:24
Tags:Flume 使用 大全 kafka source-kafka channel-hdfs kerberos 认证


agent.sources = kafkaSource1
agent.channels = kafkaChannel
agent.sinks = hdfsSink
agent.sources.kafkaSource1.channels = kafkaChannel
agent.sinks.hdfsSink.channel = kafkaChannel




agent.sources.kafkaSource1.type = org.apache.flume.source.kafka.KafkaSource
agent.sources.kafkaSource1.zookeeperConnect = node1:2181
agent.sources.kafkaSource1.topic = bpu_sensor_router,bpu_sensor_record_present,bpu_group_status_present,bpu_gateway_heartbeat,bpu_gateway_router,bpu_sensor_heartbeat
agent.sources.kafkaSource1.consumer.group.id = flume
agent.sources.kafkaSource1.kafka.consumer.timeout.ms = 100
agent.sources.kafkaSource1.kafka.bootstrap.servers = node7:9092
agent.sources.kafkaSource1.batchSize = 500000
agent.sources.kafkaSource1.batchDurationMillis = 2000
agent.sources.kafkaSource1.kafka.consumer.security.protocol = SASL_PLAINTEXT
agent.sources.kafkaSource1.kafka.consumer.sasl.mechanism = GSSAPI
agent.sources.kafkaSource1.kafka.consumer.sasl.kerberos.service.name = kafka






agent.channels.kafkaChannel.type = org.apache.flume.channel.kafka.KafkaChannel
agent.channels.kafkaChannel.kafka.bootstrap.servers = node7:9092
agent.channels.kafkaChannel.kafka.topic = flume-kafkaChannel
agent.channels.kafkaChannel.consumer.group.id = flume-consumer
agent.channels.kafkaChannel.kafka.producer.timeout.ms = 2000
agent.channels.kafkaChannel.kafka.producer.security.protocol = SASL_PLAINTEXT
agent.channels.kafkaChannel.kafka.producer.sasl.mechanism = GSSAPI
agent.channels.kafkaChannel.kafka.producer.sasl.kerberos.service.name = kafka


agent.channels.kafkaChannel.kafka.consumer.timeout.ms = 2000
agent.channels.kafkaChannel.kafka.consumer.security.protocol = SASL_PLAINTEXT
agent.channels.kafkaChannel.kafka.consumer.sasl.mechanism = GSSAPI
agent.channels.kafkaChannel.kafka.consumer.sasl.kerberos.service.name = kafka








#---------hdfsSink 相关配置------------------
agent.sinks.hdfsSink.type = hdfs
# 注意, 我们输出到下面一个子文件夹data中
agent.sinks.hdfsSink.hdfs.path = hdfs://nameservice1/user/hive/warehouse/%{topic}/%Y/%m/%d
agent.sinks.hdfsSink.hdfs.writeFormat = TEXT
agent.sinks.hdfsSink.hdfs.fileType = DataStream
agent.sinks.hdfsSink.hdfs.rollSize = 120000000
agent.sinks.hdfsSink.hdfs.rollInterval= 60
agent.sinks.hdfsSink.hdfs.rollCount = 0
agent.sinks.hdfsSink.hdfs.batchSize = 15000
agent.sinks.hdfsSink.hdfs.round = true
agent.sinks.hdfsSink.hdfs.roundUnit = day
agent.sinks.hdfsSink1.hdfs.roundValue = 1
agent.sinks.hdfsSink.hdfs.threadsPoolSize = 25
agent.sinks.hdfsSink.hdfs.useLocalTimeStamp = true
agent.sinks.hdfsSink.hdfs.minBlockReplicas = 1
agent.sinks.hdfsSink.hdfs.idleTimeout = 30
agent.sinks.hdfsSink.hdfs.filePrefix= %{topic}

】【打印繁体】【投稿】【收藏】 【推荐】【举报】【评论】 【关闭】 【返回顶部
上一篇flume通过thrift协议收集日志-Pyt.. 下一篇SODBASE CEP学习进阶篇(二)续:..

最新文章

热门文章

Hot 文章

Python

C 语言

C++基础

大数据基础

linux编程基础

C/C++面试题目