flume hdfs文件输出普通文本正确行数

最新推荐文章于 2023-01-03 15:36:18 发布

sunrising_hill

最新推荐文章于 2023-01-03 15:36:18 发布

阅读量834

点赞数

CC 4.0 BY-SA版权

分类专栏：消息系统

本文链接：https://blog.youkuaiyun.com/sunrising_hill/article/details/82800229

消息系统专栏收录该内容

3 篇文章

订阅专栏

本文详细介绍了如何正确配置Flume从Kafka读取数据并输出到HDFS的过程，解决了数据行数不匹配的问题。通过调整源、通道和接收器设置，确保了数据的准确传输。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

本次调试flume读取kafka数据输出至hdfs总是内容不对，使用fs -cat查看行数不对，经过测试后发现需要如下配置：

agent.sources = midlog
agent.channels = hdfsChannel
agent.sinks = hdfsSink

# the source of hdfs
agent.sources.midlog.type = org.apache.flume.source.kafka.KafkaSource
agent.sources.midlog.channels = hdfsChannel
agent.sources.midlog.batchSize = 5000
agent.sources.midlog.batchDurationMillis = 2000
agent.sources.midlog.kafka.bootstrap.servers = master.server.com:9092
agent.sources.midlog.kafka.topics = mid_analog
agent.sources.midlog.kafka.consumer.group.id = custom.g.id

# the channel of hdfs
agent.channels.hdfsChannel.type = file
agent.channels.hdfsChannel.checkpointDir = /data/flume/checkpointDir
agent.channels.hdfsChannel.dataDirs = /data/flume/dataDirs

# the sink of hdfs
agent.sinks.hdfsSink.type = hdfs
agent.sinks.hdfsSink.channel = hdfsChannel

agent.sinks.hdfsSink.hdfs.path = /flumelogs/midlog
agent.sinks.hdfsSink.hdfs.fileType = DataStream
agent.sinks.hdfsSink.hdfs.filePrefix = src
agent.sinks.hdfsSink.hdfs.round = true
agent.sinks.hdfsSink.hdfs.roundValue = 10
agent.sinks.hdfsSink.hdfs.roundUnit = minute

agent.sinks.hdfsSink.hdfs.rollInterval = 0
agent.sinks.hdfsSink.hdfs.rollSize = 0
agent.sinks.hdfsSink.hdfs.rollCount = 1000