tree: ee33c7ede892ec372e4fb1098dbea19f0a69bb9c [path history] [tgz]
  1. src/
  2. pom.xml
  3. README.md
htrace-flume/README.md

htrace-flume

htrace-flume provides the span receiver which sends tracing spans to Flume collector.

Tutorial

  1. build and deploy

    $ cd htrace/htrace-flume $ mvn compile assembly:single $ cp target/htrace-flume-*-jar-with-dependencies.jar $HADOOP_HOME/share/hadoop/hdfs/lib/

  2. Edit hdfs-site.xml to include the following:

  3. Setup flume collector

Create flume-conf.properties file. Below is a sample that sets up an hdfs sink.

agent.sources = avro-collection-source
agent.channels = memoryChannel
agent.sinks = loggerSink hdfs-sink

# avro source - should match the configurations in hdfs-site.xml
agent.sources.avro-collection-source.type = avro
agent.sources.avro-collection-source.bind = 127.0.0.1
agent.sources.avro-collection-source.port = 60000
agent.sources.avro-collection-source.channels = memoryChannel

#sample hdfs-sink, change to any sink that flume supports
agent.sinks.hdfs-sink.type = hdfs
agent.sinks.hdfs-sink.hdfs.path = hdfs://127.0.0.1:9000/flume
agent.sinks.hdfs-sink.channel = memoryChannel
agent.sinks.hdfs-sink.hdfs.fileType = DataStream
agent.sinks.hdfs-sink.hdfs.writeFormat = Text
agent.sinks.hdfs-sink.hdfs.rollSize = 0
agent.sinks.hdfs-sink.hdfs.rollCount = 10000
agent.sinks.hdfs-sink.hdfs.batchSize = 100

# memory channel
agent.channels.memoryChannel.capacity = 10000
agent.channels.memoryChannel.transactionCapacity = 1000

Run flume agent using command “flume-ng agent -c ./conf/ -f ./conf/flume-conf.properties -n agent”