Apache Flume without hadoop

Question

I am new to flume and hadoop. I have one doubt: Whether we can use flume without configuring hadoop?

Can any one share their knowledge?

Answer 1

Yes Flume can work totally independent of hadoop. However it does have some package dependencies on HDFS. If you are not planning to use HDFS sink, you can use it without Hadoop

Answer 2

Yes! As for my recent flume configuration[source=Twitter, channel=memory], I used a loggerSink to put the streamed 'data' into a log file in the file system(unix file dir), you just need to do something for the data to make sense. There are other sinks available though as part of the flume install(HDFSSink, HBaseSink, IRC, File Roll, check https://flume.apache.org/FlumeUserGuide.html for more details and usage), or you can create a new one.