We are needing to develop a custom interceptor to count the number of lines on the text file we are ingesting using Flume Spool Dir source.
Apache document discusses the ability to develop custom interceptors using its Interceptor interface.
After developing and compiling the custom interceptor code, documentation talkd about needing to copy the compiled jar to /usr/lib/flume-ng/lib/ so that it can be then used in the flume agent's configuration.
Although, i am not sure how it needs to work for the cloudera CDH. Where do we need to copy the jar file to. Copying to - /opt/cloudera/parcels/CDH-5.11.xxxx/lib/flume-ng/lib does not seem to work.