flume-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Arvind Prabhakar <arv...@apache.org>
Subject Re: Understanding the data flow between Source and Sink in Flume
Date Thu, 29 Mar 2012 16:34:19 GMT
Hi Som,

Look at the implementation of NullSink to get an idea of how to code
up the sink. Once it is ready, make sure that you have it packaged in
the lib directory of the distribution and reference it by using the
fully qualified class name of the implementation in  the type
attribute.

Thanks,
Arvind

On Thu, Mar 29, 2012 at 2:16 AM, shekhar sharma <shekhar2581@gmail.com> wrote:
> Hi,
> I am using Flume-728 (Flume NG) and i am trying to build a custom sink for
> Esper to apply EPL queries on the events generated. I have already custom
> sink, but it is not at all instantiating my sink class...I would like to
> understand how the flow actually happens from source to sink.
>
> Lets say my source is avro who is collecting events from an avro client
> (AvroCLI) and i am sending the events to custom sink, then wat all things i
> need to do so that source is able to send the events to custom sink...
> I have done some study on the code, and following things are mandatory
>
> custom sink should be derived from Abstract Sink and optionally implements
> Configurable. The class should override process() method which does the
> actual processing of events..
> Can you please let me know what all things are required inorder to recognize
> the custom sink by flume...
>
> Regards,
> Som

Mime
View raw message