08-16-2018 04:49 AM - edited 08-16-2018 05:25 AM
I am trying to work with kafka for data ingestion but being new to this, i kind of pretty much confused.
I have multiple crawlers, who extract data for me from web platforms, now the issue is i want to ingest that extracted data to hadoop using kafka without any middle scripts/service file . Main commplication is that, platforms are disparate in nature and one web platform is providing real-time data other batch based. Can integrate my crawlers some how with kafka producers ? and they keep running all by themselves. is it possible ? I think it is but i am not getting in right direction. any help would be appreciated.
02-03-2019 05:09 AM
Loading data directly to Kafka without any Service seems unlikely.
However, you can use execute a simple kafka console producer to send all your data to the kafka service. But if your requirement is to save data to HDFS you need to include a few more services along with Kafka.
For example, Crawlers >> kafka console producer (or) Spark Streaming >> Flume >> HDFS
As your requirement is to store the data in HDFS and not stream the data. I suggest you execute a Spark job, it will store your data to HDFS. Refer mentioned commands to execute a spark job to move data to HDFS.
Initiate a spark-shell
Execute the mentioned command in the Spark shell in the same order.
val moveFile = sc.textFile("file:///path/to/Sample.log")