I'm creating a new beat application, primarily to remove the logstash requirements for getting this data into Elastic Search. The file being read is new line delimited JSON, and gets rotated by logrotate.
Anyone reusing filebeat code for the "tailing" and "registry" tracking in their custom beat applications? Or should I just roll my own.
Filebeat 5.0 alpha1 adds json per line support. The JSON is parsed and send as event to elasticsearch. Try console output with pretty: true to check out your events. Even better, Elasticsearch 5.0 will have Ingest-Node adding some filtering/parsing capabilities to Elasticsearch input.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.