I have a Magento2 instance for which I collect the logs using prospectors in filebeat-5.2.1. This is happening on a docker container based on ubuntu xenial (16.04).
Problem is, lines are split on newline. The json parsing and multiline filters are applied thereafter.
Given JSON is structured, I would like to beats to not requiring the newline, but just stream through the JSON document, parsing object after object after object. Without requiring newlines, but also supporting pretty printed JSON. This is unfortunately not possible as of now. But we're well aware of this use-case. For this and other common use-cases I would like the filebeat readers to be somewhat chainable, so we can plugin decompression, json, xml, multiline, newline splitting, content splitting on custom pattern in any order required.
Unfortunately I didn't find a ticket, but feel free to open an enhancement request.
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.