I am still learning grok and have certain basic query which I tried searching but could not find the perfect answer hence asking to this community. My apologies if that was already asked.
What is [type] in filter? is this the extracted field? Or what field the parser to look for - to match the message?
I assume you are asking about the type in the this construct: %{NUMBER:bytes:int}. The fisrt component NUMBER is the pattern to use and bytes is the name of the field the matched data will be stored in. The int portion is the type and specifies how this should be formatted in the JSON document. This allows you to automatically convert it to a int or float instead of the default string format but should not be confused with the types specified in the mappings used when Elasticsearch is parsing the data.
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.