This post says that multiline got added to the tags field. Sadly, it doesn't work for me. When I view the processed multiline message in kibana, the tags field says "beats_input_codec_plain_applied".
I need to do some special filtering when a multiline message is processed. I technically have a workaround by adding a new field that triggers with the same condition as the multiline event. (as it is explained here)
I also found someone else having troubles with this ( here ).
Does anyone know if we are just missing out on a parameter to add the multiline to tags, or is there something else wrong?
Looking at the relevant PR, it doesn't look like a tag is supposed to be set, but rather the field multiline.lines. Have you checked to see if that was set?
I checked that now, but when I view the message in kibana, it doesn't show the field multiline.lines. In addition, I checked the fields in the index pattern, and the mentioned field isn't listed there either.
The multiline part works, but it basically ignores the whole processors part. Am I doing something wrong? Is there any other way to to tell my filter that there is a multiline event happening?
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.