Hello,
I'm trying to write a pipeline to parse following part of data
ua="192.168.211.109:443"
I implement it as
ua=\"%{IP:wkaddress}:%{NUMBER:port:integer}\"
but always I have wkaddress and port as Text field.
When I try its using simulate in develop tools port appears as integer but into the log it always is set up as text.
Someone have the same problem?
Have you checked your mappings? This looks more a Elasticsearch question but anyhow, I've done the following test and it worked as expected (7.9.1). From the DevTools console:
Create an index with a strict mapping with IP and port fields
Create a pipeline that processes a message like yours and extract the fields
Simulate the pipeline to be sure it works
Test the pipeline with a new document
Perform a IP range search to ensure the document is returned
Moving this to the Elasticsearch forum since this is not Kibana related.
My suggestion would be to try to reproduce your parsing error in the simplest and reproducible environment, try to use the /_ingest/pipeline/{your_pipeline}/_simulate to get your pipeline working as you expect, then maybe create a new index with the mapping with dynamic: strict to avoid types that you don't want and test your documents on that index and see if that works.
Sorry but running out of ideas with the information available.
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.