I am trying to test out parsing some new logs via Custom File stream Integration. I did an initial upload and then worked backwards by creating the ingest pipeline using the sample documents from the initial upload.
I had made a mistake with ecs fields. I had parsed out the fields for
I made the mistake of not converting them. So, I added a convert processor to change them into numeric fields.
The issue is my data view doesn't have a mapping conflict error. I assumed there would be one since it was first a keyword and now a numeric field. It appears as if the convert processor is not working or perhaps, I am not understanding how this works. The goal is to make the ecs fields as expected numeric fields.
Cool! Yeah keep _rollover in the back of your mind, tricky part with metrics because of TSDS etc... the rollover takes about 30 mins to take affect (very long discussion, because how the documents are routed by time etc. )
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.