I'm currently trying to solve a problem regarding log display in Kibana.
I'm using elasticsearch 6.6.1 and Kibana 6.6.1.
When creating vizualization for my log, the data set is missing every long log message (>~300 char) despite the messages are present in elasticsearch.
I've tried to use the following configuration change in elasticsearch.yml
indices.query.bool.max_clause_count: 4096
But this doesn't solve my problem. Is there a issue with long message with 6.6.1 ?
Or do I miss a configuration ?
Thanks !
That setting is about the maximum number of terms that can be used in a query so is not appropriate.
I think the issue is that long messages are actually missing from the index.
Lucene has a hard limit on the length of indexed values so index mappings tend to set limits for
keyword fields (see ignore_above setting)
You end up with too many large unique values in these situations so it would make more sense to do one of the following:
Truncate the values (this can be done in the mapping)
Pre-process them to recognise certain long messages and swap for shorter reason-codes
Hash the values of the text messages (this can be done in the mapping or an ingest pipeline)
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.