It is an empty index (all documents deleted by curl request) in order to avoid reindexing. According https://www.elastic.co/blog/changing-mapping-with-zero-downtime, I did following:
mpx@mqzhlmpx07:/var/opt/six/mpx/Kibana> curl -XPUT 'localhost:9200/filebeat-2017.06.13/_mapping/log?pretty' -H 'Content-Type: application/json' -d'
{
"log": {
"properties": {
"message": {
"type": "multi_field",
"fields": {
"message": { "type": "text" },
"message_k": { "type": "keyword" }
}
}
}
}
}
'
{
"error" : {
"root_cause" : [
{
"type" : "mapper_parsing_exception",
"reason" : "No handler for type [multi_field] declared on field [message]"
}
],
"type" : "mapper_parsing_exception",
"reason" : "No handler for type [multi_field] declared on field [message]"
},
"status" : 400
}
Please tell me what is now wrong? (something to do with norms property?)
Thanks to my valuable team members, I could elegantly solve the issue: Quoting the expression is solving the issue.
Example:
GET _search
{
"query": {
"query_string": {"query": "message: "215.SODHK""}
}
}
Thus there is no need for an update on generated mapping. For me is this case closed
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.