I'm using Elasticsearch and Kibana for the first time. I have an index "q4", with type "tweet" which contains Twitter data. I want to map these tweets onto a Kibana Tile Map visualization, so I have added a geo_point property "location" to the mapping of the tweets. However, when I attempt to read this data in Kibana, I get the following message:
No Compatible Fields: The "q4" index pattern does not contain any of the
following field types: geo_point
I'm not sure why I'm getting this error, as the mapping of index data seems to match the documentation.
Below is the mapping of my index:
{u'q4': {
u'mappings': {
u'tweet': {
u'properties': {
...(a bunch of irrelevant properties here)...
u'location': {
u'type': u'geo_point'},
}
}
}
}
}
My geo_points are currently represented as a comma-separated string - some example values are 41.1751247162,73.965773519
and 33.0732112811,111.596972035
. Kibana recognizes these points as strings, but not as geo_points.
I am using the below links as reference for ES geo_points and Kibana visualizations.
- https://www.elastic.co/guide/en/elasticsearch/reference/current/geo-point.html
- https://www.digitalocean.com/community/tutorials/how-to-map-user-location-with-geoip-and-elk-elasticsearch-logstash-and-kibana
Another SO question seems to cover a similar issue, but that situation involved the use of Logstash. I am not using Logstash in my application - rather, I am sending the tweets directly to ElasticSearch using Python's Elasticsearch API wrapper library.