Hello Elastic Community,
I'm exploring the use of JDBC to connect Kibana(bypassing Elastic) directly to MySQL. This would help us overcome the challenge of joining data from different indices in Elasticsearch. I'd appreciate any guidance, tips, or insights from the community.
I would greatly appreciate any guidance, insights, or tips from the community regarding this endeavor. If you have experience with setting up a JDBC connection in Kibana, or if you have suggestions on best practices, potential challenges, or alternative approaches, please do share your knowledge.
@Kumar_Abhinav I'm unsure about a JDBC driver going directly to Kibana, but here's our docs on the use of the JDBC connector in Elasticsearch -> SQL
Can you provide some context of what you mean with bypassing Elastic?
Kibana is the visualization and management tool of the Elastic Stack, it does not store any data, all data is stored in Elasticsearch and Kibana reads from it.
I mean to say that instead of getting data from Elastic, Kibana gets the data directly from MYSQL.
That is simply not how Kibana works ... Kibana reads data only from elasticsearch.
You can go the other way use a SQL tool to connect to elasticsearch via the JDBC / ODBC drivers.
Thanks for the KT. But Elastic search does not allow us to join two different indexes and that's why I was trying to connect MYSQL directly to Kibana. Could you please guide me, on how to join two different indexes in Elasticsearch or how to get the data from the two different indexes in a single visualization?
Elasticsearch does not support query time joins between indices... this is a fundamental aspect of Elasticsearch.
So if that is your sole requirement elasticsearch may not be the right choice.
BUT depending on what you are actually trying to accomplished elasticsearch may be able to support the outcome you are looking for.
Data can be denormalized / or enriched (which is join on ingest) on ingest.
Or depending on the data and the visualization you are looking for there may be other approaches that work.
Now that we have established that Kibana does not read from RDBMS and Elasticsearch does not do query time joins.
My suggestion would be to open a new topic with a good title samples of the data and what you want to visualize and perhaps someone can help you.
I guess I would ask if you already have your data MYSQL then why not use on of the many visualization tools for RDBMS
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.