Query Hydrolix tables as Elasticsearch indices

In this scenario, user's only data source is Hydrolix, which contains two tables named logs and siem. Quesma is configured to expose Hydrolix tables via Elasticsearch API, making them visible in Kibana/OpenSearch Dashboard (OSD) as Elasticsearch indices. Note: ingest to Hydrolix is currently not supported, let us know if you are interested in this functionality.

Note

In this approach, we require a minimal Elasticsearch/OpenSearch cluster for:

• storing Kibana/OSD metadata
• storing user data and authentication (unless its explicitly disabled)

Architecture diagram:

Quesma installation

Prerequisites:

• Hydrolix is running.
• Kibana/OSD and Elasticsearch/OpenSearch are running.
• Docker is installed, at least 20.10 version.

Installation steps:

1. Create a configuration file named quesma.yaml with the following content, make sure to replace placeholders (#PLACE_YOUR*) with actual values.

   licenseKey: #PLACE_YOUR_LICENSE_KEY_HERE 
   # license key is required for backend connector of `hydrolix` type, 
   # please contact Quesma support ([email protected]) to obtain yours  
   frontendConnectors:
     - name: elastic-query
       type: elasticsearch-fe-query
       config:
         listenPort: 8080
   backendConnectors:
     - name: minimal-elasticsearch
       type: elasticsearch
       config:
         url: #PLACE_YOUR_ELASTICSEARCH_URL_HERE
         user: #PLACE_YOUR_ELASTICSEARCH_USERNAME_HERE
         password: #PLACE_YOUR_ELASTICSEARCH_PASSWORD_HERE
     - name: hydrolix-instance
       type: hydrolix
       config:
         url: #PLACE_YOUR_HYDROLIX_URL_HERE
         user: #PLACE_YOUR_HYDROLIX_USER_HERE
         password: #PLACE_YOUR_HYDROLIX_PASSWORD_HERE
         database: #PLACE_YOUR_HYDROLIX_DATABASE_NAME_HERE
   processors:
     - name: query-processor
       type: quesma-v1-processor-query
       config:
         indexes:      # `siem` and `logs` are just examples, 
           siem:       # make sure to replace them with your actual table names in your Hydrolix instance
             target: [ hydrolix-instance ]
           logs:
             target: [ hydrolix-instance ]
           '*':       # DO NOT remove, always required
             target: [ minimal-elasticsearch ]
   pipelines:
     - name: elasticsearch-proxy-read
       frontendConnectors: [ elastic-query ]
       processors: [ query-processor ]
       backendConnectors: [ minimal-elasticsearch, hydrolix-instance ]

Note: To learn more about configuration options, refer to Configuration primer

2. Run Quesma with the following command:

   docker run --name quesma -p 8080:8080 \
    -e QUESMA_CONFIG_FILE=/configuration/quesma.yaml \
   -v $(pwd)/quesma.yaml:/configuration/quesma.yaml quesma/quesma:latest

   You have now a running Quesma instance running with Elasticsearch API endpoint on port 8080. You can also enable Quesma's admin panel at localhost:9999 by adding -p 9999:9999 to docker run command.
3. Reconfigure client endpoint:
   • For Kibana: update your Kibana configuration, so that it points to Quesma Elasticsearch API endpoint mentioned above, instead of Elasticsearch original endpoint. In your Kibana configuration file, replace the elasticsearch.hosts value with Quesma's host and port, e.g.:

   elasticsearch.hosts: ["http://quesma:8080"]

   or optionally use ELASTICSEARCH_HOSTS environment variable.
   • For OpenSearchDashboards: modify opensearch_dashboards.yml file and change opensearch.hosts property. Also, make sure to update the Elasticsearch/OpenSearch endpoint configuration for any clients ingesting data.
4. Restart Kibana/OSD.
5. Add DataViews/Index Patterns:
   • For Kibana: in order to view your Hydrolix tables in Kibana, you need to create Data Views for tables (indexes) from the config. If you're unsure how to do it, follow the Data Views creation guide for more information.
   • For OpenSearchDashboards: you may need to add Index Patterns. See instructions.