Query ClickHouse/Hydrolix tables as Elasticsearch indices

WARNING

Note: If you are using Hydrolix as your data source please follow to Query Hydrolix tables as Elasticsearch indices subsection of this document for more information. If you plan to use ClickHouse, continue below.

In this scenario, user's only data source is ClickHouse, which contains two tables named logs and siem. Quesma is configured to expose ClickHouse tables via Elasticsearch API, making them visible in Kibana/OpenSearch Dashboard (OSD) as Elasticsearch indices and also to ingest data to these two tables in ClickHouse only.

If you are interesed in more details about ingesting your data, refer to Ingest page.

Note

In this approach, Quesma still requires a minimal Elasticsearch/OpenSearch cluster for:

• storing Kibana/OSD metadata
• storing user data and authentication (unless its explicitly disabled)
• storing any additional indexes appearing at ingest endpoint and not explicitely listed in Quesma ingest processor

Architecture diagram:

Quesma installation

Prerequisites:

• ClickHouse is running.
• Kibana/OSD and minimal instance of 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 `clickhouse` type, 
   # please contact Quesma support (support@quesma.com) to obtain yours
   # in case you are using self-hosted clickhouse instance, you can use `clickhouse-os` backend connector and continue without the license key  
   frontendConnectors:
     - name: elastic-query
       type: elasticsearch-fe-query
       config:
         listenPort: 8080
    - name: elastic-ingest
       type: elasticsearch-fe-ingest
       config:
         listenPort: 8080
   backendConnectors:
     - name: minimal-elasticsearch
       type: elasticsearch
       config:
         url: #PLACE_YOUR_ELASTICSEARCH_URL_HERE, for example: http://192.168.0.7:9200
         user: #PLACE_YOUR_ELASTICSEARCH_USERNAME_HERE
         password: #PLACE_YOUR_ELASTICSEARCH_PASSWORD_HERE
     - name: clickhouse-instance
       type: clickhouse-os
       #type: clickhouse # use for ClickHouse cloud service only
       config:
         url: #PLACE_YOUR_CLICKHOUSE_URL_HERE, for example: clickhouse://192.168.0.7:9000
         user: #PLACE_YOUR_CLICKHOUSE_USER_HERE
         password: #PLACE_YOUR_CLICKHOUSE_PASSWORD_HERE
         database: #PLACE_YOUR_CLICKHOUSE_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 ClickHouse instance
             target:
               - clickhouse-instance
           logs:
             target:
               - clickhouse-instance
           '*':        # DO NOT remove, always required
             target:
               - minimal-elasticsearch
     - name: ingest-processor
       type: quesma-v1-processor-ingest
       config:
         indexes:      # `siem` and `logs` are just examples, 
           siem:       # make sure to replace them with your actual table or index names in your ClickHouse instance
             target:
               - clickhouse-instance
           logs:
             target:
               - clickhouse-instance
           '*':        # DO NOT remove, always required
             target:
               - minimal-elasticsearch
   pipelines:
     - name: elasticsearch-proxy-read
       frontendConnectors: [ elastic-query ]
       processors: [ query-processor ]
       backendConnectors: [ minimal-elasticsearch, clickhouse-instance ]
    - name: elasticsearch-proxy-write # remove if you do NOT want to write data to ClickHouse
       frontendConnectors: [ elastic-ingest ]
       processors: [ ingest-processor ]
       backendConnectors: [ minimal-elasticsearch, clickhouse-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 using ELASTICSEARCH_HOSTS environment variable.

     If you use Elasticsearch/Kibana without authentication, please modify the frontendConnectors section in the following way:

     frontendConnectors:
       - name: elastic-query
         type: elasticsearch-fe-query
         config:
           listenPort: 8080
           disableAuth: true
       - name: elastic-ingest
         type: elasticsearch-fe-ingest
         config:
           listenPort: 8080
           disableAuth: true

   • 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.

     If you use OpenSearch/OpenSearchDashboards without authentication, please modify the frontendConnectors section in the following way:

     frontendConnectors:
       - name: elastic-query
         type: elasticsearch-fe-query
         config:
           listenPort: 8080
           disableAuth: true
       - name: elastic-ingest
         type: elasticsearch-fe-ingest
         config:
           listenPort: 8080
           disableAuth: true

4. Restart Kibana/OSD.

5. Add DataViews/Index Patterns:

   • For Kibana: in order to view your ClickHouse 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.