Embedded in ClickHouse

ClickStack is bundled directly into the ClickHouse server binary. This means you can access the ClickStack UI (HyperDX) from your ClickHouse instance without deploying any additional components. This deployment is similar to the public demo at play-clickstack.clickhouse.com, but running against your own ClickHouse instance and data.

Suitable for

• Trying ClickStack with minimal setup
• Exploring your own ClickHouse data with an observability UI
• Demos and evaluations

Limitations

This embedded version is not designed for production use. The following features are not available compared to the production-ready OSS deployments:

• Alerting
• Dashboard and search persistence — dashboards and saved searches are not retained across sessions
• Customizable query settings
• Event patterns

Deployment steps

Docker

Start ClickHouse

Pull and run the ClickHouse server image with a password set:

docker run --rm -it -p 8123:8123 -e CLICKHOUSE_PASSWORD=password clickhouse/clickhouse-server:head-alpine

Running without a password

If you prefer to run without a password, you must explicitly enable default access management:

docker run --rm -it -p 8123:8123 -e CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT=1 clickhouse/clickhouse-server:head-alpine

Open the ClickStack UI

Open http://localhost:8123 in your browser and click ClickStack.

Enter the username default and the password password to connect to the local instance.

Create a source

If you have existing OpenTelemetry tables, ClickStack will detect them and create sources automatically.

On a fresh installation, you will be prompted to create a source. Complete the Table field with the appropriate table name (e.g. otel_logs) and click Save New Source.

If you don't have data yet, see Ingesting data for available options.

Binary

Start ClickHouse

Download and start ClickHouse:

curl https://clickhouse.com/ | sh

Optional: Enable system log tables

To explore ClickHouse's own internal logs and metrics, create a configuration snippet in your working directory before starting the server:

mkdir -p config.d && cat > config.d/query_logs.xml << 'EOF'
<clickhouse>
    <query_log>
        <database>system</database>
        <table>query_log</table>
    </query_log>
    <query_thread_log>
        <database>system</database>
        <table>query_thread_log</table>
    </query_thread_log>
    <query_views_log>
        <database>system</database>
        <table>query_views_log</table>
    </query_views_log>
    <metric_log>
        <database>system</database>
        <table>metric_log</table>
    </metric_log>
    <asynchronous_metric_log>
        <database>system</database>
        <table>asynchronous_metric_log</table>
    </asynchronous_metric_log>
</clickhouse>
EOF

With this enabled, you can create a Log Source pointing to system.query_log after opening ClickStack:

Setting	Value
Name	Query Logs
Database	system
Table	query_log
Timestamp Column	event_time
Default Select	event_time, query_kind, query, databases, tables, initial_user, projections, memory_usage, written_rows, read_rows, query_duration_ms

Start the server:

./clickhouse server

Open the ClickStack UI

Open http://localhost:8123 in your browser and click ClickStack. A connection to the local instance is created automatically.

Create a source

If you have existing OpenTelemetry tables, ClickStack will detect them and create sources automatically.

If you don't have data yet, see Ingesting data for available options.

Next steps

If you're ready to move beyond evaluation, consider a production-ready deployment:

• All-in-One — single container with all components, including persistence and authentication
• Docker Compose — individual components for more control
• Helm — recommended for production Kubernetes deployments
• Managed ClickStack — fully managed on ClickHouse Cloud