Ingesting data

Fully managed
Self-managed

ClickHouse Observability includes a managed OpenTelemetry (OTel) collector that receives logs, metrics, traces, and session replay data and writes them to ClickHouse Cloud. ClickHouse Cloud operates the ingestion collector, database, and ClickStack UI. You instrument your applications and configure them or your data collection agents to send telemetry to the managed service.For more details about these components, see Architecture.

Find your connection details

In your service, open Team Settings → Data and find Managed Telemetry. This section provides the OTLP endpoint and API key for sending telemetry to your service.Select URL, Collector config, or Env vars to view the connection details in the format you need, then use Copy connection details. Configure your SDK or collector with the provided endpoint and authentication settings.See Managed Telemetry for details. You can use Send a Test Log to submit a sample log to the service.

Choose how to send telemetry

Application instrumentation

Language SDKs collect telemetry from inside your applications, such as request traces and application logs. Configure your SDK to export telemetry using OTLP to the managed endpoint, or to a collector in your environment that forwards it to the service.You can use existing OpenTelemetry instrumentation. ClickStack SDKs also provide features such as browser session replay; see Session replay for a managed setup example.

Infrastructure and collection agents

Run collection agents in your environment to gather telemetry such as host metrics, container logs, and Kubernetes events. An OpenTelemetry collector can run alongside an application, on a host, or as a Kubernetes DaemonSet, depending on what you need to collect.Configure these agents to forward telemetry to the managed OTLP endpoint using your service’s connection details. You can also route application telemetry through these collectors for processing before it reaches ClickHouse Observability.

Verify ingestion

After configuring your application or collector, generate some telemetry and open Search. Select the Logs or Traces source, choose a time range that includes the new events, and search for your service name to confirm that the data has arrived.

All data is ingested into Self-managed ClickHouse Observability via an OpenTelemetry (OTel) collector. This acts as the primary entry point for logs, metrics, traces, and session data.You run an OpenTelemetry Collector to receive telemetry data from your applications. With Self-managed ClickHouse Observability, ClickHouse and the UI are hosted and managed in ClickHouse Cloud, with integrated authentication and operational management handled for you.The collector exposes two OTLP endpoints:

HTTP - port 4318
gRPC - port 4317

You can send data to these endpoints either directly from language SDKs or OTel-compatible data collection agents e.g. other OTel collectors collecting infrastructure metrics and logs.More specifically:

Language SDKs are responsible for collecting telemetry from within your application - most notably traces and logs - and exporting this data to the OpenTelemetry collector, via the OTLP endpoint, which handles ingestion into ClickHouse. For more details on the language SDKs available see SDKs.
Data collection agents are agents deployed at the edge — on servers, Kubernetes nodes, or alongside applications. They collect infrastructure telemetry (e.g. logs, metrics) or receive events directly from applications instrumented with SDKs. In this case, the agent runs on the same host as the application, often as a sidecar or DaemonSet. These agents forward data to the central ClickStack OTel collector, which acts as a gateway, typically deployed once per cluster, data center, or region. The gateway receives OTLP events from agents or applications and handles ingestion into ClickHouse. See OTel collector for more details. These agents can be other instances of the OTel collector or alternative technologies such as Fluentd or Vector.

OpenTelemetry compatibilityWhile ClickStack offers its own language SDKs and a custom OpenTelemetry collector, with enhanced telemetry and features, you can also use your existing OpenTelemetry SDKs and agents seamlessly.

Introduction

Features

Administration

Guides

Integration partners

Self-managed

Find your connection details

Choose how to send telemetry

Application instrumentation

Infrastructure and collection agents

Verify ingestion

​Find your connection details

​Choose how to send telemetry

​Application instrumentation

​Infrastructure and collection agents

​Verify ingestion

Find your connection details

Choose how to send telemetry

Application instrumentation

Infrastructure and collection agents

Verify ingestion