Documentation¶
Click-Dog reads native query spans and query-log context from ClickHouse, exports them to OTLP-compatible backends and Splunk HEC, and provides local query-family, baseline, and regression analysis. Start with the guided Install, then branch into your backend, privacy policy, and operations from the sections below.
Get started¶
-
From a fresh host to a verified, useful systemd service.
-
Ansible · Kubernetes · Docker
Self-contained deployment guides for fleets and orchestrators.
-
The full YAML reference: every key, default, and env-var override.
-
Choose raw, redacted, normalized-only, or no query text and understand each data boundary.
Integrations¶
-
OTLP export plus three provisioned dashboards for query analysis, exported user activity, and Click-Dog self-monitoring.
-
Export traces to Honeycomb through an OpenTelemetry Collector.
-
Send to any OTLP/gRPC backend: Tempo, Jaeger, Elastic APM, and more.
-
Fan out span-shaped query events to Splunk HTTP Event Collector.
Query analysis¶
-
Group query families, compare against an explicit known-good baseline, gate on regression findings, and notify webhooks or Datadog Events.
-
Every attribute carried on the spans Click-Dog exports.
Operate¶
-
Watch Click-Dog itself: structured logs, Prometheus metrics, webhooks.
-
Check source readiness, test exporter acceptance, and prove native parent-child tracing before enabling scheduled mode.
-
Circuit breaker and adaptive backoff that protect ClickHouse and Click-Dog.
-
Diagnose common issues, starting with debug logging.
-
What the public beta guarantees and what it does not.
-
Config-aware systemd updates, deployment-specific updates, and rollback.
Reference¶
-
Scheduled, backfill, validation, dry-run, smoke testing, analysis, and manifest generation.
-
The layers that decide which spans get exported.
-
How the single Go binary sits beside your database.
-
What shipped in the current release.
-
Community help and direct production support from Click-Dog's maintainer.