Kubernetes Installation

Run DataBuff on a Kubernetes cluster — Doris, Ingest, and Web are deployed automatically in order.

1. Prerequisites

• A Kubernetes cluster
• kubectl with cluster access
• Recommended: 8C16G (8 CPU cores / 16 GB RAM)

2. Install the Platform

curl -fsSL https://databuff.ai/databuff/ai-apm-k8s-install.sh | bash

After installation, the terminal prints the Web UI URL, namespace, and access details. Default namespace: databuff.

Purpose	Address
Web UI	http://<node-ip>:32703
Default login	admin / Databuff@123
OTLP gRPC	<node-ip>:30417
OTLP HTTP	http://<node-ip>:30418

In-cluster agents: http://ai-apm-ingest:4318 (gRPC ai-apm-ingest:4317). NodePorts are for external access. To connect your own apps, see OpenTelemetry OTLP Ingestion; for zero-instrumentation collection see eBPF Ingestion, for entry Nginx see Nginx Ingestion.

Offline Images (Optional)

When nodes cannot pull images:

curl -fsSL https://databuff.ai/databuff/ai-apm-k8s-download-images.sh | bash

See Kubernetes Operations — Offline Images.

Install a specific version:

curl -fsSL https://databuff.ai/databuff/ai-apm-k8s-install.sh | bash -s -- --version 0.1.10
# or
APM_VERSION=0.1.10 curl -fsSL https://databuff.ai/databuff/ai-apm-k8s-install.sh | bash

Operations: Kubernetes Operations Reference — start/stop, NodePort access, offline images, scaling.

3. Install the Demo (Optional)

Let the demo app report traces to the platform and quickly see call chains and topology.

curl -fsSL https://databuff.ai/databuff/ai-apm-demo-k8s-install.sh | bash

4. Post-Install Verification

1. Open http://<node-ip>:32703 and sign in with the default account
2. APM → Service List — confirm demo services have data
3. (Optional) Configuration → Model Settings — enter your API key to enable AI:

Example AI query: "Query service list for the last hour"