OTel Studio

Generate realistic OpenTelemetry data for any observability stack

OTel Studio is a Docker-deployable telemetry data generator that produces correlated traces, logs and metrics using realistic service topologies. Point it at any OTel-compatible backend and get production-like data flowing in 60 seconds.

Quick Start

docker run -d --name otel-studio -p 3002:3002 \
  -e OTEL_ENDPOINT="http://your-collector:4317" \
  -e OTEL_PROTOCOL="grpc" \
  -e ADMIN_PASS="your-password" \
  otel-studio:latest

Open http://localhost:3002, log in with admin / your password, enable a scenario and click Start.

Navigating the App

OTel Studio uses a slim icon rail on the left to switch between five pages. Each gets the full window width — nothing is squeezed into a sidebar.

The top control bar (Real System Data / Simulation toggles, signal switches, quota timer, Inject Anomaly, global rate multiplier, Start/Stop/Reset) stays visible on every page — these are global controls, not scoped to any one page.

Architecture

OTel Studio uses the official OpenTelemetry JS SDK. Each scenario creates a BasicTracerProvider per service name, with one BatchSpanProcessor per active endpoint. Logs use LoggerProvider with one BatchLogRecordProcessor per active endpoint. Metrics use a shared MeterProvider with one PeriodicExportingMetricReader per active endpoint — this is what makes multi-endpoint fanout work for all three signal types, not just traces.

All three signals are correlated — logs emitted during a trace span inherit the span's trace ID and span ID via OpenTelemetry context propagation. This means you can search for a trace ID in your backend and see the matching logs and metrics.

Docker Configuration

docker run -d --name otel-studio -p 3002:3002 \
  -e OTEL_ENDPOINT="http://otelcol:4317" \
  -e OTEL_PROTOCOL="grpc" \
  -e SERVICE_NAME="otel-agent" \
  -e ENVIRONMENT="production" \
  -e ADMIN_PASS="secure-password" \
  -e MODE_REAL="true" \
  -e MODE_SIM="true" \
  -v /var/log:/var/log:ro \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v otel-data:/data \
  --memory=256m \
  otel-studio:latest

25 Scenarios

Each scenario simulates a realistic service topology with parent-child span relationships, correlated log records per service, and matching metric instrument recordings.

Web App

Infrastructure

APM

Cloud Native

Chaos / Stress

Signal Types & Ratios

Per the OpenTelemetry specification, the three signal types have different volume characteristics:

Each of Traces, Logs, and Metrics can be toggled independently in the top control bar — turning a signal off stops it from being generated across every active scenario at once.

Advanced Scenario Controls

Every scenario card has an Advanced Settings panel with telemetrygen-style controls for testing specific pipeline behaviors:

Geo / IP Simulation

Tags the root span and logs of each generated request with a client.address and GeoIP-resolvable attributes (geo.country.iso_code, geo.city.name, geo.location.lat/lon, cloud.region), so requests appear to originate from realistic global locations. Two modes, set globally on the Endpoints page:

Synthetic PII

Tags the root span and logs with a synthetic user.id, user.name, and user.email, generated from a fixed name pool and RFC 2606-reserved domains (example.com/.org/.net) — never real people's data. Every synthetic record is marked with pii.synthetic: true so a downstream redaction/DLP pipeline being tested can positively identify it. Recommended for scenarios like Banking, Social Network, and E-Commerce where testing PII handling is relevant.

Custom Attributes

Arbitrary key=value pairs, comma-separated, added to the root span and logs of every request in that scenario — useful for tagging generated data with a team name, region, or any attribute your dashboards filter on.

Span Duration Override

Set a min/max range in milliseconds to override the scenario's natural randomized timing. Leave blank to use built-in realistic chain timing.

Status Code Weights

code:weight pairs, comma-separated (e.g. 200:90,404:5,500:5) — controls the distribution of HTTP status codes on generated spans. When set, this overrides the scenario's plain error-rate slider.

Anomaly Injection

Click ⚡ Inject Anomaly to trigger a burst across all active scenarios. For 30 seconds (or until you click Stop Anomaly):

Anomaly state is tracked server-side and stays in sync with the main Start/Stop control — stopping the simulation (manually or via quota expiry) also clears an in-progress anomaly, and the button/timer reflect this on any connected browser tab, not just the one that triggered it.

Sequential Mode

Enable Sequential Mode in the Scenarios panel to run one scenario at a time, rotating through all enabled scenarios on a configurable timer (default 30 seconds each). Useful for sequential load testing — each scenario runs in isolation before rotating to the next.

Time Limit / Quota

Set a time limit (1m / 5m / 10m / 30m / 1h / no limit) before clicking Start. The top bar shows a live countdown; when it reaches zero, sending stops automatically across all signals — including metrics, which export on their own periodic schedule independent of the trace/log generation loop.

Multi-Endpoint Fanout

OTel Studio can send telemetry to multiple collectors simultaneously. On the Endpoints page, each endpoint has its own toggle — check any combination to activate them together; at least one must stay active. The SDK creates one BatchSpanProcessor, one BatchLogRecordProcessor, and one metric reader per active endpoint, so traces, logs, and metrics all fan out identically. Use this to compare two o11y backends side-by-side or to send to both production and staging collectors at once.

TLS / mTLS

Each endpoint supports an optional TLS/mTLS configuration, expandable in the Add/Edit Endpoint form: a CA certificate, a client certificate, a client private key (all PEM format), and an "insecure" toggle to skip certificate verification for testing against self-signed backends.

Stack Presets

One-click endpoint configuration for 12 popular observability backends:

Webhook on Quota Expire

Set an HTTP endpoint (Slack incoming webhook or any URL) on the Endpoints page. When the time-limit quota expires and sending stops automatically, OTel Studio fires a POST with {event: "quota_expired", ts: "<ISO timestamp>"} — useful for notifying a team channel when an automated test run finishes.

Real System Data

Alongside simulated scenarios, OTel Studio can collect real telemetry from the host it's running on, on the Real Data page:

Real System Data and Simulation are independent toggles — run either alone or both together.

Import / Export

Export your scenario configuration as JSON to share setups with your team. Import a config file to restore a saved setup. Export the activity stream as CSV for offline analysis.

Environment Variables

OTel Studio v2.0 — Built by XplurData