Skip to content

Latest commit

 

History

History
238 lines (191 loc) · 6.56 KB

README.md

File metadata and controls

238 lines (191 loc) · 6.56 KB

otelgen

A simple and reliable CLI tool for generating OpenTelemetry traces, metrics, and logs to test OTLP endpoints.

Features

  • Generate traces, metrics, and logs
  • Support for gRPC, gRPCs, HTTP, and HTTPS protocols
  • Automatic port selection (443 for gRPC/gRPCs/HTTPS, 80 for HTTP)
  • Configurable rate and duration
  • Simple and intuitive CLI interface

Installation

Build from source

go build -o otelgen ./cmd/otelgen

Build Docker image

docker build -t otelgen .

The Docker image uses a multi-stage build with scratch as the final base image for minimal size (~20MB).

Usage

Traces

Generate trace data:

# Using gRPCs (secure gRPC)
./otelgen traces \
  --otlp-endpoint grpcs://example.com:443 \
  --service my-service \
  --rate 1 \
  --duration 10s

# Using HTTP
./otelgen traces \
  --otlp-endpoint http://localhost:80 \
  --service my-service \
  --rate 10 \
  --duration 1m

# Port is optional - will use default 443 for gRPC
./otelgen traces \
  --otlp-endpoint grpc://localhost \
  --service my-service

Metrics

Generate metrics data:

# Using gRPCs
./otelgen metrics \
  --otlp-endpoint grpcs://example.com:443 \
  --service my-service \
  --rate 1 \
  --duration 10s

# Using HTTPS
./otelgen metrics \
  --otlp-endpoint https://localhost:443 \
  --service my-service \
  --rate 5 \
  --duration 30s

Logs

Generate log data:

# Using gRPCs
./otelgen logs \
  --otlp-endpoint grpcs://example.com:443 \
  --service my-service \
  --rate 1 \
  --duration 10s

# Using HTTP with custom port
./otelgen logs \
  --otlp-endpoint http://localhost:8080 \
  --service my-service \
  --rate 20 \
  --duration 5s

Docker Usage

# Traces
docker run --rm otelgen traces \
  --otlp-endpoint grpcs://your-endpoint:443 \
  --service my-service \
  --rate 1 \
  --duration 10s

# Metrics
docker run --rm otelgen metrics \
  --otlp-endpoint grpcs://your-endpoint:443 \
  --service my-service \
  --rate 1 \
  --duration 10s

# Logs
docker run --rm otelgen logs \
  --otlp-endpoint grpcs://your-endpoint:443 \
  --service my-service \
  --rate 1 \
  --duration 10s

Flags

Flag Description Default Required
--otlp-endpoint OTLP endpoint URL (grpc://, grpcs://, http://, https://) - Yes
--service Service name for telemetry otelgen No
--rate Number of telemetry items per second 1 No
--duration How long to generate telemetry (e.g., 10s, 1m, 1h) 10s No
--size Payload size to increase data volume (e.g., 1kb, 1mb, 500b) - No
--batch-size Maximum number of logs to batch before sending (logs only) 512 No
--headers Additional headers (e.g., key1=value1,key2=value2) - No
--verbose Enable verbose logging false No
--insecure-skip-verify Skip TLS certificate verification (insecure) false No

Protocol Support

  • grpc:// - Insecure gRPC (default port: 443)
  • grpcs:// - Secure gRPC with TLS (default port: 443)
  • http:// - Insecure HTTP (default port: 80)
  • https:// - Secure HTTPS with TLS (default port: 443)

Default Ports

If you don't specify a port in the endpoint URL, the following defaults are used:

  • gRPC/gRPCs: 443
  • HTTP: 80
  • HTTPS: 443

Examples

# Quick test with local collector
./otelgen traces --otlp-endpoint grpc://localhost --service test-app --duration 5s

# High-volume test
./otelgen metrics --otlp-endpoint http://localhost:80 --service load-test --rate 100 --duration 1m

# Production endpoint test
./otelgen logs --otlp-endpoint grpcs://prod.example.com:443 --service prod-app --rate 10 --duration 30s

# Test with increased payload size (1KB per trace)
./otelgen traces --otlp-endpoint grpc://localhost --service test-app --size 1kb --duration 10s

# Load test with large payloads (1MB per metric)
./otelgen metrics --otlp-endpoint http://localhost:80 --service load-test --size 1mb --rate 10 --duration 30s

# Test with custom payload size (500 bytes)
./otelgen logs --otlp-endpoint grpcs://example.com:443 --service test-app --size 500b --rate 5 --duration 1m

# Large logs with smaller batch size to avoid gRPC message size limit (4MB)
# For 1MB logs, use batch size of 3 to keep messages under 4MB
./otelgen logs --otlp-endpoint grpcs://example.com:443 --service test-app --size 1mb --rate 10 --batch-size 3 --duration 10s

What Gets Generated

Traces

  • Parent spans with child spans
  • Random operation types and IDs
  • Realistic timing and nesting
  • Optional payload padding via attributes when --size is specified

Metrics

  • Counter: otelgen.requests
  • Histogram: otelgen.duration
  • Gauge: otelgen.cpu_usage
  • Optional payload padding via attributes when --size is specified

Logs

  • Proper OTLP log records with resource attributes
  • Various log levels (INFO, WARN, ERROR, DEBUG) mapped to appropriate severity
  • Log body contains realistic JSON structured data including:
    • Timestamp, service name, environment, version
    • HTTP request details (method, endpoint, status code, duration, user agent, client IP)
    • User information (ID, email, role, organization)
    • Error details with stack traces (for ERROR level)
    • Database query metrics (30% of logs)
  • Additional attributes: component, request_id, user_id
  • When --size is specified, the JSON body is expanded to reach target size
  • Batch Size: Logs are batched before sending to improve efficiency. The default batch size is 512 logs. When using large log sizes (e.g., --size=1mb), you should reduce the batch size using --batch-size to avoid exceeding the gRPC message size limit (typically 4MB). For example, with 1MB logs, use --batch-size=3 to keep messages under the limit.

Sample Log Output

{
  "timestamp": "2025-01-23T10:15:30.123456789Z",
  "level": "INFO",
  "message": "Request processed successfully",
  "service": "api-gateway",
  "environment": "production",
  "version": "v1.2.3",
  "host": "server-3",
  "pod_id": "pod-42-aBcDeFgH",
  "request_id": "req-XyZ123AbC456-1737628530",
  "trace_id": "a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6",
  "span_id": "q1w2e3r4t5y6u7i8",
  "http": {
    "method": "POST",
    "endpoint": "/api/v1/orders",
    "status_code": 201,
    "duration_ms": 234,
    "user_agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36",
    "client_ip": "10.142.15.89"
  },
  "user": {
    "id": "user_1234",
    "email": "user1234@example.com",
    "role": "user",
    "org_id": "org_56"
  },
  "database": {
    "query_time_ms": 45,
    "rows_affected": 1,
    "connection_id": 23,
    "database": "orders_db"
  }
}