Tutorial

A guided tour through every part of Sonda you will reach for in real testing work. Each page below adds one concept on top of the last; by the end, you can build multi-signal scenarios, push them to a real backend, and drive them from the HTTP API.

This tutorial picks up where Getting Started leaves off. You have Sonda installed and have run your first metric and log; now you want to know what every knob does.

What you need:

• Sonda installed -- see Getting Started -- Installation.
• Docker -- only for the Server API page.

The tour

1. Generators -- the eight value shapes Sonda can produce, when to reach for each, and how jitter adds realism.
2. Encoders -- wire formats your backend speaks: Prometheus text, InfluxDB line protocol, JSON lines, syslog, OTLP, remote write.
3. Sinks -- where the encoded bytes go: stdout, files, HTTP, TCP/UDP, Loki, Kafka, OTLP/gRPC.
4. Generating logs -- template mode with field pools, replay mode, and pairing with the syslog encoder.
5. Scheduling -- gaps and bursts -- inject the irregularities real telemetry has, so your alerts and pipeline see real-shaped data.
6. Multi-scenario runs -- sonda run, phase offsets, and clock_group for compound-alert testing.
7. The Server API -- submit scenarios over HTTP, scrape live stats, manage long-running runs.

After the tour

When you have walked through all seven pages you have everything you need to:

• Push synthetic data into a real backend end-to-end -- the canonical start-the-stack, push, query loop with VictoriaMetrics, Loki, Kafka, and OTLP.
• Test alert rules with predictable threshold crossings.
• Validate a pipeline change end-to-end.
• Plan capacity for high-volume runs.

If you would rather not write YAML by hand, run sonda init for an interactive wizard, or browse Built-in Scenarios for ready-to-run patterns.