Stroppy Benchmark Action

GitHub Action for running Stroppy database benchmarks in CI/CD pipelines.

Stroppy is a command-line tool built on k6/xk6 for database load testing. This action installs Stroppy, runs benchmarks (custom scripts or built-in presets), and uploads results as workflow artifacts.

Usage

Run a built-in preset

- uses: stroppy-io/stroppy-action@main
  with:
    preset: tpcb
    driver-url: ${{ secrets.DATABASE_URL }}
    duration: 10m
    vus: 4
    scale-factor: 10

Run a custom script

- uses: stroppy-io/stroppy-action@main
  with:
    script: ./benchmarks/my-bench.ts
    sql-file: ./benchmarks/schema.sql
    driver-url: ${{ secrets.DATABASE_URL }}
    duration: 30m
    vus: 8

Pin a specific Stroppy version

- uses: stroppy-io/stroppy-action@main
  with:
    version: v1.2.3
    preset: simple
    driver-url: ${{ secrets.DATABASE_URL }}

Use outputs in subsequent steps

- uses: stroppy-io/stroppy-action@main
  id: bench
  with:
    preset: tpcb
    driver-url: ${{ secrets.DATABASE_URL }}

- run: echo "Benchmark exited with code ${{ steps.bench.outputs.exit-code }}"

Inputs

Input	Required	Default	Description
version	No	latest	Stroppy release tag (e.g. v1.2.3) or latest
script	No	''	Path to a custom .ts benchmark script (mutually exclusive with preset)
sql-file	No	''	Path to a .sql file (only used with script)
preset	No	''	Built-in preset: tpcb, tpcc, tpcds, simple, execute_sql
driver-url	Yes		Database connection URL
scale-factor	No	''	Scale factor for the workload
duration	No	''	Test duration (e.g. 1h, 30m)
vus	No	''	Number of virtual users
log-level	No	info	Log level: debug, info, warn, error
k6-args	No	''	Additional k6 CLI arguments
artifact-name	No	stroppy-results	Name for the uploaded results artifact

Either script or preset must be provided (but not both).

Outputs

Output	Description
exit-code	Exit code from the Stroppy run
results-file	Path to the JSON results file
artifact-id	ID of the uploaded artifact

How it works

1. Install — Downloads the Stroppy binary from GitHub Releases (with tool-cache for fast re-runs).
2. Run — Executes the benchmark:
   • Custom script: stroppy run <script> [sql] -- <k6-args>
   • Preset: generates files via stroppy gen, then runs them.
3. Collect — Sets outputs and uploads the JSON results file as a workflow artifact.

License

MIT