Testing & Simulation

Vier Test-Ebenen, alle pre-release verifiziert. Stand: 520 Rust-Tests grün + 3 ignored (regtest opt-in), 298 Svelte-Check-Files / 0 errors.

1. Unit-Tests

Pro Crate, reine Logik. cargo test --workspace läuft alle 16 Library- Crates + 5 Binaries.

Wichtige Test-Counts (Stand Phase 19):

Crate	Tests	Was
`warppool-config`	~24	TOML-Schema, coinbase_splits, VarDiffSettings::validate
`warppool-storage`	~75	sqlx-roundtrips, migrations, alle 10 Tabellen, vardiff, audit_log, api_tokens, admin_2fa, pool_settings
`warppool-api`	~80	axum-handler-Tests gegen in-memory SQLite + mock-GitHub/SMTP-Server. Inkl. Phase 16-Tests (pool_metrics + notifier in /metrics)
`warppool-stratum-v1`	~44 + 4 e2e	session-state-machine, ratelimit, vardiff (17 isoliert), TLS-roundtrip
`warppool-stratum-v2`	~71	noise-handshake, codec-roundtrips, session-frames, TDP-foundation
`warppool-translator`	~22	extended-channel-state-machine, BIP-320-version-rolling, slushpool-prev-hash-reverse
`warppool-share-validator`	~17	sha256d-vectors, dedup, network/pool-target-checks
`warppool-job-builder`	~32	coinbase-construction, BIP-34-height-encoding, merkle-branches
`warppool-bitcoin-rpc`	~14	RPC-envelope-parsing, ZMQ-frame-parsing
`warppool-notifier`	19	render_text, EmailSink/SlackSink-construction, sink-skip-when-env-missing
`warppool-autoupdate`	~25	version-compare, asset-matching, download_verified, atomic_swap, cosign-subprocess
`warppool-health`	~24	RPC-call, bitcoin-conf-parser, snippet-generator, warnings-aggregation
`warppool-telemetry`	~27	vendor-probes (AxeOS/cgminer-mock), mDNS-discovery, PoolMetrics+histogram
`warppool-profiles`	6	profile-resolution, retention-monotonie
`warppool-hwdetect`	6	sysinfo-detection, container-env, recommendation
`warppool-simulator`	~9	sim-miner-personas, scenarios

# Alles
cargo test --workspace

# Ein bestimmter Crate
cargo test -p warppool-share-validator

# Mit output
cargo test -p warppool-stratum-v2 -- --nocapture

2. Integration-Tests

Mehrere Crates in-process, axum-mock-server für externe Dienste. Pro Crate unter tests/.

Typische Cases:

Test	Crate	Was
`e2e.rs`	stratum-v1	TCP-Listener + ShareValidator + Sim-Miner-Roundtrip mit echtem subscribe/authorize/submit-Flow
`e2e_regtest.rs`	bitcoin-rpc	`#[ignore]` — gegen laufendes bitcoind regtest
`update_apply_with_cosign_verify_but_no_env_returns_500`	api	axum-mock-GitHub + mock-asset-server
`metrics_renders_pool_counters_when_pool_metrics_set`	api	/metrics-output gegen Prometheus-Format

3. Simulation (`dvb-warppool-sim`)

Gegen den echten Daemon-Prozess (oder in-process spawned). Szenarien sind End-to-End-Stories.

Eingebaute Szenarien

Szenario	Dauer	Status	Was es prüft
`solo-block-found`	30s	✓ executable	Sim-Miner connectet, Block-Found-Event landet
`connection-storm-defense`	60s	✓ executable	Rate-Limit greift bei N parallelen Auth-Versuchen
`enterprise-load-smoke`	5min	✓ executable	N parallele Miner gemischter Persona laufen sauber
`small-24h-stability`	24h	Plan-only	Klein-Profil bleibt unter 200 MB, keine Panics
`rpc-outage-recovery`	10min	Plan-only	Pool übersteht 5-Minuten-RPC-Outage

solo-block-found läuft self-contained — spawnt einen Stratum-V1-Server in-process mit network_target=[0xff;32] (jeder valid Share wird BlockFound) und einen JsonStratumMiner (Persona BitaxeUltra). Kein Bitcoin Core nötig:

dvb-warppool-sim scenario solo-block-found --duration 10
# → Result: BLOCK FOUND after 55 ms (worker=bc1qsim-bitaxe, job_id=sim-job-1)

dvb-warppool-sim scenario connection-storm-defense --duration 15 --attackers 200
# → Success: 10, RateLimit: 190, OtherErr: 0
# → OK: rate-limit hat 190x gegriffen (burst war 10)

dvb-warppool-sim scenario enterprise-load-smoke --duration 15 --miners 200
# → Accepted: 57400 (~3826 shares/s), Rejected: 0, Disconnects: 200

Personas

Sim-Miner ahmen reale Hardware nach (Hashrate, User-Agent, TLS-Support, Reconnect-Pattern):

Persona	Nominal HRate	TLS	Anmerkung
`nerdminer-v2`	50 GH/s	nein	Plain only
`bitaxe-ultra`	2 TH/s	ja	Sauberer Reconnect
`nerd-octaxe`	4.5 TH/s	ja	NerdNOS-Quirks
`avalon-q`	90 TH/s	nein	CGMiner-User-Agent
`antminer-s23-pro`	580 TH/s	ja	ASIC-Boost overt
`adversary`	—	—	Boshafte Inputs

Failure-Modes

dvb-warppool-sim failures <mode> injiziert Production-Probleme:

rpc-down — Bitcoin Core unreachable
zmq-disconnect — Block-Notify weg
tls-handshake-fail — abgelaufenes Cert
miner-hang — Connection hängt nach N Shares
network-latency — künstliche RTT
packet-loss — N% Paket-Loss
invalid-template — kaputte CB-Tx
reorg — Tip wird zurückgerollt
memory-pressure — Pool unter RAM-Druck
disk-full — SQLite-Write-Fail

Security-Suite

dvb-warppool-sim security fährt adversarielle Inputs (Line-Flood, Connection-Storm, Slowloris, Auth-Flood, Weak-Cipher, ...).

Correct outcome: jede Connection sauber abgelehnt, kein Panic, kein OOM.

4. Regtest E2E (Phase 15b — opt-in)

Für volles E2E mit echtem bitcoind regtest:

eval "$(scripts/regtest-up.sh)"   # startet bitcoind regtest + setzt env-vars
cargo test -p warppool-bitcoin-rpc --test regtest_e2e -- --ignored --nocapture
scripts/regtest-down.sh --purge   # cleanup

Drei #[ignore]-Tests gegen das laufende regtest:

Test	Was
`regtest_blockchain_info_returns_regtest_chain`	`chain == "regtest"`
`regtest_getblocktemplate_works`	Template hat version + bits + prev-hash
`regtest_submit_invalid_block_is_rejected`	Trivially invalid hex → Rejected (nicht Accepted, nicht Error)

CI führt sie nicht aus (kein bitcoind im Runner-Image). Lokal sind sie sehr nützlich vor Release-Tags. Siehe Configuration Reference → WARPPOOL_E2E_REGTEST_*.

5. Performance Benchmarks (Phase 19 — opt-in)

Criterion-Suites für Hot-Paths:

Crate	Bench	Hot-Path
`warppool-share-validator`	`validate`	per accepted+rejected share
`warppool-job-builder`	`build_job`	per neuem Template
`warppool-stratum-v1`	`vardiff`	per accepted share

cargo bench --workspace
# oder
cargo bench -p warppool-share-validator --bench validate

Siehe Performance Benchmarks für Baseline-Zahlen und Interpretation.

6. UI-Side (svelte-check)

cd ui
pnpm install
pnpm svelte-check
# → 294 files / 0 errors / 1 pre-existing warning

Pre-existing warning: Unused CSS selector "th.num" in blocks/+page.svelte — historisches Artefakt aus Phase 4, kein Funktionsproblem.

CI

.github/workflows/ci.yml Matrix:

cargo fmt --check (no diff)
cargo clippy --workspace --all-targets -- -D warnings
cargo test --workspace
cargo deny check (license + advisories)
cargo audit
UI-Job: pnpm install + pnpm svelte-check + pnpm build

Separate Workflows (siehe Architecture):

release.yml — Tag-Triggered: Multi-arch Docker + native binaries + SLSA-3 + SBOM
repro.yml — Reproducible-Builds-Verify (zweimal bauen + sha256-diff)
docs.yml — mdBook → GitHub Pages
benches.yml — Criterion (manual-dispatch + tag-push, nicht-PR)

dvb-WarpPool