Important
Under active development. Not production-ready.
make start # Start K8s stack
make test # Run tests
make stop # Stop (preserves data)Run specific suites:
cargo test --test integration auth_jwt
cargo test --test integration vault_isolation
cargo test --test integration cache| Category | Tests | Scope |
|---|---|---|
| Authentication | 7 | JWT validation, Ed25519, expiration, scopes |
| Vault Isolation | 4 | Multi-tenant separation, cross-vault prevention |
| Cache Behavior | 4 | Hit/miss patterns, expiration, concurrency |
| Concurrency | 5 | Parallel requests, race conditions |
| E2E Workflows | 2 | Registration → authorization journeys |
| Management | 5 | Org suspension, client deactivation |
| Resilience | 6 | Recovery, degradation, error propagation |
| Script | Purpose |
|---|---|
k8s-local-start.sh |
Deploy stack to local K8s |
k8s-local-stop.sh |
Stop services, preserve data |
k8s-local-status.sh |
Check deployment health |
k8s-local-update.sh |
Rebuild and redeploy images |
k8s-local-purge.sh |
Remove all resources |
k8s-local-run-integration-tests.sh |
Execute test suite |
Tests run inside K8s using service DNS. Override for local development:
| Variable | Default (in K8s) | Purpose |
|---|---|---|
CONTROL_URL |
http://inferadb-control:9090 |
Control HTTP endpoint |
ENGINE_URL |
http://inferadb-engine:8080 |
Engine HTTP endpoint |
ENGINE_GRPC_URL |
http://inferadb-engine:8081 |
Engine gRPC endpoint |
ENGINE_MESH_URL |
http://inferadb-engine:8082 |
Engine mesh endpoint |
#[tokio::test]
async fn test_my_feature() {
let fixture = TestFixture::create().await.expect("setup failed");
let jwt = fixture.generate_jwt(None, &["inferadb.check"]).unwrap();
let response = fixture
.call_engine_evaluate(&jwt, "document:1", "viewer", "user:alice")
.await
.unwrap();
assert_eq!(response.status(), StatusCode::OK);
}| Issue | Solution |
|---|---|
| Services not starting | kubectl get pods -n inferadb && kubectl logs -n inferadb deployment/inferadb-engine |
| Port in use | make purge && make start |
| Tests timing out | Check Docker RAM (4GB+ recommended), check pod logs for errors |
Join us on Discord to discuss InferaDB, get help with your projects, and connect with other developers. Whether you have questions, want to share what you're building, or are interested in contributing, we'd love to have you!
Licensed under either of:
at your option.