#Next Steps

You built a complete fleet from the ground up: you imported real Terraform state as entities, gave them proper entity types, grouped your clusters into a cell that decides placement, connected tenants to clusters with the runs_on relationship, and wired an OnboardTenant Action that opens a pull request your CI applies. That's the whole cattle loop — see it, place into it, connect it, grow it — running on your machine, with Terraform still in charge of every apply.

Here's where to go from here, depending on what you want next.

#Go deeper on the end-to-end flow

The Quick Start — Advanced tutorial walks a multi-stack greenfield scenario: model several cells, fire an OnboardTenant cascade that opens PRs across dependent stacks, and let your CI apply them in dependency order. It's the same primitives you just learned, scaled up to a realistic onboarding cascade. If your real fleet has tenants whose infrastructure spans more than one stack, start there.

If you came in through the five-minute on-ramp and want to revisit the fast path, the Quick Start — Simple page is the canonical import → dashboard flow.

#Now do it well

You can build a fleet; the Best Practices section is about building one that stays healthy as it grows:

• Modeling entities & cells — how to choose entity-type boundaries and cell granularity so the model fits how your team actually thinks about the fleet.
• Relationship & cascade design — cardinality, cascade phases, and dependsOn ordering that tears down cleanly every time.
• Actions & PR hygiene — keeping the PRs your Actions open reviewable, scoped, and easy for CI to apply.
• Multi-tenancy & self-hosting — running the whole thing yourself, with zero SaaS dependencies.

#Look up the exact details

When you need the precise field name, flag, or function — not the narrative — the Reference is the lookup surface:

• Catalog Schema Reference — every field of every kind, one page each: EntityType, Cell, RelationshipType, Action, Secret, and the apply primitive.
• Triggers Reference — the full menu of triggers (pull-request, Terraform Cloud, Atlantis, Atmos workflow, webhook) and the GitHub App that powers PR-based triggers end to end.
• Interpolation Reference — every {{ ... }} function you can use in parameters, properties, and triggers.
• CLI Reference — every command and flag, including the full surface of import terraform, apply, secrets, and dashboard.
• Configuration Reference — project config, environment variables, and tokens for pointing the CLI at a remote project.

#When something goes wrong

The Troubleshoot section covers the failure modes you're most likely to hit: import / ingest issues, Action / PR debugging, and drift & state issues.

#See it all at full size

The canonical demos in examples/ ship complete, runnable blueprints for each substrate — the same model you just built, in full:

• bare-tf-gh-actions-onefile-per-tenant — bare Terraform + GitHub Actions (pull-request trigger), the shape this tutorial mirrored.
• cattle-saas-tenants — Terraform Cloud (terraform-cloud trigger).
• cattle-saas-tenants-atmos — Atmos (atmos-workflow trigger).
• cattle-saas-tenants-atlantis — Atlantis (atlantis trigger).

Browse them in the Examples gallery, pick the one closest to how you run Terraform, and adapt its blueprint to your real fleet.

---

Prev: ← Step 4 — Wire an Action · Back to: Your First Fleet (overview)