Skip to content

Deployment Overview

This section helps you choose the right deployment target for Validibot.

These pages are meant for customer-facing self-host deployments. Hosted-cloud operator workflows belong in the private internal docs, not here.

If you are new to Validibot and just want to get it running, start with Run Validibot Locally.

The three deployment targets

Validibot has three deployment targets that share the same Django codebase but have different audiences and substrates:

Target Substrate Driver Audience Stages
local docker-compose.local.yml on the developer's laptop just local <cmd> one Validibot developer testing the app dev only
self-hosted docker-compose.production.yml on a single Linux VM just self-hosted <cmd> a customer running their own copy on DigitalOcean, AWS EC2, on-prem single env per VM
GCP Cloud Run, Cloud SQL, Cloud Tasks, GCS just gcp <cmd> the Validibot team operating its own cloud dev / staging / prod

The same operator capabilities (bootstrap, doctor, smoke-test, backup, restore, upgrade, collect-support-bundle) exist for both self-hosted and GCP. Cross-target parity is a quality gate: if a recipe exists for self-hosted but not GCP (or vice versa), assume the design is wrong until proven otherwise.

Choose a target

Target Best for Start here
Local first-time evaluation, local sandboxing, development Run Validibot Locally
Self-hosted single-host self-hosting on a VPS, VM, or on-prem server Self-Hosting Overview (operator-facing) or Deploy with Docker Compose (developer-facing)
GCP managed cloud deployment on Google Cloud Deploy to GCP
AWS future target, not yet implemented Deploy to AWS

Command style

All deployment targets use the Just command runner.

Typical commands look like this:

just local up
just self-hosted bootstrap
just gcp deploy-all dev

Which page should I read first?

Use this shortcut:

  • If you want the quickest path to a running app on your machine, read Run Validibot Locally.
  • If you want a production deployment on infrastructure you control, read Deploy with Docker Compose.
  • If you want a managed cloud deployment on Google Cloud, read Deploy to GCP.
  • If you need AWS specifically, read Deploy to AWS and plan on using Docker Compose on an AWS host for now.

Once you have chosen a target, these supporting guides become relevant: