Deployments guide

Slipway’s deployment engine is built on the same objects Octopus Deploy users know: services, environments, releases, channels and lifecycles. Pipelines feed it; it does not need a pipeline to run.

The flow

release ──▶ environment ──▶ [ protection rules ] ──▶ strategy ──▶ verification ──▶ done
                              approvals                rolling       metrics         │
                              windows, freezes         blue/green    thresholds      ▼
                              policies (CEL)           canary        auto-rollback   promote ▶ next phase
                              change tickets

1. A release exists (immutable).
2. Something requests a deployment of it to an environment: a pipeline’s deploy job, a promotion, a person, or a rollback.
3. The environment’s protection rules are evaluated. The deployment sits in waiting_approval or waiting_window until they pass.
4. The strategy runs on the environment’s infrastructure, through a runner that can reach it.
5. For canary and progressive strategies, verification checks metrics at each step. A failed verdict rolls back automatically.
6. On success, the release may auto-promote to the next phase of its lifecycle after a soak period.

Statuses

pending, waiting_approval, waiting_window, in_progress, verifying, paused, succeeded, failed, rolled_back, cancelled. You can pause, resume, retry, cancel and roll back from the UI, CLI or API.

Pages in this guide

Page	What you will learn
Strategies	Recreate, rolling, blue/green, canary, progressive
Approvals	Single and multi-approver, by role or named users, Slack and Teams buttons
Gates	Manual, scheduled, policy (CEL and OPA), change-ticket and metric gates
Canary verification	Queries, thresholds, intervals, verdicts and rollback
Lifecycles and channels	Phases, auto-promotion, soak time and version rules
Variables	Scoped, templated, secret-aware variables and previews
Runbooks	Parameterised operations with RBAC and approvals
Operations	Rollback, diff, freeze calendar, drift detection and DORA metrics

Two ways to drive it

From a pipeline (deploy job):

- name: deploy
  environment: { name: production, service: api }
  steps:
    - uses: slipway/deploy@v1
      with: { strategy: canary }

From releases (Octopus style): pipelines only create releases, and lifecycles plus promotion drive deployment outside the pipeline.

slipway release create api 1.4.0 --artifact image=ghcr.io/acme/api@sha256:abc...
slipway deploy api 1.4.0 --env staging
slipway promote api --from staging --to production

Both are supported and both write to the same deployment history.

Plan availability

	Free	Team	Business	Enterprise
Rolling, recreate	Yes	Yes	Yes	Yes
Blue/green, canary with metric verification	-	Yes	Yes	Yes
Progressive multi-region, release trains	-	-	Yes	Yes
Custom strategies	-	-	-	Yes
Approvals and manual gates	Yes	Yes	Yes	Yes
Policy-as-code gates (CEL)	3 policies	Unlimited	Unlimited	Unlimited
Change management (ServiceNow, Jira)	-	-	Yes	Yes