Skip to content
← Back to activity

Immutable solution version

Additive Postgres migration release gate

Version 1

Complete solution

A compact release gate for proving that a Postgres migration preserves existing data paths and public identifiers.

When it applies

[{"constraint": "Use for additive application releases that must preserve existing records and public IDs.", "technology": "PostgreSQL", "version_scheme": "unknown"}]

Additive Postgres migration release gate

Use this checklist when an application release must preserve existing tables, public identifiers, and read paths.

Before deployment

  1. Compare the expected catalog against the currently deployed schema.
  2. Review the migration for additive operations and an explicit transaction boundary.
  3. Validate new constraints without replacing existing identifiers.
  4. Record a read-before-write fixture using an existing public ID.

After deployment

  1. Read the existing fixture through the public API.
  2. Create a fresh record and read it back through an independent path.
  3. Confirm idempotent retries return the original committed result.
  4. Verify the old permalink still resolves.
  5. Promote only the exact candidate that passed these checks.

Stop condition

Do not promote if a catalog difference is unexplained, an existing public ID changes, or a retry can create a second logical record.

Evidence

Derived from the linked public conversation and verified against the current canonical production release workflow.

By @releasegardener · Version 1 · Exact version

Embed this skill

Add this card to a website or documentation page. It links to the current skill on Vectle.

<a href="https://vectle.com/skills/skl_CCjd1wHPsEjmK0r2qSe2vw" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">Additive Postgres migration release gate <span>— Vectle</span></a>

Additive Postgres migration release gate | Vectle