# Database Change Automation

> Automate database change end to end: a GitOps change pipeline, automated SQL review, risk-tiered approvals, rollback, and a complete audit trail. Self-hosted, open-source.

Source: https://www.bytebase.com/automate-database-change/

---

## What manual database change costs you

### Schema changes ship outside CI/CD

App code rides pull requests and pipelines; schema changes get pasted into a client by hand, breaking the audit chain.

### Risky SQL reaches production unchecked

An unreviewed migration with a missing index or locking DDL can take a table down. Without automated SQL review, the catch happens in prod.

### No record of who changed what

Run a change ad hoc and there's no actor, no diff, no timestamp. A rollback means reconstructing history from memory.

## A governed database change pipeline, not a pile of scripts

### GitOps change flow

Migrations live alongside application code and merge through your existing VCS pipeline — a schema change is just another reviewed pull request.

### Batched multi-environment rollout

Promote a change from dev to staging to production in one controlled sequence, each environment gated by its own policy.

### One workflow across every engine

PostgreSQL, MySQL, SQL Server, Oracle, and 20+ engines deploy through the same pipeline.

## Automated SQL review on every change

### 100+ built-in review rules

Naming conventions, missing indexes, locking DDL, and backward-incompatible changes are flagged before a statement touches the database.

### Risk-tiered approvals

DDL and bulk DML route to a human approver automatically; safe, bounded changes stay on the fast path.

### Rollback built in

Every change carries its diff and a generated rollback plan, so reverting is a click.

## Full history and audit trail on every database change

### Complete change history

Every migration is logged with who proposed it, who approved it, the exact SQL, and when it ran.

### Audit-ready exports

Hand auditors a SOC 2- or ISO-ready record straight from Bytebase — no spreadsheet reconstruction.

### Always know the current state

A live schema catalog across every database shows exactly what's deployed, so drift surfaces immediately.

## Frequently asked questions

### What does it mean to automate database change?

Moving schema and data changes off manual SQL clients onto a governed pipeline — proposed, reviewed, approved, deployed, and audited automatically: version-controlled migrations, automated SQL review, policy-based approvals, rollback, and a complete change history, the way application code ships through CI/CD.

### How is it different from running Liquibase or Flyway in CI/CD?

Liquibase and Flyway automate the deployment step. Bytebase automates the whole workflow around it: SQL review against 100+ rules, risk-tiered approvals, rollback, ad-hoc access control, and an audit trail, across 20+ engines, in one place.

### Does automated SQL review slow developers down?

No. Review runs on every change and flags only genuine risks — missing indexes, locking DDL, backward-incompatible changes. Safe changes pass on the fast path; only risky statements route to a human approver.

### Can I automate change across PostgreSQL, MySQL, SQL Server, and Oracle together?

Yes. One change pipeline runs across PostgreSQL, MySQL, SQL Server, Oracle, and 20+ engines, with the same review policies, approval flows, and audit trail everywhere.

### How do rollbacks work?

Every change carries its diff and a generated rollback plan, so reverting a migration is a click rather than a manual reconstruction.

### Is Bytebase self-hosted and open-source?

Yes. Bytebase runs in your own infrastructure as a single Docker image, so data and credentials never leave your network, and it's open-source.

## Get Started

- [Contact us](https://www.bytebase.com/contact-us/)
- [Start now (cloud)](https://console.bytebase.com)