Database CI/CD and Version Control with Oracle SQLcl and Liquibase

Treating Database Schemas as Code

In modern DevOps, application code is continuously tested, reviewed, and deployed through automated CI/CD pipelines. Yet, in many organizations, database deployments are still handled manually: a developer emails a migration.sql script to a DBA who executes it in production during late-night maintenance windows.

This manual process is error-prone, untraceable, and dangerous. By pairing Oracle SQLcl with Liquibase, teams can implement automated, version-controlled database migrations with full rollback capability.

---

Generating Changelogs with SQLcl

SQLcl includes native Liquibase commands. You can generate a complete changelog of an existing Oracle schema with a single command:

sql hr/password@localhost:1521/XEPDB1
lb generate-schema -split

This splits your schema into discrete XML/YAML/SQL files for tables, packages, views, triggers, and sequences.

---

Executing Automated Pipeline Deployments

In your GitHub Actions or GitLab CI/CD runner:

name: Deploy Database Changes
on:
  push:
    branches: [main]
jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Run Liquibase Update
        run: |
          sql -cloud ${{ secrets.DB_CONN_STR }}
          lb update -changelog-file controller.xml

---

Key Benefits

• Automatic Rollbacks: If a migration step encounters a failure, Liquibase rolls back changes, preserving database integrity.
• Audit Trail: Every applied change is logged in the DATABASECHANGELOG tracking table with execution timestamps and checksums.
• Multi-Environment Parity: Dev, Test, Staging, and Production environments always stay in perfect synchronization.