data-migration

Plan zero-downtime MySQL and PostgreSQL migrations using Expand-and-Contract, Dual-Write, and Shadow Table patterns.

1|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/iceflower/agent-skills --skill data-migration-iceflower
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: data-migration
Source: https://github.com/iceflower/agent-skills/tree/main/data-migration
Command: npx skills add https://github.com/iceflower/agent-skills --skill data-migration-iceflower

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Data migration and ETL patterns including zero-downtime migration strategies, backfills, validation, and rollback planning to keep systems in sync during complex migrations.

Core Features & Use Cases

  • Expand-and-Contract, Dual-Write, and Shadow Table patterns for safe schema evolution
  • Online schema migration tooling (gh-ost, pt-online-schema-change, pgroll) plus backward-compatible changes
  • Data validation, reconciliation, CDC, and rollback strategies across MySQL and PostgreSQL

Quick Start

Start by selecting a zero-downtime migration pattern and outline the phased rollout steps.

Frequently Asked Questions about data-migration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute zero-downtime data migrations across MySQL and PostgreSQL?

Zero-downtime data migrations use Expand-and-Contract, Dual-Write, and Shadow Table patterns to maintain system sync. They enforce backward-compatible changes, data validation, and rollback procedures across MySQL and PostgreSQL.

When do I need CDC and shadow tables for ETL migrations?

CDC and shadow tables are needed for ETL migrations requiring continuous data synchronization and validation without downtime. They capture incremental changes during complex migrations to keep source and target databases aligned.

What is the best way to plan a phased rollout for online schema migrations?

The best way to plan phased rollouts for online schema migrations is applying backward-compatible changes using gh-ost, pt-online-schema-change, or pgroll, ensuring data validation and rollback readiness at each stage.

Can I use gh-ost and pt-online-schema-change for backward-compatible schema evolution?

Yes, gh-ost and pt-online-schema-change support backward-compatible schema evolution by applying online schema migrations. They enable zero-downtime table alterations while maintaining data integrity and system availability.

Why does dual-write require data reconciliation during database migrations?

Dual-write requires data reconciliation during database migrations to detect discrepancies between simultaneously updated sources. Validation ensures both systems remain synchronized before completing the phased rollout and contracting the old schema.