Migration Path from v1 to v2

[emicarito]

Feature Request: Migration Path from v1 to v2

Summary

I noticed there’s no clear migration path from v1 to v2 in the GPT-RAG project. This makes it challenging for me because I have existing v1 deployments and want to upgrade without starting from scratch.

Why are we doing this?

• Problem / opportunity: As a user I need a reliable and documented way to migrate from v1 to v2 without losing data, configurations, or customizations.
• Business value / outcome: A migration path lowers adoption barriers, speeds up upgrades, and ensures continuity for enterprise environments.
• Success metrics (how we know it worked):
  • Clear migration documentation is available.
  • Scripts or tools exist to automate key migration steps.
  • Positive feedback from users who successfully migrated.

What does it do? (Functional Overview)

• Core behavior: Provide a documented process and/or automation scripts to migrate:
  • Configuration files
  • Ingestion pipelines
  • Orchestrator settings
  • UI customizations
• Data collection / storage needs: Ensure compatibility between existing indexes and metadata with the v2 architecture.
• Data analysis / reporting needs: Validate data integrity and performance after migration.
• Nice to have (stretch goals):
  • A migration validation tool to check completeness.
  • A rollback option in case of migration failure.

Components

• Components (check all that apply):
  • gpt-rag-orchestrator
  • gpt-rag-ingestion
  • gpt-rag-ui
  • gpt-rag-mcp