OpenStack Lightspeed Operator documentation

OpenStack Lightspeed is an AI-powered assistant, tailored for Red Hat OpenStack Services on OpenShift (RHOSO), that answers questions in plain English using product documentation. Goose CLI is the primary supported interface for RHOSO 19 Beta.

Ask it something like “How do I create a VM using the OpenStack CLI?” or “Why would a Nova compute service show as down?” — see Usage for more.

You don’t need an existing RHOSO deployment to try it — an OpenShift cluster and an LLM you can point it at is enough (see Quickstart).

Important

This is a community release. Support is provided upstream only, via GitHub Issues — there is no separate commercial support channel for this project:

• lightspeed-operator issues

• lightspeed-rag-content issues

See Troubleshooting for what to check, and what to include, before filing an issue.

Contents:

• Quickstart
  • Install the operator
  • Create the secret and CR
  • Use the assistant
• Installation Guide
  • Prerequisites
  • Access to registry images
  • Installing the operator
  • Setting up LLM credentials
  • Deploying OpenStackLightspeed
  • Verifying the deployment
  • Accessing the assistant
  • Accessing the API
• Configuration
  • Core fields
  • Models (models[])
  • Supported providers
  • Logging
  • Data collection
  • Persistent storage (database)
  • Container resources
  • Container images
  • Offline knowledge portal
  • Quota enforcement
  • Developer / experimental options (dev)
• Development
  • Local cluster with CRC
  • Releasing to the community catalog
  • Architecture
  • RHOKP container image
  • Pod security defaults
• Troubleshooting
  • Start with the resource’s conditions
  • Deployment-specific issues
  • Getting operator logs
• Available Features
  • Asking questions
  • Multi-model request routing
  • Quota enforcement (optional)
  • Feedback and transcripts
• Testing OpenStackAssistant with OpenStack Lightspeed
  • 1. Deploy the OpenStack Lightspeed operator
  • 2. Create the LLM credentials and OpenStackLightspeed instance
  • 3. Wait for OpenStack Lightspeed to become ready
  • 4. Deploy OpenStackAssistant
  • 5. Start Goose in the assistant pod