Brohammad / Brohammad/VoxForge
Write a horizontal scaling guide (multi-worker + sticky sessions)
- Dominant language
- Python
- Stars
- 4
- Forks
- 1
- Avg merge
- 5d 13m
- Merged PRs (30d)
- 10
Description
## Summary
Production compose uses one uvicorn worker. `docs/deployment/operations.md` has three bullets. Roadmap v1.1 wants a real scaling guide.
## Context
- Known limitation: single uvicorn worker
- WebSocket voice needs sticky sessions if you add replicas
- Postgres/Redis should move to managed services before app replicas
## Scope
New doc e.g. `docs/deployment/scaling.md` covering:
- Why `--workers 1` today
- Sticky sessions / WebSocket routing
- Session state in Redis vs in-process
- Knowledge worker and LiveKit worker as separate services
- What breaks if you naively set workers > 1
Link it from operations.md, deployment README, and known-limitations.
## Out of scope
Implementing multi-instance production in this repo’s default compose.
Contributor guide
Research direction
Start with docs/deployment/operations.md, the deployment README, and the known-limitations documentation to understand the current single-worker setup and existing links. Create the scaling guide covering the listed worker, session, routing, and service-separation topics, then link it from all three locations. Done means the documentation explains the multi-worker risks and scaling prerequisites without changing the default compose deployment.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker, fastapi, postgres, python, redis
- Domain
- backend, devops, documentation, infrastructure
- Issue type
- Documentation
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Clearly specified
- Newbie friendliness
- 75/100