Skip to content

getting-started: inline Ollama setup with extra_hosts note#72

Merged
smaramwbc merged 1 commit into
mainfrom
docs/quickstart-ollama-extra-hosts
May 29, 2026
Merged

getting-started: inline Ollama setup with extra_hosts note#72
smaramwbc merged 1 commit into
mainfrom
docs/quickstart-ollama-extra-hosts

Conversation

@smaramwbc

Copy link
Copy Markdown
Owner

What

Adds a concrete Ollama setup block to the "Turn on a real LLM provider" section of getting-started.md, so quickstart readers don't have to dig into the deployment guide to discover a required networking step.

Includes:

  • the Ollama .env block (ollama/llama3 model, local API base),
  • the Docker Desktop vs Docker Engine note — host.docker.internal resolves automatically on Desktop but needs extra_hosts: ["host.docker.internal:host-gateway"] on the api service under Engine,
  • the ollama pull llama3 reminder.

Why

A tester following the guide on Docker Engine hit a wall: the API container could not reach host Ollama, and the fix was only documented in deployment/guide.md. This surfaces it where the reader already is.

Notes

Companion to the statewave quickstart fix (fix/quickstart-demo-default), which makes the keyless demo-mode default actually work.

Quickstart readers no longer have to dig into the deployment guide to
learn that `host.docker.internal` needs `extra_hosts` on Linux/Docker
Engine. Adds a concrete Ollama `.env` block, the Docker Desktop vs
Engine networking note, and the `ollama pull` reminder.
@smaramwbc smaramwbc merged commit 622f685 into main May 29, 2026
2 checks passed
@smaramwbc smaramwbc deleted the docs/quickstart-ollama-extra-hosts branch May 29, 2026 19:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant