cediackermann
ef27add5d3
- Extract shared helpers into steps/common.sh (colors, print_*, wait_for_user, ask_confirmation) - Split each install step into its own sourced script under steps/ - setup.sh now sources step files instead of duplicating logic - Add setup_ollama.sh: Ollama-only device setup (no Docker) - Add setup_onyx.sh: Onyx-only device setup with auto-detection of Ollama host via network scan - setup_ollama.sh prints ready-to-run setup_onyx.sh command with detected IP at completion - Docker memory recommendation is MAX when machine has <20 GB RAM, 20 GB otherwise - cleanup.sh and check-status.sh source steps/common.sh with inline curl fallback - check-status.sh detects device mode (single/ollama/onyx) and skips irrelevant checks - cleanup.sh reinstall message lists all three setup one-liners - README updated with split-device one-liner commands
104 lines
2.9 KiB
Markdown
104 lines
2.9 KiB
Markdown
# RIoT AI Setup Guide
|
|
|
|
> Step-by-step instructions for setting up a Mac Mini from scratch.
|
|
|
|
## Presentation Link
|
|
|
|
[Link](https://docs.google.com/presentation/d/12XRHySp_W6Mrh64nA_6Hi3s8XvliUoANP-p_ha0FHEA/edit?usp=sharing)
|
|
|
|
---
|
|
|
|
## Quick Start
|
|
|
|
### Single device (Ollama + Onyx on the same machine)
|
|
|
|
```bash
|
|
bash <(curl -fsSL https://raw.githubusercontent.com/cediackermann/riotsecure/main/setup.sh)
|
|
```
|
|
|
|
### Two devices (Ollama on one machine, Onyx on another)
|
|
|
|
**On the Ollama device** — installs Ollama and the RIoT models, no Docker needed:
|
|
|
|
```bash
|
|
bash <(curl -fsSL https://raw.githubusercontent.com/cediackermann/riotsecure/main/setup_ollama.sh)
|
|
```
|
|
|
|
At the end the script prints the exact command to run on the Onyx device.
|
|
|
|
**On the Onyx device** — installs Docker and Onyx, auto-detects the Ollama device on the network:
|
|
|
|
```bash
|
|
bash <(curl -fsSL https://raw.githubusercontent.com/cediackermann/riotsecure/main/setup_onyx.sh)
|
|
```
|
|
|
|
The script scans the local network for Ollama. If it can't find it automatically, it will ask for the IP address.
|
|
|
|
---
|
|
|
|
## Status & Maintenance
|
|
|
|
**Check everything is working:**
|
|
|
|
```bash
|
|
cd ~/riotsecure && ./check-status.sh
|
|
```
|
|
|
|
**Update Ollama models:**
|
|
|
|
```bash
|
|
cd ~/riotsecure && ./updateModels.sh modelfiles
|
|
```
|
|
|
|
**Uninstall:**
|
|
|
|
```bash
|
|
bash <(curl -fsSL https://raw.githubusercontent.com/cediackermann/riotsecure/main/cleanup.sh)
|
|
```
|
|
|
|
---
|
|
|
|
## Manual Steps Reference
|
|
|
|
The scripts pause and prompt when manual action is required. Here is what to expect at each pause.
|
|
|
|
### Docker Desktop resources
|
|
|
|
Go to **Docker → Settings → Resources** and set:
|
|
|
|
| Setting | Value |
|
|
| ---------------- | ---------------------------- |
|
|
| CPU limit | MAX |
|
|
| Memory limit | 20 GB (or MAX if less than 20 GB available) |
|
|
| Disk usage limit | MAX |
|
|
|
|
Hit **Apply & Restart**.
|
|
|
|
### Onyx installer prompts
|
|
|
|
1. Press **Enter** to acknowledge
|
|
2. Choose **2** for Standard
|
|
3. Press **Enter** for Edge
|
|
|
|
### Web interface configuration
|
|
|
|
1. Open `http://localhost:3000` and create an admin account.
|
|
2. Go to **Admin Panel → Language Models**:
|
|
- Select **Ollama** as provider and give it a name
|
|
- Set the API base URL to the address shown by the setup script
|
|
- Refresh the model list, select at least the three RIoT models, and click **Connect**
|
|
3. Go to **Admin Panel → Chat Preferences → System Prompt → Modify Prompt**, delete the entire prompt, and save.
|
|
|
|
### RAG content upload
|
|
|
|
**Option A — File upload (recommended)**
|
|
|
|
```bash
|
|
cd ~/riotsecure && ./fetchContent.sh riot-sources.txt
|
|
```
|
|
|
|
Then in the web interface: **Admin Panel → Add Connector → File**, upload the files from `~/riotsecure/content`, and wait ~30 seconds for indexing.
|
|
|
|
**Option B — URL connector**
|
|
|
|
In the web interface: **Admin Panel → Add Connector → Web**, add each URL from `riot-sources.txt` as a separate connector using scrape method **Single**, and wait ~30 seconds for indexing.
|