Skip to content

Latest commit

 

History

History
80 lines (55 loc) · 2.43 KB

File metadata and controls

80 lines (55 loc) · 2.43 KB

First Boot

Boot Process (6-Stage Model)

The boot follows the VOLT 6-stage sequence defined in VOLT Architecture Chapter 3:

  1. GRUB — Detects your CPU architecture (x86_64 or aarch64) and loads the matching Linux kernel
  2. Linux Kernel — Boots Debian Trixie base with live filesystem
  3. Plymouth Splash — Samaris-branded boot animation
  4. Systemd Services — Volt daemons initialize in order:
    • ASC (Adaptive System Config) — detects hardware, generates /run/samaris/adaptive.generated.toml
    • Kernel B — backend kernel starts on ws://localhost:9998
    • VRM / VUM / VGM / DWP — RAM manager, update service, GPU monitor, worker pool
    • Kernel A — frontend kernel starts on ws://localhost:9999
    • Unifier — Volt Unifier bridges Kernel A and Kernel B
    • Desktop — Electron shell loads with React Samaris UI
  5. Onboarding — First-boot welcome flow (theme, WiFi, tour)
  6. Desktop Ready — Fully interactive desktop

Boot Timeline

BIOS → GRUB → Kernel (10.7s) → Userspace (37.8s) → Desktop Ready

Total: ~48 seconds in QEMU VM (4 GB RAM, 4 vCPU). Times may vary on bare metal.


What Happens at First Boot

  1. Adaptive Config — ASC detects hardware and generates /run/samaris/adaptive.generated.toml
  2. AI Models — Qwen3, Whisper, and OuteTTS models load into VRM-managed memory from /opt/volt/ai-models/
  3. Desktop — The onboarding flow welcomes you with language selection, theme selection, and WiFi setup
  4. Orbit AI — The local AI assistant is ready after the first ~8s Metal shader compilation

Verifying the System

# Check kernel health
curl ws://localhost:9999

# Check service status
systemctl status volt-kernel.service

# Verify daemons
ls -la /run/samaris/

# View boot logs
journalctl -b | grep volt

Troubleshooting

If the desktop doesn't appear after boot:

  • Wait up to 60 seconds for first-time shader compilation (Metal/Vulkan)
  • Check journalctl -u volt-desktop.service for errors
  • Verify display hardware in lspci | grep VGA
  • Ensure VRM has sufficient memory allocated (at least 2048 MB for LLM)

Related



← Back: Documentation Index