fix: remove Ollama and nvidia-ctk runtime configure from DGX installer
- Remove Ollama installation functionality (not needed on DGX) - Remove nvidia-ctk runtime configure command that breaks DGX OS 7 (DGX uses override.conf, not daemon.json) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
@@ -24,7 +24,6 @@ RELEASE_URL="https://api.github.com/repos/GT-Edge-AI-Internal/gt-ai-os-community
|
||||
UNATTENDED=false
|
||||
SKIP_DOCKER_CHECK=false
|
||||
USE_RELEASE=false
|
||||
INSTALL_OLLAMA=false
|
||||
BRANCH="main"
|
||||
RUNNING_FROM_REPO=false
|
||||
|
||||
@@ -68,7 +67,6 @@ OPTIONS:
|
||||
-b, --branch BRANCH Git branch to clone (default: main)
|
||||
--skip-docker-check Skip Docker installation check
|
||||
--use-release Download latest release instead of cloning git repo
|
||||
--install-ollama Install and configure Ollama for local LLM inference
|
||||
|
||||
REQUIREMENTS:
|
||||
- NVIDIA DGX OS or Ubuntu with NVIDIA drivers
|
||||
@@ -82,8 +80,8 @@ EXAMPLES:
|
||||
# Interactive installation
|
||||
sudo $0
|
||||
|
||||
# Unattended installation with Ollama
|
||||
sudo $0 --unattended --install-ollama
|
||||
# Unattended installation
|
||||
sudo $0 --unattended
|
||||
|
||||
# Custom directory with pre-built release
|
||||
sudo $0 --dir /data/gt2 --use-release
|
||||
@@ -122,10 +120,6 @@ parse_args() {
|
||||
USE_RELEASE=true
|
||||
shift
|
||||
;;
|
||||
--install-ollama)
|
||||
INSTALL_OLLAMA=true
|
||||
shift
|
||||
;;
|
||||
*)
|
||||
print_error "Unknown option: $1"
|
||||
show_help
|
||||
@@ -298,71 +292,6 @@ check_disk_space() {
|
||||
print_success "Available disk space: ${available_gb}GB"
|
||||
}
|
||||
|
||||
check_ollama() {
|
||||
if [ "$INSTALL_OLLAMA" = false ] && [ "$UNATTENDED" = true ]; then
|
||||
return
|
||||
fi
|
||||
|
||||
print_info "Checking for Ollama..."
|
||||
|
||||
if command -v ollama &> /dev/null; then
|
||||
print_success "Ollama found: $(ollama --version 2>&1 | head -1)"
|
||||
return
|
||||
fi
|
||||
|
||||
print_info "Ollama not found"
|
||||
|
||||
if [ "$INSTALL_OLLAMA" = false ] && [ "$UNATTENDED" = false ]; then
|
||||
read -p "Install Ollama for local LLM inference? (y/n): " install_ollama_choice
|
||||
if [ "$install_ollama_choice" = "y" ]; then
|
||||
INSTALL_OLLAMA=true
|
||||
fi
|
||||
fi
|
||||
|
||||
if [ "$INSTALL_OLLAMA" = true ]; then
|
||||
install_ollama_service
|
||||
fi
|
||||
}
|
||||
|
||||
install_ollama_service() {
|
||||
print_info "Installing Ollama..."
|
||||
|
||||
# Download and install Ollama
|
||||
curl -fsSL https://ollama.ai/install.sh | sh
|
||||
|
||||
# Configure Ollama to listen on all interfaces
|
||||
mkdir -p /etc/systemd/system/ollama.service.d
|
||||
|
||||
cat > /etc/systemd/system/ollama.service.d/override.conf << 'EOF'
|
||||
[Service]
|
||||
Environment="OLLAMA_HOST=0.0.0.0:11434"
|
||||
Environment="OLLAMA_ORIGINS=*"
|
||||
Environment="CUDA_VISIBLE_DEVICES=0"
|
||||
Environment="OLLAMA_NUM_PARALLEL=4"
|
||||
Environment="OLLAMA_MAX_LOADED_MODELS=1"
|
||||
EOF
|
||||
|
||||
# Reload and start Ollama
|
||||
systemctl daemon-reload
|
||||
systemctl enable ollama
|
||||
systemctl start ollama
|
||||
|
||||
# Wait for Ollama to be ready
|
||||
sleep 5
|
||||
|
||||
print_success "Ollama installed and running"
|
||||
|
||||
# Pull a default model
|
||||
if [ "$UNATTENDED" = false ]; then
|
||||
read -p "Pull llama3.1:8b model now? (y/n): " pull_model
|
||||
if [ "$pull_model" = "y" ]; then
|
||||
print_info "Pulling llama3.1:8b (this may take several minutes)..."
|
||||
ollama pull llama3.1:8b
|
||||
print_success "Model downloaded"
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
cleanup_existing_containers() {
|
||||
# Find all containers with gentwo- prefix (running or stopped)
|
||||
local existing_containers=$(docker ps -a --filter "name=gentwo-" --format "{{.Names}}" 2>/dev/null || true)
|
||||
@@ -654,11 +583,6 @@ display_access_info() {
|
||||
echo -e "${BLUE}Access URLs:${NC}"
|
||||
echo -e " Control Panel: ${GREEN}http://localhost:3001${NC}"
|
||||
echo -e " Tenant App: ${GREEN}http://localhost:3002${NC}"
|
||||
|
||||
if [ "$INSTALL_OLLAMA" = true ]; then
|
||||
echo -e " Ollama API: ${GREEN}http://localhost:11434${NC}"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
echo -e "${BLUE}Default Credentials:${NC}"
|
||||
echo -e " Username: ${GREEN}gtadmin@test.com${NC}"
|
||||
@@ -670,16 +594,6 @@ display_access_info() {
|
||||
echo " - Platform: NVIDIA DGX with Grace ARM architecture"
|
||||
echo " - Embeddings: Optimized for 20-core ARM CPU"
|
||||
echo " - Memory: High-memory configuration enabled"
|
||||
|
||||
if [ "$INSTALL_OLLAMA" = true ]; then
|
||||
echo " - Ollama: Running on port 11434 for local LLM inference"
|
||||
echo ""
|
||||
echo -e "${BLUE}Ollama Commands:${NC}"
|
||||
echo " Test Ollama: ollama list"
|
||||
echo " Ollama status: systemctl status ollama"
|
||||
echo " Pull models: ollama pull <model-name>"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
echo -e "${BLUE}Useful Commands:${NC}"
|
||||
echo " View logs: cd ${INSTALL_DIR} && docker compose logs -f"
|
||||
@@ -848,7 +762,6 @@ main() {
|
||||
check_nvidia_container_runtime
|
||||
check_ram
|
||||
check_disk_space
|
||||
check_ollama
|
||||
|
||||
# Check for existing installation FIRST (prompts user before any cleanup)
|
||||
# Skip if we're running from within the repo we'd be checking
|
||||
|
||||
Reference in New Issue
Block a user