#!/bin/bash
# 空白的 Ubuntu 22.04 使用。不要在已经跑着业务的 prod102 上再执行。
set -eu
export DEBIAN_FRONTEND=noninteractive

if [ "$(hostname)" = "prod102" ] && [ -f /root/.kube/config ]; then
  echo "prod102 上已经有集群。这份脚本是给空机器的，停在这里。"
  exit 1
fi

apt-get update
apt-get install -y ca-certificates curl gnupg python3

apt-get install -y nvidia-driver-550
if ! nvidia-smi >/dev/null 2>&1; then
  echo "驱动已安装。重启后再继续，直到 nvidia-smi 能看到 NVIDIA H20。"
  exit 0
fi

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey \
  | gpg --dearmor --yes -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg
curl -fsSL https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \
  | sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' \
  > /etc/apt/sources.list.d/nvidia-container-toolkit.list
apt-get update
apt-get install -y nvidia-container-toolkit=1.17.8-1

modprobe br_netfilter || true
cat > /etc/modules-load.d/k8s.conf << 'EOF'
br_netfilter
EOF
cat > /etc/sysctl.d/k8s.conf << 'EOF'
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-ip6tables = 1
net.ipv4.ip_forward = 1
EOF
sysctl --system >/dev/null

mkdir -p /usr/cluster/models /usr/cluster/docker /opt/new-api/data /opt/new-api/logs \
  /usr/cluster/dify/xinference /usr/cluster/server-status /usr/cluster/llm-nav \
  /usr/cluster/observability/metrics
ln -sfn /usr/cluster/models /models
echo PREP_OK