# 在 VPS 上部署 Node Exporter
docker run -d \
  --name node-exporter \
  --network host \
  --pid host \
  -v /proc:/host/proc:ro \
  -v /sys:/host/sys:ro \
  -v /:/rootfs:ro \
  --mount type=bind,source=/etc/hostname,target=/etc/hostname \
  prom/node-exporter

# 通过 Prometheus 抓取指标
# prometheus.yml
scrape_configs:
  - job_name: 'vps-metrics'
    static_configs:
      - targets: ['localhost:9100']

#!/bin/bash
# cost-tracker.sh - 每日成本记录
VPS_NAME=$1
DAILY_COST=$2
DATE=$(date +%Y-%m-%d)

echo "$DATE,$VPS_NAME,$DAILY_COST" >> /opt/cost-tracking/history.csv

# 查看历史 CPU 利用率（最后 30 天）
promtool query instant prometheus:9090 \
  'avg by (instance) (rate(node_cpu_seconds_total{mode="idle"}[30d])) * 100'

# 检查内存使用情况
free -h
# 注意：buff/cache 不是真实占用

# 找出内存占用最多的进程
ps aux --sort=-%mem | head -10

# 检查是否有内存泄漏
vmstat 1 5

#!/bin/bash
# storage-tiering.sh - 自动归档冷数据

# 查找 90 天未访问的文件
find /data -type f -mtime +90 -exec mv {} /mnt/archive/ \;

# 定期压缩归档
tar czf /mnt/archive/backup-$(date +%Y%m%d).tar.gz /mnt/archive/old-files/

# 上传到廉价对象存储（可选）
rclone copy /mnt/archive/ backup:my-bucket/cold-storage/

# Kubernetes HPA 配置示例
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: web-service
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: web-service
  minReplicas: 1
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 70

# 使用 DigitalOcean 快照迁移
doctl compute snapshot create my-vps-snapshot --instance-id 123456
doctl compute droplet create my-vps-small \
  --image my-vps-snapshot \
  --size s-1vcpu-1gb \
  --region nyc1

#!/bin/bash
# auto-shutdown.sh

# 工作时间：9:00-18:00
START_HOUR=9
END_HOUR=18

CURRENT_HOUR=$(date +%H)

if [ "$CURRENT_HOUR" -lt "$START_HOUR" ] || [ "$CURRENT_HOUR" -ge "$END_HOUR" ]; then
    # 非工作时间，关停开发服务器
    doctl compute droplet action stop 123456
fi

# AWS EC2 竞价实例启动
aws ec2 run-instances \
  --instance-market-options 'SpotOptions={InstanceInterruptionBehavior=terminate}' \
  --instance-type m5.large \
  --min-count 1 \
  --max-count 1

# AWS Reserved Instance 购买示例
aws ec2 purchase-reserved-instances-offering \
  --reserved-instances-offering-id <offer-id> \
  --instance-count 1

# 启用 ZFS 压缩
zfs set compression=zstd rpool/data

# 定期清理无用包
apt autoremove --purge -y
apt clean

# 清理 Docker 资源
docker system prune -a --force
docker volume prune --force

# docker-compose.yml - 限制资源使用
services:
  web:
    image: nginx:latest
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 256M
        reservations:
          cpus: '0.25'
          memory: 128M

# Alertmanager 规则
groups:
  - name: cost-optimization
    rules:
      - alert: HighCPUUtilization
        expr: node_cpu_seconds_total{mode="idle"} < 0.3
        for: 10m
        labels:
          severity: warning
        annotations:
          summary: "CPU 利用率超过 70%，考虑扩容"
      
      - alert: StorageNearCapacity
        expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.2
        for: 5m
        labels:
          severity: critical
        annotations:
          summary: "磁盘空间不足 20%，需要清理或扩容"