#!/bin/bash # cf-vps-monitor - Cloudflare Worker VPS监控脚本 # 版本: 1.1.0 # 支持所有常见Linux系统，无需root权限 set -euo pipefail # 初始化系统类型变量 OS=$(uname -s) export OS # 颜色定义 RED='\033[0;31m' GREEN='\033[0;32m' YELLOW='\033[1;33m' BLUE='\033[0;34m' PURPLE='\033[0;35m' CYAN='\033[0;36m' NC='\033[0m' # No Color # 全局变量 - 集中式文件管理 SCRIPT_DIR="$HOME/.cf-vps-monitor" CONFIG_FILE="$SCRIPT_DIR/config/config" LOG_FILE="$SCRIPT_DIR/logs/monitor.log" PID_FILE="$SCRIPT_DIR/run/monitor.pid" SERVICE_FILE="$SCRIPT_DIR/bin/vps-monitor-service.sh" INSTALL_MANIFEST="$SCRIPT_DIR/system/install.manifest" # 默认配置 DEFAULT_INTERVAL=10 DEFAULT_WORKER_URL="" DEFAULT_SERVER_ID="" DEFAULT_API_KEY="" # 打印带颜色的消息 print_message() { local color=$1 local message=$2 echo -e "${color}${message}${NC}" } # 日志函数（环境适配） log() { local message="$1" local timestamp=$(date '+%Y-%m-%d %H:%M:%S') echo "[$timestamp] $message" >> "$LOG_FILE" # 只在非服务模式下输出到控制台（避免重复日志） if [[ "${SERVICE_MODE:-false}" != "true" ]]; then echo "[$timestamp] $message" fi } # 错误处理 error_exit() { local message="$1" print_message "$RED" "错误: $message" log "ERROR: $message" exit 1 } # 检查命令是否存在 command_exists() { command -v "$1" >/dev/null 2>&1 } # ==================== 系统兼容性层 ==================== # 检测systemd可用性 is_systemd_available() { command_exists systemctl && systemctl --version >/dev/null 2>&1 } # 检测用户级systemd可用性 is_user_systemd_available() { if is_root_user; then is_systemd_available else is_systemd_available && \ [[ -n "${XDG_RUNTIME_DIR:-}" ]] && \ systemctl --user --version >/dev/null 2>&1 fi } # 跨平台sed命令 safe_sed() { local pattern="$1" local file="$2" if [[ "$OS" == "FreeBSD" ]] || [[ "$OS" == "Darwin" ]]; then sed -i '' "$pattern" "$file" 2>/dev/null || true else sed -i "$pattern" "$file" 2>/dev/null || true fi } # 安全的systemctl命令 safe_systemctl() { if is_systemd_available; then systemctl "$@" 2>/dev/null || true else return 1 fi } # 检查系统资源（防止fork错误） check_system_resources() { # 检查进程数限制（特别针对FreeBSD） local max_proc=$(ulimit -u 2>/dev/null || echo "1024") local current_proc=$(ps aux 2>/dev/null | wc -l || echo "100") if [[ $current_proc -gt $((max_proc * 80 / 100)) ]]; then print_message "$YELLOW" "警告: 进程数接近限制 ($current_proc/$max_proc)" if [[ "$OS" == "FreeBSD" ]]; then print_message "$CYAN" "FreeBSD建议: 增加用户进程限制或稍后重试" fi return 1 fi return 0 } # 验证PID有效性 validate_pid() { local pid="$1" [[ "$pid" =~ ^[0-9]+$ ]] && [[ "$pid" != "$$" ]] && kill -0 "$pid" 2>/dev/null } # 获取进程命令行（FreeBSD兼容） get_process_command() { local pid="$1" if [[ "$OS" == "FreeBSD" ]]; then # FreeBSD兼容语法 ps -p "$pid" -o command 2>/dev/null | tail -n +2 | head -1 || echo "unknown" else # Linux标准语法 ps -p "$pid" -o cmd= 2>/dev/null || echo "unknown" fi } # 统一的监控进程检测函数（精确检测） find_monitor_processes() { local pids="" # 层次1: PID文件检测（最可靠） if [[ -f "$PID_FILE" ]]; then local file_pid=$(cat "$PID_FILE" 2>/dev/null) if validate_pid "$file_pid"; then pids="$file_pid" fi fi # 层次2: 精确脚本路径匹配 if [[ -z "$pids" ]] && [[ -f "${SERVICE_FILE:-}" ]]; then if [[ "$OS" == "FreeBSD" ]]; then pids=$(ps axww | grep "$SERVICE_FILE" | grep -v grep | awk '{print $1}') else pids=$(ps aux | grep "$SERVICE_FILE" | grep -v grep | awk '{print $2}') fi fi # 层次3: 验证所有PID并确认命令行 local valid_pids="" for pid in $pids; do if validate_pid "$pid"; then local cmd=$(get_process_command "$pid") # 确认命令行确实包含我们的脚本 if [[ "$cmd" =~ (vps-monitor-service|cf-vps-monitor) ]]; then valid_pids="$valid_pids $pid" fi fi done echo "$valid_pids" | sed 's/^[[:space:]]*//' | sed 's/[[:space:]]*$//' } # 检查监控服务是否运行 is_monitor_running() { local pids=$(find_monitor_processes) [[ -n "$pids" ]] } # 获取用户类型描述 get_user_type_description() { if is_root_user; then echo "系统管理员" else echo "普通用户" fi } # 简洁的监控服务诊断 diagnose_monitor_service() { print_message "$CYAN" "=== 监控服务诊断 ===" # 检查关键文件 print_message "$BLUE" "文件状态:" [[ -f "$SERVICE_FILE" ]] && print_message "$GREEN" " ✓ 服务脚本存在" || print_message "$RED" " ✗ 服务脚本不存在" [[ -f "$CONFIG_FILE" ]] && print_message "$GREEN" " ✓ 配置文件存在" || print_message "$RED" " ✗ 配置文件不存在" [[ -d "$(dirname "$LOG_FILE")" ]] && print_message "$GREEN" " ✓ 日志目录存在" || print_message "$RED" " ✗ 日志目录不存在" # 显示相关进程（FreeBSD优化） print_message "$BLUE" "相关进程:" if [[ "$OS" == "FreeBSD" ]]; then local processes=$(ps axww | grep -E "(monitor|vps)" | grep -v grep | grep -v diagnose) else local processes=$(ps aux | grep -E "(monitor|vps)" | grep -v grep | grep -v diagnose) fi if [[ -n "$processes" ]]; then echo "$processes" else print_message "$YELLOW" " 无相关进程" fi print_message "$CYAN" "====================" } # 移除所有自启动设置 remove_autostart_settings() { print_message "$BLUE" "移除自启动设置..." local removed_count=0 # 1. 移除systemd服务 local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi if [[ -f "$service_path" ]] && is_systemd_available; then local systemd_cmd="systemctl" [[ ! $(is_root_user) ]] && systemd_cmd="systemctl --user" $systemd_cmd stop cf-vps-monitor.service 2>/dev/null || true $systemd_cmd disable cf-vps-monitor.service 2>/dev/null || true rm -f "$service_path" $systemd_cmd daemon-reload 2>/dev/null || true print_message "$GREEN" " ✓ systemd服务已移除" removed_count=$((removed_count + 1)) fi # 2. 移除crontab条目 if command_exists crontab; then local current_crontab=$(crontab -l 2>/dev/null || echo "") if echo "$current_crontab" | grep -q "cf-vps-monitor"; then echo "$current_crontab" | grep -v "cf-vps-monitor" | crontab - 2>/dev/null print_message "$GREEN" " ✓ crontab自启动已移除" removed_count=$((removed_count + 1)) fi fi # 3. 移除shell profile自启动（FreeBSD兼容） local profile_files=(".bashrc" ".bash_profile" ".profile") for profile in "${profile_files[@]}"; do local profile_path="$HOME/$profile" if [[ -f "$profile_path" ]] && grep -q "cf-vps-monitor auto-start" "$profile_path" 2>/dev/null; then # FreeBSD兼容的sed语法 if [[ "$OS" == "FreeBSD" ]] || [[ "$OS" == "Darwin" ]]; then sed -i '' '/# === cf-vps-monitor auto-start BEGIN ===/,/# === cf-vps-monitor auto-start END ===/d' "$profile_path" 2>/dev/null else sed -i '/# === cf-vps-monitor auto-start BEGIN ===/,/# === cf-vps-monitor auto-start END ===/d' "$profile_path" 2>/dev/null fi print_message "$GREEN" " ✓ 已从 $profile 移除自启动代码" removed_count=$((removed_count + 1)) break fi done # 显示结果 if [[ $removed_count -gt 0 ]]; then print_message "$GREEN" "✓ 已移除 $removed_count 种自启动设置" else print_message "$YELLOW" "未找到需要移除的自启动设置" fi } # 添加自启动设置 add_autostart_settings() { print_message "$BLUE" "配置自启动设置..." local added_count=0 # 1. 尝试配置systemd服务 local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi if is_systemd_available && [[ ! -f "$service_path" ]]; then # 创建服务目录 mkdir -p "$(dirname "$service_path")" 2>/dev/null # 创建systemd服务文件 cat > "$service_path" << EOF [Unit] Description=CF VPS Monitor Service After=network.target [Service] Type=simple ExecStart=$SERVICE_FILE Restart=always RestartSec=10 User=$USER WorkingDirectory=$HOME [Install] WantedBy=default.target EOF local systemd_cmd="systemctl" [[ ! $(is_root_user) ]] && systemd_cmd="systemctl --user" $systemd_cmd daemon-reload 2>/dev/null || true $systemd_cmd enable cf-vps-monitor.service 2>/dev/null || true print_message "$GREEN" " ✓ systemd服务已配置" added_count=$((added_count + 1)) fi # 2. 配置crontab自启动 if command_exists crontab; then local current_crontab=$(crontab -l 2>/dev/null || echo "") if ! echo "$current_crontab" | grep -q "cf-vps-monitor"; then local crontab_entry="@reboot sleep 30 && pgrep -f 'cf-vps-monitor|vps-monitor-service' >/dev/null || $SERVICE_FILE" (echo "$current_crontab"; echo "$crontab_entry") | crontab - 2>/dev/null print_message "$GREEN" " ✓ crontab自启动已配置" added_count=$((added_count + 1)) fi fi # 3. 配置shell profile自启动 local profile="$HOME/.bashrc" if [[ -f "$profile" ]] && ! grep -q "cf-vps-monitor auto-start" "$profile" 2>/dev/null; then cat >> "$profile" << EOF # === cf-vps-monitor auto-start BEGIN === # VPS监控服务自启动检测 (最后保障) if [ -n "\$PS1" ] && [ "\$TERM" != "dumb" ]; then if ! pgrep -f 'cf-vps-monitor|vps-monitor-service' >/dev/null 2>&1; then (sleep 5 && nohup "$SERVICE_FILE" >/dev/null 2>&1 &) & fi fi # === cf-vps-monitor auto-start END === EOF print_message "$GREEN" " ✓ shell profile自启动已配置" added_count=$((added_count + 1)) fi # 显示结果 if [[ $added_count -gt 0 ]]; then print_message "$GREEN" "✓ 已配置 $added_count 种自启动设置" else print_message "$YELLOW" "自启动设置已存在，无需重复配置" fi } # 统一的命令接口 get_system_command() { local cmd_type="$1" local fallback="${2:-}" case "$cmd_type" in "memory_info") if [[ "$OS" == "FreeBSD" ]] || [[ "$OS" == "OpenBSD" ]] || [[ "$OS" == "NetBSD" ]]; then echo "sysctl" elif [[ -f /proc/meminfo ]]; then echo "proc" elif command_exists free; then echo "free" else echo "$fallback" fi ;; "disk_usage") if command_exists df; then echo "df" elif command_exists du; then echo "du" else echo "$fallback" fi ;; "network_stats") if [[ -f /proc/net/dev ]]; then echo "proc" elif command_exists netstat; then echo "netstat" elif command_exists ss; then echo "ss" else echo "$fallback" fi ;; "process_info") if command_exists ps; then echo "ps" elif [[ -d /proc ]]; then echo "proc" else echo "$fallback" fi ;; "cpu_info") if [[ "$OS" == "FreeBSD" ]] || [[ "$OS" == "OpenBSD" ]] || [[ "$OS" == "NetBSD" ]]; then echo "sysctl" elif [[ -f /proc/stat ]]; then echo "proc" elif command_exists top; then echo "top" elif command_exists vmstat; then echo "vmstat" else echo "$fallback" fi ;; *) echo "$fallback" ;; esac } # 跨平台的命令执行 execute_system_command() { local cmd_type="$1" local command_method="$2" shift 2 local args=("$@") case "$cmd_type:$command_method" in "memory_info:sysctl") sysctl vm.stats.vm 2>/dev/null || sysctl hw.physmem hw.usermem 2>/dev/null ;; "memory_info:proc") cat /proc/meminfo 2>/dev/null ;; "memory_info:free") free -b 2>/dev/null || free 2>/dev/null ;; "disk_usage:df") df -B1 "${args[@]}" 2>/dev/null || df "${args[@]}" 2>/dev/null ;; "network_stats:proc") cat /proc/net/dev 2>/dev/null ;; "network_stats:netstat") netstat -i 2>/dev/null ;; "cpu_info:sysctl") sysctl kern.cp_time 2>/dev/null ;; "cpu_info:proc") cat /proc/stat 2>/dev/null ;; "cpu_info:top") timeout 3 top -bn1 2>/dev/null | head -10 ;; "cpu_info:vmstat") timeout 3 vmstat 1 2 2>/dev/null | tail -1 ;; *) return 1 ;; esac } # 检测系统信息（优化版 - 减少fork操作） detect_system() { # 一次性获取系统基本信息（减少fork） local system_info=$(uname -srm) IFS=' ' read -r OS KERNEL_VERSION ARCH <<< "$system_info" # FreeBSD特殊优化（避免不必要的检测） if [[ "$OS" == "FreeBSD" ]]; then IS_CONTAINER="false" CONTAINER_TYPE="none" VIRTUALIZATION="none" VER=$(echo "$KERNEL_VERSION" | cut -d'-' -f1) DISTRO_ID="freebsd" DISTRO_NAME="FreeBSD" print_message "$GREEN" "检测到系统: FreeBSD $VER" elif [[ "$OS" == "Darwin" ]]; then IS_CONTAINER="false" CONTAINER_TYPE="none" VIRTUALIZATION="none" VER=$(sw_vers -productVersion 2>/dev/null || echo "$KERNEL_VERSION") DISTRO_ID="macos" DISTRO_NAME="macOS" print_message "$GREEN" "检测到系统: macOS $VER" else # Linux系统的简化检测 IS_CONTAINER="false" CONTAINER_TYPE="none" VIRTUALIZATION="none" # 简化的容器检测（只检查明显标志） if [[ -f /.dockerenv ]]; then IS_CONTAINER="true" CONTAINER_TYPE="docker" fi # 简化的发行版检测 if [[ -f /etc/os-release ]]; then local os_info=$(cat /etc/os-release 2>/dev/null) DISTRO_ID=$(echo "$os_info" | grep '^ID=' | cut -d= -f2 | tr -d '"' || echo "linux") VER=$(echo "$os_info" | grep '^VERSION_ID=' | cut -d= -f2 | tr -d '"' || echo "unknown") DISTRO_NAME=$(echo "$os_info" | grep '^NAME=' | cut -d= -f2 | tr -d '"' || echo "Linux") else DISTRO_ID="linux" VER="unknown" DISTRO_NAME="Linux" fi print_message "$GREEN" "检测到系统: $DISTRO_NAME $VER" fi # 确保变量在全局可用 export OS ARCH KERNEL_VERSION VER DISTRO_ID DISTRO_NAME export IS_CONTAINER CONTAINER_TYPE VIRTUALIZATION } # 检测包管理器（增强版） detect_package_manager() { PKG_MANAGER="" PKG_INSTALL="" PKG_UPDATE="" PKG_SEARCH="" PKG_INFO="" # 根据系统类型和发行版检测包管理器 case "$OS" in FreeBSD|OpenBSD|NetBSD) if command_exists pkg; then PKG_MANAGER="pkg" PKG_INSTALL="pkg install -y" PKG_UPDATE="pkg update" PKG_SEARCH="pkg search" PKG_INFO="pkg info" elif command_exists pkg_add && [[ "$OS" == "OpenBSD" ]]; then PKG_MANAGER="pkg_add" PKG_INSTALL="pkg_add" PKG_UPDATE="pkg_add -u" PKG_SEARCH="pkg_info -Q" PKG_INFO="pkg_info" fi ;; Darwin) if command_exists brew; then PKG_MANAGER="brew" PKG_INSTALL="brew install" PKG_UPDATE="brew update" PKG_SEARCH="brew search" PKG_INFO="brew info" elif command_exists port; then PKG_MANAGER="port" PKG_INSTALL="port install" PKG_UPDATE="port selfupdate" PKG_SEARCH="port search" PKG_INFO="port info" fi ;; Linux|*) # 按优先级和发行版特性检测包管理器 if command_exists apt-get; then PKG_MANAGER="apt-get" PKG_INSTALL="apt-get install -y" PKG_UPDATE="apt-get update" PKG_SEARCH="apt-cache search" PKG_INFO="apt-cache show" elif command_exists apt; then PKG_MANAGER="apt" PKG_INSTALL="apt install -y" PKG_UPDATE="apt update" PKG_SEARCH="apt search" PKG_INFO="apt show" elif command_exists dnf; then PKG_MANAGER="dnf" PKG_INSTALL="dnf install -y" PKG_UPDATE="dnf update -y" PKG_SEARCH="dnf search" PKG_INFO="dnf info" elif command_exists yum; then PKG_MANAGER="yum" PKG_INSTALL="yum install -y" PKG_UPDATE="yum update -y" PKG_SEARCH="yum search" PKG_INFO="yum info" elif command_exists zypper; then PKG_MANAGER="zypper" PKG_INSTALL="zypper install -y" PKG_UPDATE="zypper refresh" PKG_SEARCH="zypper search" PKG_INFO="zypper info" elif command_exists pacman; then PKG_MANAGER="pacman" PKG_INSTALL="pacman -S --noconfirm" PKG_UPDATE="pacman -Sy" PKG_SEARCH="pacman -Ss" PKG_INFO="pacman -Si" elif command_exists apk; then PKG_MANAGER="apk" PKG_INSTALL="apk add" PKG_UPDATE="apk update" PKG_SEARCH="apk search" PKG_INFO="apk info" elif command_exists emerge; then PKG_MANAGER="emerge" PKG_INSTALL="emerge" PKG_UPDATE="emerge --sync" PKG_SEARCH="emerge --search" PKG_INFO="emerge --info" elif command_exists xbps-install; then PKG_MANAGER="xbps" PKG_INSTALL="xbps-install -y" PKG_UPDATE="xbps-install -S" PKG_SEARCH="xbps-query -Rs" PKG_INFO="xbps-query -R" elif command_exists swupd; then PKG_MANAGER="swupd" PKG_INSTALL="swupd bundle-add" PKG_UPDATE="swupd update" PKG_SEARCH="swupd search" PKG_INFO="swupd bundle-info" elif command_exists nix-env; then PKG_MANAGER="nix" PKG_INSTALL="nix-env -i" PKG_UPDATE="nix-channel --update" PKG_SEARCH="nix-env -qa" PKG_INFO="nix-env -qa --description" elif command_exists snap; then PKG_MANAGER="snap" PKG_INSTALL="snap install" PKG_UPDATE="snap refresh" PKG_SEARCH="snap find" PKG_INFO="snap info" elif command_exists flatpak; then PKG_MANAGER="flatpak" PKG_INSTALL="flatpak install -y" PKG_UPDATE="flatpak update -y" PKG_SEARCH="flatpak search" PKG_INFO="flatpak info" fi ;; esac if [[ -n "$PKG_MANAGER" ]]; then print_message "$GREEN" "检测到包管理器: $PKG_MANAGER" else print_message "$YELLOW" "警告: 未检测到支持的包管理器，将尝试手动安装依赖" fi # 导出变量供其他函数使用 export PKG_MANAGER PKG_INSTALL PKG_UPDATE PKG_SEARCH PKG_INFO } # 检查并安装依赖（无需root权限的方法） install_dependencies() { print_message "$BLUE" "检查系统依赖..." local missing_deps=() # 检查必需的命令 if ! command_exists curl; then missing_deps+=("curl") fi if ! command_exists bc; then missing_deps+=("bc") fi # 检查可选的命令 local optional_missing=() if ! command_exists ifstat; then optional_missing+=("ifstat") fi if ! command_exists jq; then optional_missing+=("jq") fi # 报告可选依赖状态 if [[ ${#optional_missing[@]} -gt 0 ]]; then print_message "$YELLOW" "可选依赖未安装: ${optional_missing[*]}" print_message "$YELLOW" "这些依赖缺失不会影响基本功能，但可能限制某些特性" fi if [[ ${#missing_deps[@]} -eq 0 ]]; then print_message "$GREEN" "所有必需依赖已安装" return 0 fi print_message "$YELLOW" "缺少必需依赖: ${missing_deps[*]}" # 根据不同发行版调整包名 local adjusted_deps=() for dep in "${missing_deps[@]}"; do case "$dep" in "bc") if [[ "$DISTRO_ID" == "alpine" ]]; then adjusted_deps+=("bc") else adjusted_deps+=("bc") fi ;; "curl") adjusted_deps+=("curl") ;; *) adjusted_deps+=("$dep") ;; esac done # 尝试安装依赖 if [[ -n "$PKG_MANAGER" ]]; then if command_exists sudo && sudo -n true 2>/dev/null; then print_message "$BLUE" "尝试使用sudo安装依赖..." # 先更新包列表（对于某些包管理器） if [[ "$PKG_MANAGER" == "apt-get" ]] || [[ "$PKG_MANAGER" == "apt" ]]; then sudo $PKG_UPDATE fi for dep in "${adjusted_deps[@]}"; do print_message "$BLUE" "安装 $dep..." if ! sudo $PKG_INSTALL "$dep"; then print_message "$YELLOW" "警告: 无法安装 $dep" fi done else print_message "$YELLOW" "需要sudo权限安装依赖，请手动执行:" print_message "$CYAN" " sudo $PKG_INSTALL ${adjusted_deps[*]}" fi else print_message "$YELLOW" "未检测到包管理器，请手动安装依赖" print_message "$CYAN" "常见安装命令:" print_message "$CYAN" " Ubuntu/Debian: sudo apt-get install ${adjusted_deps[*]}" print_message "$CYAN" " CentOS/RHEL: sudo yum install ${adjusted_deps[*]}" print_message "$CYAN" " Fedora: sudo dnf install ${adjusted_deps[*]}" print_message "$CYAN" " Alpine: sudo apk add ${adjusted_deps[*]}" fi # 简化的依赖检查 if ! command_exists curl && ! command_exists wget; then print_message "$RED" "错误: curl和wget都不可用" print_message "$CYAN" "请安装curl或wget后重试" return 1 fi if ! command_exists bc; then print_message "$YELLOW" "警告: bc未安装，某些计算功能可能受限" fi print_message "$GREEN" "依赖检查完成" } # 创建集中式目录结构 create_directories() { print_message "$BLUE" "创建集中式目录结构..." # 创建主目录和子目录 mkdir -p "$SCRIPT_DIR"/{bin,config,logs,tmp,cache,run,system/{templates,backups}} || error_exit "无法创建目录结构" # 创建安装清单文件 touch "$INSTALL_MANIFEST" # 设置临时目录环境变量 export TMPDIR="$SCRIPT_DIR/tmp" print_message "$GREEN" "✓ 集中式目录结构创建完成" print_message "$CYAN" " 主目录: $SCRIPT_DIR" } # 加载配置 load_config() { if [[ -f "$CONFIG_FILE" ]]; then source "$CONFIG_FILE" else WORKER_URL=$(echo "$DEFAULT_WORKER_URL" | tr -d ' \n\r') SERVER_ID=$(echo "$DEFAULT_SERVER_ID" | tr -d ' \n\r') API_KEY=$(echo "$DEFAULT_API_KEY" | tr -d ' \n\r') INTERVAL="$DEFAULT_INTERVAL" fi } # 记录安装项到安装清单 record_installation() { local type="$1" # 文件类型 local path="$2" # 文件路径 local action="$3" # 执行的操作 local backup="$4" # 备份信息 echo "$type:$path:$action:$backup" >> "$INSTALL_MANIFEST" } # 保存配置 save_config() { # 确保保存前清理空白字符 WORKER_URL=$(echo "$WORKER_URL" | tr -d ' \n\r') SERVER_ID=$(echo "$SERVER_ID" | tr -d ' \n\r') API_KEY=$(echo "$API_KEY" | tr -d ' \n\r') cat > "$CONFIG_FILE" << EOF # VPS监控配置文件 WORKER_URL="$WORKER_URL" SERVER_ID="$SERVER_ID" API_KEY="$API_KEY" INTERVAL="$INTERVAL" EOF print_message "$GREEN" "配置已保存到 $CONFIG_FILE" } # 获取CPU使用率 get_cpu_usage() { local cpu_usage local cpu_load # FreeBSD系统 if [[ "$OS" == "FreeBSD" ]]; then # 使用sysctl获取CPU使用率 if command_exists sysctl; then local cpu_idle=$(sysctl -n kern.cp_time 2>/dev/null | awk '{print $5}' 2>/dev/null || echo "0") local cpu_total=$(sysctl -n kern.cp_time 2>/dev/null | awk '{sum=0; for(i=1;i<=NF;i++) sum+=$i; print sum}' 2>/dev/null || echo "0") # 确保获取到有效数值 cpu_idle=$(sanitize_integer "$cpu_idle" "0") cpu_total=$(sanitize_integer "$cpu_total" "0") if [[ $cpu_total -gt 0 && $cpu_idle -le $cpu_total ]]; then cpu_usage=$(echo "scale=1; 100 - ($cpu_idle * 100 / $cpu_total)" | bc 2>/dev/null || echo "0") # 确保cpu_usage是有效的数字 cpu_usage=$(sanitize_number "$cpu_usage" "0") else cpu_usage="0" fi else cpu_usage="0" fi # FreeBSD负载平均值 local load1="0" load5="0" load15="0" if command_exists sysctl; then load1=$(sysctl -n vm.loadavg 2>/dev/null | awk '{print $2}' 2>/dev/null || echo "0") load5=$(sysctl -n vm.loadavg 2>/dev/null | awk '{print $3}' 2>/dev/null || echo "0") load15=$(sysctl -n vm.loadavg 2>/dev/null | awk '{print $4}' 2>/dev/null || echo "0") # 清理负载数值 load1=$(sanitize_number "$load1" "0") load5=$(sanitize_number "$load5" "0") load15=$(sanitize_number "$load15" "0") fi cpu_load="$load1,$load5,$load15" else # Linux系统 - 多种方法提高兼容性 cpu_usage="0" # 方法1: 使用/proc/stat（最准确的方法） if [[ -f /proc/stat ]]; then local cpu_line=$(head -n1 /proc/stat 2>/dev/null) if [[ -n "$cpu_line" ]]; then local cpu_times=($cpu_line) if [[ ${#cpu_times[@]} -ge 8 ]]; then local idle=${cpu_times[4]} local iowait=${cpu_times[5]:-0} local total=0 # 计算总CPU时间（user + nice + system + idle + iowait + irq + softirq + steal） for i in {1..7}; do if [[ -n "${cpu_times[i]}" && "${cpu_times[i]}" =~ ^[0-9]+$ ]]; then total=$((total + cpu_times[i])) fi done if [[ $total -gt 0 ]]; then cpu_usage=$(echo "scale=1; 100 - (($idle + $iowait) * 100 / $total)" | bc 2>/dev/null || echo "0") fi fi fi fi # 方法2: 使用top命令（如果/proc/stat不可用） if [[ "$cpu_usage" == "0" ]] && command_exists top; then # 尝试不同的top输出格式 local top_output=$(timeout 3 top -bn1 2>/dev/null | head -10) if [[ -n "$top_output" ]]; then # 匹配不同格式的CPU行 if [[ "$top_output" =~ %Cpu\(s\):[[:space:]]*([0-9.]+)[[:space:]]*us.*[[:space:]]+([0-9.]+)[[:space:]]*id ]]; then # 格式: %Cpu(s): 12.5 us, 2.1 sy, 0.0 ni, 85.4 id local idle_percent="${BASH_REMATCH[2]}" cpu_usage=$(echo "scale=1; 100 - $idle_percent" | bc 2>/dev/null || echo "0") elif [[ "$top_output" =~ CPU:[[:space:]]*([0-9.]+)%[[:space:]]*us.*[[:space:]]+([0-9.]+)%[[:space:]]*id ]]; then # 格式: CPU: 12.5% us, 2.1% sy, 85.4% id local idle_percent="${BASH_REMATCH[2]}" cpu_usage=$(echo "scale=1; 100 - $idle_percent" | bc 2>/dev/null || echo "0") fi fi fi # 方法3: 使用vmstat命令（备用方法） if [[ "$cpu_usage" == "0" ]] && command_exists vmstat; then local vmstat_output=$(timeout 3 vmstat 1 2 2>/dev/null | tail -1) if [[ -n "$vmstat_output" ]]; then local idle_percent=$(echo "$vmstat_output" | awk '{print $(NF-2)}' 2>/dev/null || echo "100") if [[ "$idle_percent" =~ ^[0-9]+$ ]]; then cpu_usage=$((100 - idle_percent)) fi fi fi # 确保cpu_usage是有效的数字 cpu_usage=$(sanitize_number "$cpu_usage" "0") # 获取负载平均值 - 多种方法 local load1="0" load5="0" load15="0" if [[ -f /proc/loadavg ]]; then local load_data=$(cat /proc/loadavg 2>/dev/null | awk '{print $1" "$2" "$3}' || echo "0 0 0") read -r load1 load5 load15 <<< "$load_data" elif command_exists uptime; then # 尝试从uptime命令获取负载 local uptime_output=$(uptime 2>/dev/null) if [[ "$uptime_output" =~ load[[:space:]]+average:[[:space:]]*([0-9.]+),[[:space:]]*([0-9.]+),[[:space:]]*([0-9.]+) ]]; then load1="${BASH_REMATCH[1]}" load5="${BASH_REMATCH[2]}" load15="${BASH_REMATCH[3]}" elif [[ "$uptime_output" =~ ([0-9.]+)[[:space:]]+([0-9.]+)[[:space:]]+([0-9.]+)$ ]]; then load1="${BASH_REMATCH[1]}" load5="${BASH_REMATCH[2]}" load15="${BASH_REMATCH[3]}" fi fi # 清理和验证每个负载值 load1=$(sanitize_number "$load1" "0") load5=$(sanitize_number "$load5" "0") load15=$(sanitize_number "$load15" "0") cpu_load="$load1,$load5,$load15" fi echo "{\"usage_percent\":$cpu_usage,\"load_avg\":[$cpu_load]}" } # 获取内存使用情况 get_memory_usage() { local total used free usage_percent # FreeBSD系统 if [[ "$OS" == "FreeBSD" ]]; then if command_exists sysctl; then # FreeBSD内存信息 local page_size=$(sysctl -n hw.pagesize 2>/dev/null || echo "4096") local total_pages=$(sysctl -n vm.stats.vm.v_page_count 2>/dev/null || echo "0") local free_pages=$(sysctl -n vm.stats.vm.v_free_count 2>/dev/null || echo "0") local inactive_pages=$(sysctl -n vm.stats.vm.v_inactive_count 2>/dev/null || echo "0") local cache_pages=$(sysctl -n vm.stats.vm.v_cache_count 2>/dev/null || echo "0") # 清理和验证数值 page_size=$(sanitize_integer "$page_size" "4096") total_pages=$(sanitize_integer "$total_pages" "0") free_pages=$(sanitize_integer "$free_pages" "0") inactive_pages=$(sanitize_integer "$inactive_pages" "0") cache_pages=$(sanitize_integer "$cache_pages" "0") # 计算内存（转换为KB） if [[ $page_size -gt 0 && $total_pages -gt 0 ]]; then total=$(( (total_pages * page_size) / 1024 )) free=$(( ((free_pages + inactive_pages + cache_pages) * page_size) / 1024 )) used=$((total - free)) # 确保数值合理 if [[ $used -lt 0 ]]; then used=0 fi if [[ $free -lt 0 ]]; then free=0 fi else total=0 used=0 free=0 fi else total=0 used=0 free=0 fi else # Linux系统 - 修复内存计算逻辑，确保 used + free = total total=0 used=0 free=0 # 方法1: 使用free命令（最常用且最准确） if command_exists free; then local mem_info=$(free -k 2>/dev/null | grep "^Mem:") if [[ -n "$mem_info" ]]; then total=$(echo "$mem_info" | awk '{print $2}') # 尝试获取available列（第7列，现代Linux系统） local available=$(echo "$mem_info" | awk '{print $7}' 2>/dev/null || echo "") if [[ "$available" =~ ^[0-9]+$ ]]; then # 如果有available列，使用它作为真正的可用内存 free=$available used=$((total - free)) else # 如果没有available列，使用传统方法计算 local mem_free=$(echo "$mem_info" | awk '{print $4}' 2>/dev/null || echo "0") local buff_cache=$(echo "$mem_info" | awk '{print $6}' 2>/dev/null || echo "0") # 验证数据有效性 if [[ "$mem_free" =~ ^[0-9]+$ ]] && [[ "$buff_cache" =~ ^[0-9]+$ ]]; then free=$((mem_free + buff_cache)) used=$((total - free)) else # 如果解析失败，使用第3列作为used，但需要重新计算free local raw_used=$(echo "$mem_info" | awk '{print $3}' 2>/dev/null || echo "0") if [[ "$raw_used" =~ ^[0-9]+$ ]]; then used=$raw_used free=$((total - used)) fi fi fi fi fi # 方法2: 直接读取/proc/meminfo（备用方法） if [[ "$total" == "0" ]] && [[ -f /proc/meminfo ]]; then total=$(grep "^MemTotal:" /proc/meminfo | awk '{print $2}' 2>/dev/null || echo "0") local mem_free=$(grep "^MemFree:" /proc/meminfo | awk '{print $2}' 2>/dev/null || echo "0") local buffers=$(grep "^Buffers:" /proc/meminfo | awk '{print $2}' 2>/dev/null || echo "0") local cached=$(grep "^Cached:" /proc/meminfo | awk '{print $2}' 2>/dev/null || echo "0") local sreclaimable=$(grep "^SReclaimable:" /proc/meminfo | awk '{print $2}' 2>/dev/null || echo "0") # 计算实际可用内存（包括可回收的内存） free=$((mem_free + buffers + cached + sreclaimable)) used=$((total - free)) fi # 方法3: 容器环境特殊处理 (Cgroup V1 & V2) # 即使没有CONTAINER_ENV变量，如果检测到cgroup限制且限制合理，也优先使用 local cgroup_limit="0" local cgroup_usage="0" local cgroup_found=false # 尝试检测 Cgroup V2 if [[ -f /sys/fs/cgroup/memory.max ]]; then local max_raw=$(cat /sys/fs/cgroup/memory.max 2>/dev/null) if [[ "$max_raw" != "max" && "$max_raw" =~ ^[0-9]+$ ]]; then cgroup_limit="$max_raw" if [[ -f /sys/fs/cgroup/memory.current ]]; then cgroup_usage=$(cat /sys/fs/cgroup/memory.current 2>/dev/null || echo "0") cgroup_found=true fi fi fi # 尝试检测 Cgroup V1 (如果V2没找到) if [[ "$cgroup_found" == "false" && -f /sys/fs/cgroup/memory/memory.limit_in_bytes ]]; then local limit_raw=$(cat /sys/fs/cgroup/memory/memory.limit_in_bytes 2>/dev/null || echo "0") # 忽略极大的值 (未限制) if [[ "$limit_raw" =~ ^[0-9]+$ && "$limit_raw" -lt 9223372036854771712 ]]; then cgroup_limit="$limit_raw" cgroup_usage=$(cat /sys/fs/cgroup/memory/memory.usage_in_bytes 2>/dev/null || echo "0") cgroup_found=true fi fi # 如果找到了有效的cgroup限制，并且限制值小于宿主机物理内存(total)，则使用cgroup数据 # 或者如果total是0（上面获取失败），直接使用cgroup数据 if [[ "$cgroup_found" == "true" ]]; then local cgroup_total_kb=$((cgroup_limit / 1024)) # 只有当cgroup限制明显即使有效（例如小于宿主机内存，或者我们确定是在容器里）时才使用 # 这里如果不确定宿主机内存，或者cgroup限制小于宿主机内存，就采用 if [[ "$total" == "0" || "$cgroup_total_kb" -lt "$total" ]]; then total=$cgroup_total_kb used=$((cgroup_usage / 1024)) free=$((total - used)) fi fi # 确保所有值都是有效数字 total=$(sanitize_integer "$total" "0") used=$(sanitize_integer "$used" "0") free=$(sanitize_integer "$free" "0") # 数据一致性验证和修正 - 优化版本 if [[ $total -gt 0 ]]; then # 确保所有值都是有效数字 total=$(sanitize_integer "$total" "0") used=$(sanitize_integer "$used" "0") free=$(sanitize_integer "$free" "0") # 确保 used + free = total 的一致性 local sum=$((used + free)) local diff=$((sum - total)) # 如果差异超过1%，说明数据有问题，需要修正 local tolerance=$((total / 100)) if [[ $tolerance -lt 1024 ]]; then tolerance=1024 # 最小容差1MB fi if [[ ${diff#-} -gt $tolerance ]]; then # 数据不一致，优先保证total的准确性 if [[ $free -gt $total ]]; then # free过大，重置为total free=$total used=0 elif [[ $used -gt $total ]]; then # used过大，重置 used=$total free=0 else # 重新计算used，保证一致性 used=$((total - free)) fi # 最终安全检查 if [[ $used -lt 0 ]]; then used=0 free=$total fi if [[ $free -lt 0 ]]; then free=0 used=$total fi fi else # 如果没有获取到数据，设置默认值 total=0 used=0 free=0 fi fi # 计算使用百分比 if [[ $total -gt 0 ]]; then usage_percent=$(echo "scale=1; $used * 100 / $total" | bc 2>/dev/null || echo "0") # 确保usage_percent是有效的数字 if ! [[ "$usage_percent" =~ ^[0-9]+\.?[0-9]*$ ]]; then usage_percent="0" fi else usage_percent="0" fi echo "{\"total\":$total,\"used\":$used,\"free\":$free,\"usage_percent\":$usage_percent}" } # 获取磁盘使用情况 get_disk_usage() { local total used free usage_percent # 多种方法获取磁盘信息，提高兼容性 if command_exists df; then # 使用-k参数确保输出单位一致（KB） local disk_info=$(df -k / 2>/dev/null | tail -1) if [[ -n "$disk_info" ]]; then # 从KB转换为GB，使用awk进行更安全的计算 total=$(echo "$disk_info" | awk '{printf "%.2f", $2 / 1024 / 1024}' 2>/dev/null || echo "0") used=$(echo "$disk_info" | awk '{printf "%.2f", $3 / 1024 / 1024}' 2>/dev/null || echo "0") free=$(echo "$disk_info" | awk '{printf "%.2f", $4 / 1024 / 1024}' 2>/dev/null || echo "0") usage_percent=$(echo "$disk_info" | awk '{print $5}' | tr -d '%' 2>/dev/null || echo "0") # 验证数据有效性 total=$(sanitize_number "$total" "0") used=$(sanitize_number "$used" "0") free=$(sanitize_number "$free" "0") usage_percent=$(sanitize_integer "$usage_percent" "0") else total="0" used="0" free="0" usage_percent="0" fi else # 如果df不可用，尝试其他方法 total="0" used="0" free="0" usage_percent="0" fi # 容器环境特殊处理 if [[ "${CONTAINER_ENV:-false}" == "true" && "$total" == "0" ]]; then # 在容器中，尝试获取当前目录的磁盘使用情况 if command_exists df; then local container_disk=$(df -k . 2>/dev/null | tail -1) if [[ -n "$container_disk" ]]; then total=$(echo "$container_disk" | awk '{printf "%.2f", $2 / 1024 / 1024}' 2>/dev/null || echo "0") used=$(echo "$container_disk" | awk '{printf "%.2f", $3 / 1024 / 1024}' 2>/dev/null || echo "0") free=$(echo "$container_disk" | awk '{printf "%.2f", $4 / 1024 / 1024}' 2>/dev/null || echo "0") usage_percent=$(echo "$container_disk" | awk '{print $5}' | tr -d '%' 2>/dev/null || echo "0") total=$(sanitize_number "$total" "0") used=$(sanitize_number "$used" "0") free=$(sanitize_number "$free" "0") usage_percent=$(sanitize_integer "$usage_percent" "0") fi fi fi echo "{\"total\":$total,\"used\":$used,\"free\":$free,\"usage_percent\":$usage_percent}" } # 获取网络使用情况 get_network_usage() { local upload_speed=0 local download_speed=0 local total_upload=0 local total_download=0 # FreeBSD系统 if [[ "$OS" == "FreeBSD" ]]; then # 获取默认网络接口 local interface="" # FreeBSD使用不同的route命令格式 if command_exists route; then # 获取默认路由的接口 interface=$(route -n get default 2>/dev/null | grep 'interface:' | awk '{print $2}') fi # 如果没有找到，尝试查找活跃接口 if [[ -z "$interface" ]] && command_exists netstat; then # 尝试1: 查找有流量的接口（排除lo） interface=$(netstat -i -b | awk 'NR>1 && $1 !~ /^lo/ && ($8 > 0 || $11 > 0) {print $1; exit}') # 尝试2: 如果没有找到活跃接口，使用第一个非lo接口 if [[ -z "$interface" ]]; then interface=$(netstat -i -b | awk 'NR>1 && $1 !~ /^lo/ {print $1; exit}') fi fi # 如果还是没找到，使用第一个非lo接口 if [[ -z "$interface" ]] && command_exists ifconfig; then interface=$(ifconfig -l | tr ' ' '\n' | grep -v '^lo' | head -1) fi if [[ -n "$interface" ]] && command_exists netstat; then # 使用netstat获取网络统计 # FreeBSD netstat -i -b 输出格式： # Name Mtu Network Address Ipkts Ierrs Idrop Ibytes Opkts Oerrs Obytes Coll # 同一接口可能有多行，我们只取第一行（Link层的统计） local net_stats=$(netstat -i -b 2>/dev/null | grep "^$interface" | grep "/dev/null || echo "") if [[ -n "$net_stats" ]]; then local raw_download=$(echo "$net_stats" | awk '{print $8}' 2>/dev/null || echo "0") # Ibytes local raw_upload=$(echo "$net_stats" | awk '{print $11}' 2>/dev/null || echo "0") # Obytes # 清理和验证数值 total_download=$(sanitize_integer "$raw_download" "0") total_upload=$(sanitize_integer "$raw_upload" "0") else # 如果没有找到Link统计，尝试其他方法 local net_stats_alt=$(netstat -i -b 2>/dev/null | grep "^$interface" | head -1 2>/dev/null || echo "") if [[ -n "$net_stats_alt" ]]; then local raw_download=$(echo "$net_stats_alt" | awk '{print $8}' 2>/dev/null || echo "0") local raw_upload=$(echo "$net_stats_alt" | awk '{print $11}' 2>/dev/null || echo "0") total_download=$(sanitize_integer "$raw_download" "0") total_upload=$(sanitize_integer "$raw_upload" "0") fi fi # 计算速度（简单方法） local speed_file="${TMPDIR:-/tmp}/vps_monitor_net_${interface}_$(whoami)" mkdir -p "$(dirname "$speed_file")" 2>/dev/null local current_time=$(date +%s) if [[ -f "$speed_file" ]]; then local last_data=$(cat "$speed_file") local last_time=$(echo "$last_data" | cut -d' ' -f1) local last_rx=$(echo "$last_data" | cut -d' ' -f2) local last_tx=$(echo "$last_data" | cut -d' ' -f3) local time_diff=$((current_time - last_time)) if [[ $time_diff -gt 0 ]]; then download_speed=$(( (total_download - last_rx) / time_diff )) upload_speed=$(( (total_upload - last_tx) / time_diff )) # 确保速度不为负数 [[ $download_speed -lt 0 ]] && download_speed=0 [[ $upload_speed -lt 0 ]] && upload_speed=0 fi fi # 保存当前数据供下次使用 echo "$current_time $total_download $total_upload" > "$speed_file" fi else # Linux系统 # 获取默认网络接口 - 多种方法提高兼容性 local interface="" # 方法1: 使用ip命令（现代Linux） if command_exists ip; then interface=$(ip route show default 2>/dev/null | awk '/default/ {print $5}' | head -1) fi # 方法2: 使用route命令（传统方法） if [[ -z "$interface" ]] && command_exists route; then interface=$(route -n 2>/dev/null | awk '/^0.0.0.0/ {print $8}' | head -1) fi # 方法3: 检查/proc/net/route（直接读取内核路由表） if [[ -z "$interface" && -f "/proc/net/route" ]]; then interface=$(awk '/^[^I]/ && $2 == "00000000" {print $1; exit}' /proc/net/route 2>/dev/null) fi # 方法4: 查找活跃的网络接口（改进版） if [[ -z "$interface" && -f "/proc/net/dev" ]]; then # 查找有流量的接口（排除lo和虚拟接口） # 查找有流量的接口（排除lo和虚拟接口） interface=$(awk '/^ *[^:]*:/ { gsub(/:/, "", $1) # 排除回环和虚拟接口 if ($1 != "lo" && $1 !~ /^(docker|br-|veth|tun|tap|virbr|vmnet)/) { # 只要不是排除的接口，就认为是活动接口，不设置流量阈值 print $1 exit } }' /proc/net/dev) fi # 方法5: 如果还是没找到，使用第一个物理网络接口 if [[ -z "$interface" && -f "/proc/net/dev" ]]; then # 优先选择常见的物理接口名称 for prefix in eth ens enp eno wlan wlp; do interface=$(awk -v prefix="$prefix" '/^ *[^:]*:/ { gsub(/:/, "", $1) if ($1 ~ "^" prefix) { print $1 exit } }' /proc/net/dev) if [[ -n "$interface" ]]; then break fi done fi # 方法6: 最后的备选方案 if [[ -z "$interface" && -f "/proc/net/dev" ]]; then interface=$(awk '/^ *[^:]*:/ { gsub(/:/, "", $1) if ($1 != "lo" && $1 !~ /^(docker|br-|veth|tun|tap|virbr|vmnet)/) { print $1 exit } }' /proc/net/dev) fi if [[ -n "$interface" && -f "/proc/net/dev" ]]; then # 获取总流量 local net_line=$(grep "^ *$interface:" /proc/net/dev 2>/dev/null) if [[ -n "$net_line" ]]; then # 解析网络统计数据 # 格式: interface: bytes packets errs drop fifo frame compressed multicast local stats=($net_line) total_download=${stats[1]} # 接收字节数 total_upload=${stats[9]} # 发送字节数 # 确保是数字 if ! [[ "$total_download" =~ ^[0-9]+$ ]]; then total_download=0 fi if ! [[ "$total_upload" =~ ^[0-9]+$ ]]; then total_upload=0 fi fi # 尝试获取实时速度 if command_exists ifstat && [[ -n "$interface" ]]; then # 使用ifstat获取实时速度 local network_speed=$(timeout 3 ifstat -i "$interface" 1 1 2>/dev/null | tail -1) if [[ -n "$network_speed" && "$network_speed" != *"no statistics"* ]]; then download_speed=$(echo "$network_speed" | awk '{printf "%.0f", $1 * 1024}' 2>/dev/null || echo "0") upload_speed=$(echo "$network_speed" | awk '{printf "%.0f", $2 * 1024}' 2>/dev/null || echo "0") fi else # 如果没有ifstat，使用简单的方法计算速度 local speed_file="${TMPDIR:-/tmp}/vps_monitor_net_${interface}_$(whoami)" mkdir -p "$(dirname "$speed_file")" 2>/dev/null local current_time=$(date +%s) if [[ -f "$speed_file" ]]; then local last_data=$(cat "$speed_file") local last_time=$(echo "$last_data" | cut -d' ' -f1) local last_rx=$(echo "$last_data" | cut -d' ' -f2) local last_tx=$(echo "$last_data" | cut -d' ' -f3) local time_diff=$((current_time - last_time)) if [[ $time_diff -gt 0 ]]; then download_speed=$(( (total_download - last_rx) / time_diff )) upload_speed=$(( (total_upload - last_tx) / time_diff )) fi fi # 保存当前数据供下次使用 echo "$current_time $total_download $total_upload" > "$speed_file" fi fi fi # 确保所有值都是数字 [[ "$upload_speed" =~ ^[0-9]+$ ]] || upload_speed=0 [[ "$download_speed" =~ ^[0-9]+$ ]] || download_speed=0 [[ "$total_upload" =~ ^[0-9]+$ ]] || total_upload=0 [[ "$total_download" =~ ^[0-9]+$ ]] || total_download=0 echo "{\"upload_speed\":$upload_speed,\"download_speed\":$download_speed,\"total_upload\":$total_upload,\"total_download\":$total_download}" } # 获取系统运行时间 get_uptime() { local uptime_seconds=0 # FreeBSD系统 if [[ "$OS" == "FreeBSD" ]]; then if command_exists sysctl; then # FreeBSD使用sysctl获取启动时间 local boot_time_raw=$(sysctl -n kern.boottime 2>/dev/null | awk '{print $4}' | tr -d ',' 2>/dev/null || echo "0") local boot_time=$(sanitize_integer "$boot_time_raw" "0") local current_time=$(date +%s) if [[ $boot_time -gt 0 && $current_time -gt $boot_time ]]; then uptime_seconds=$((current_time - boot_time)) else # 如果无法获取启动时间，尝试其他方法 if command_exists uptime; then # 尝试解析uptime命令输出 local uptime_str=$(uptime 2>/dev/null | grep -o 'up [^,]*' | sed 's/up //' || echo "0") # 简化处理，假设格式为 "X days" 或 "X:Y" if [[ "$uptime_str" =~ ([0-9]+).*day ]]; then uptime_seconds=$((${BASH_REMATCH[1]} * 86400)) else uptime_seconds=0 fi else uptime_seconds=0 fi fi else uptime_seconds=0 fi else # Linux系统 if [[ -f /proc/uptime ]]; then uptime_seconds=$(cut -d. -f1 /proc/uptime) elif command_exists uptime; then # 解析uptime命令输出 local uptime_str=$(uptime | awk '{print $3}') # 这里简化处理，实际可能需要更复杂的解析 uptime_seconds=$(echo "$uptime_str" | sed 's/,//' | awk '{print $1 * 86400}' 2>/dev/null || echo "0") fi fi echo "$uptime_seconds" } # 验证和清理数值 sanitize_number() { local value="$1" local default_value="${2:-0}" value=$(echo "$value" | sed 's/[^0-9.]//g') if [[ "$value" =~ ^[0-9]*\.?[0-9]+$ ]] || [[ "$value" =~ ^[0-9]+\.?[0-9]*$ ]]; then [[ "$value" =~ ^\. ]] && value="0$value" [[ "$value" =~ \.$ ]] && value="${value}0" echo "$value" else echo "$default_value" fi } # 验证和清理整数 sanitize_integer() { local value="$1" local default_value="${2:-0}" value=$(echo "$value" | sed 's/[^0-9]//g') [[ "$value" =~ ^[0-9]+$ ]] && echo "$value" || echo "$default_value" } # 清理JSON字符串 clean_json_string() { local input="$1" # 移除可能的控制字符和非打印字符 echo "$input" | tr -d '\000-\037' | tr -d '\177-\377' } # 获取服务器配置（带简单重试） get_config() { local max_attempts=3 local attempt=1 while [[ $attempt -le $max_attempts ]]; do log "正在获取服务器配置... (第 $attempt/$max_attempts 次)" local response=$(curl -s -w "%{http_code}" -X GET "$WORKER_URL/api/config/$SERVER_ID" \ -H "X-API-Key: $API_KEY" 2>/dev/null || echo "000") local http_code="${response: -3}" local response_body="${response%???}" if [[ "$http_code" == "200" ]]; then log "配置获取成功" # 简化的间隔解析 local new_interval=$(echo "$response_body" | sed -n 's/.*"interval":\([0-9]\+\).*/\1/p') if [[ -n "$new_interval" && "$new_interval" =~ ^[0-9]+$ && "$new_interval" -gt 0 ]]; then if [[ "$new_interval" != "$INTERVAL" ]]; then log "服务器返回新的上报间隔: ${new_interval}秒 (当前: ${INTERVAL}秒)" INTERVAL="$new_interval" save_config log "上报间隔已更新为: ${INTERVAL}秒" fi fi save_config_cache "$response_body" return 0 else log "配置获取失败 (HTTP $http_code)" case "$http_code" in "401") log "认证失败 - 请检查API密钥" ;; "404") log "服务器不存在 - 请检查服务器ID" ;; "000") log "网络连接失败" ;; esac if [[ $attempt -lt $max_attempts ]]; then log "等待2秒后重试..." sleep 2 fi fi attempt=$((attempt + 1)) done log "配置获取最终失败" return 1 } # 本地缓存管理 CACHE_DIR="$SCRIPT_DIR/cache" CONFIG_CACHE_FILE="$CACHE_DIR/config.json" METRICS_CACHE_FILE="$CACHE_DIR/last_metrics.json" # 初始化缓存目录 init_cache() { if [[ ! -d "$CACHE_DIR" ]]; then mkdir -p "$CACHE_DIR" log "创建缓存目录: $CACHE_DIR" fi } # 保存配置到缓存 save_config_cache() { local config_data="$1" init_cache echo "$config_data" > "$CONFIG_CACHE_FILE" log "配置已缓存到本地" } # 从缓存加载配置 load_config_cache() { if [[ -f "$CONFIG_CACHE_FILE" ]]; then local cached_config=$(cat "$CONFIG_CACHE_FILE") if [[ -n "$cached_config" ]]; then log "从缓存加载配置" echo "$cached_config" return 0 fi fi return 1 } # 上报监控数据 report_metrics() { local timestamp=$(date +%s) local cpu_raw=$(get_cpu_usage) local memory_raw=$(get_memory_usage) local disk_raw=$(get_disk_usage) local network_raw=$(get_network_usage) local uptime_raw=$(get_uptime) # 验证运行时间 local uptime=$(sanitize_integer "$uptime_raw" "0") # 清理JSON数据 cpu_raw=$(clean_json_string "$cpu_raw") memory_raw=$(clean_json_string "$memory_raw") disk_raw=$(clean_json_string "$disk_raw") network_raw=$(clean_json_string "$network_raw") # 简单验证JSON格式 [[ ! "$cpu_raw" =~ ^\{.*\}$ ]] && cpu_raw='{"usage_percent":0,"load_avg":[0,0,0]}' [[ ! "$memory_raw" =~ ^\{.*\}$ ]] && memory_raw='{"total":0,"used":0,"free":0,"usage_percent":0}' [[ ! "$disk_raw" =~ ^\{.*\}$ ]] && disk_raw='{"total":0,"used":0,"free":0,"usage_percent":0}' [[ ! "$network_raw" =~ ^\{.*\}$ ]] && network_raw='{"upload_speed":0,"download_speed":0,"total_upload":0,"total_download":0}' # 构建JSON数据 local data="{\"timestamp\":$timestamp,\"cpu\":$cpu_raw,\"memory\":$memory_raw,\"disk\":$disk_raw,\"network\":$network_raw,\"uptime\":$uptime}" # 确保API KEY和ID没有多余的空格或换行 local clean_api_key=$(echo "$API_KEY" | tr -d ' \n\r') local clean_server_id=$(echo "$SERVER_ID" | tr -d ' \n\r') log "正在上报数据到 $WORKER_URL/api/report/$clean_server_id" local response=$(curl -s -w "%{http_code}" -X POST "$WORKER_URL/api/report/$clean_server_id" \ -H "Content-Type: application/json" \ -H "X-API-Key: $clean_api_key" \ -d "$data" 2>/dev/null || echo "000") local http_code="${response: -3}" local response_body="${response%???}" if [[ "$http_code" == "200" ]]; then log "数据上报成功" # 尝试从响应中解析新的间隔设置 if command_exists jq; then # 如果有jq命令，使用jq解析 local new_interval=$(echo "$response_body" | jq -r '.interval // empty' 2>/dev/null) if [[ -n "$new_interval" && "$new_interval" =~ ^[0-9]+$ && "$new_interval" -gt 0 ]]; then if [[ "$new_interval" != "$INTERVAL" ]]; then log "服务器返回新的上报间隔: ${new_interval}秒 (当前: ${INTERVAL}秒)" INTERVAL="$new_interval" # 更新配置文件 save_config log "上报间隔已更新为: ${INTERVAL}秒" # 创建重启标记文件，让主循环重启服务以应用新间隔 touch "$SCRIPT_DIR/restart_needed" fi fi else # 如果没有jq，使用简单的文本解析 local new_interval=$(echo "$response_body" | sed -n 's/.*"interval":\([0-9]\+\).*/\1/p') if [[ -n "$new_interval" && "$new_interval" =~ ^[0-9]+$ && "$new_interval" -gt 0 ]]; then if [[ "$new_interval" != "$INTERVAL" ]]; then log "服务器返回新的上报间隔: ${new_interval}秒 (当前: ${INTERVAL}秒)" INTERVAL="$new_interval" # 更新配置文件 save_config log "上报间隔已更新为: ${INTERVAL}秒" # 创建重启标记文件，让主循环重启服务以应用新间隔 touch "$SCRIPT_DIR/restart_needed" fi fi fi return 0 else # 错误分类处理 case "$http_code" in "400"|"413") log "数据上报失败 (HTTP $http_code): 数据格式或大小问题" return 1 # 不可重试的错误 ;; "401"|"403") log "数据上报失败 (HTTP $http_code): 认证失败" return 1 # 不可重试的错误 ;; "404") log "数据上报失败 (HTTP $http_code): 服务器不存在" return 1 # 不可重试的错误 ;; "429"|"500"|"502"|"503"|"504"|"000") log "数据上报失败 (HTTP $http_code): 可重试的错误" return 2 # 可重试的错误 ;; *) log "数据上报失败 (HTTP $http_code): 未知错误" return 1 # 默认不可重试 ;; esac fi } # 创建监控服务脚本 create_service_script() { # 获取当前脚本的绝对路径 local main_script_path=$(realpath "$0") cat > "$SERVICE_FILE" << EOF #!/bin/bash # cf-vps-monitor服务脚本 - 集中式文件管理 SCRIPT_DIR="$SCRIPT_DIR" CONFIG_FILE="\$SCRIPT_DIR/config/config" LOG_FILE="\$SCRIPT_DIR/logs/monitor.log" PID_FILE="\$SCRIPT_DIR/run/monitor.pid" MAIN_SCRIPT="$main_script_path" # 设置服务模式标志（避免日志重复） export SERVICE_MODE=true # 确保日志目录存在 mkdir -p "\$(dirname "\$LOG_FILE")" 2>/dev/null # 加载配置 if [[ -f "\$CONFIG_FILE" ]]; then source "\$CONFIG_FILE" else echo "配置文件不存在: \$CONFIG_FILE" exit 1 fi # 从主脚本加载监控函数（简化版） source_monitoring_functions() { # 直接source主脚本，但设置标志避免执行主程序 if [[ -f "\$MAIN_SCRIPT" ]]; then # 设置标志表示只加载函数 export FUNCTIONS_ONLY=true source "\$MAIN_SCRIPT" unset FUNCTIONS_ONLY else log "错误: 找不到主脚本 \$MAIN_SCRIPT" exit 1 fi } # 加载监控函数 source_monitoring_functions # 清理JSON字符串 clean_json_string() { local input="\$1" # 移除可能的控制字符和非打印字符 echo "\$input" | tr -d '\\000-\\037' | tr -d '\\177-\\377' } # 上报监控数据 report_metrics() { local timestamp=\$(date +%s) local cpu_raw=\$(get_cpu_usage) local memory_raw=\$(get_memory_usage) local disk_raw=\$(get_disk_usage) local network_raw=\$(get_network_usage) local uptime_raw=\$(get_uptime) # 验证运行时间 local uptime=\$(sanitize_integer "\$uptime_raw" "0") # 清理JSON数据 cpu_raw=\$(clean_json_string "\$cpu_raw") memory_raw=\$(clean_json_string "\$memory_raw") disk_raw=\$(clean_json_string "\$disk_raw") network_raw=\$(clean_json_string "\$network_raw") # 验证各个JSON组件（使用更宽松的验证） if [[ -z "\$cpu_raw" || "\$cpu_raw" == "{}" || ! "\$cpu_raw" =~ ^\{.*\}\$ ]]; then cpu_raw='{\\"usage_percent\\":0,\\"load_avg\\":[0,0,0]}' fi if [[ -z "\$memory_raw" || "\$memory_raw" == "{}" || ! "\$memory_raw" =~ ^\{.*\}\$ ]]; then memory_raw='{\\"total\\":0,\\"used\\":0,\\"free\\":0,\\"usage_percent\\":0}' fi if [[ -z "\$disk_raw" || "\$disk_raw" == "{}" || ! "\$disk_raw" =~ ^\{.*\}\$ ]]; then disk_raw='{\\"total\\":0,\\"used\\":0,\\"free\\":0,\\"usage_percent\\":0}' fi if [[ -z "\$network_raw" || "\$network_raw" == "{}" || ! "\$network_raw" =~ ^\{.*\}\$ ]]; then network_raw='{\\"upload_speed\\":0,\\"download_speed\\":0,\\"total_upload\\":0,\\"total_download\\":0}' fi # 构建JSON数据 local data="{\\"timestamp\\":\$timestamp,\\"cpu\\":\$cpu_raw,\\"memory\\":\$memory_raw,\\"disk\\":\$disk_raw,\\"network\\":\$network_raw,\\"uptime\\":\$uptime}" # 确保API KEY和ID没有多余的空格或换行 local clean_api_key=\$(echo "\$API_KEY" | tr -d ' \\n\\r') local clean_server_id=\$(echo "\$SERVER_ID" | tr -d ' \\n\\r') log "正在上报数据..." local response=\$(curl -s -w "%{http_code}" -X POST "\$WORKER_URL/api/report/\$clean_server_id" \\ -H "Content-Type: application/json" \\ -H "X-API-Key: \$clean_api_key" \\ -d "\$data" 2>/dev/null || echo "000") local http_code="\${response: -3}" local response_body="\${response%???}" if [[ "\$http_code" == "200" ]]; then log "数据上报成功" # 尝试从响应中解析新的间隔设置 # 使用简单的文本解析（避免依赖jq） local new_interval=\$(echo "\$response_body" | sed -n 's/.*"interval":\\([0-9]\\+\\).*/\\1/p') if [[ -n "\$new_interval" && "\$new_interval" =~ ^[0-9]+\$ && "\$new_interval" -gt 0 ]]; then if [[ "\$new_interval" != "\$INTERVAL" ]]; then log "服务器返回新的上报间隔: \${new_interval}秒 (当前: \${INTERVAL}秒)" INTERVAL="\$new_interval" # 更新配置文件 cat > "\$CONFIG_FILE" << EOL # VPS监控配置文件 WORKER_URL="\$WORKER_URL" SERVER_ID="\$SERVER_ID" API_KEY="\$API_KEY" INTERVAL="\$INTERVAL" EOL log "上报间隔已更新为: \${INTERVAL}秒" # 创建重启标记文件，让主循环重新加载配置 touch "\$SCRIPT_DIR/restart_needed" fi fi return 0 else log "数据上报失败 (HTTP \$http_code): \$response_body" # 简化的错误处理 case "\$http_code" in "400") log "数据格式错误" ;; "401") log "认证失败 - 请检查API密钥" ;; "404") log "服务器不存在 - 请检查服务器ID" ;; "429") log "请求过于频繁 - 将自动重试" ;; "500"|"503") log "服务器错误 - 将在下个周期重试" ;; "000") log "网络连接失败" ;; *) log "未知错误 (HTTP \$http_code)" ;; esac return 1 fi } # 获取服务器配置 get_config() { log "正在获取服务器配置..." local response=\$(curl -s -w "%{http_code}" -X GET "\$WORKER_URL/api/config/\$SERVER_ID" \\ -H "X-API-Key: \$API_KEY" 2>/dev/null || echo "000") local http_code="\${response: -3}" local response_body="\${response%???}" if [[ "\$http_code" == "200" ]]; then log "配置获取成功" # 尝试解析配置 local new_interval="" # 使用改进的文本解析（避免依赖jq） # 方法1: 使用grep + cut new_interval=\$(echo "\$response_body" | grep -o '"report_interval":[0-9]*' | cut -d':' -f2 2>/dev/null) # 方法2: 如果方法1失败，使用awk备用方案 if [[ -z "\$new_interval" ]]; then new_interval=\$(echo "\$response_body" | awk -F'"report_interval":' '{if(NF>1) print \$2}' | awk -F',' '{print \$1}' | tr -d ' ' 2>/dev/null) fi # 验证并更新间隔设置 if [[ -n "\$new_interval" && "\$new_interval" =~ ^[0-9]+\$ && "\$new_interval" -gt 0 ]]; then if [[ "\$new_interval" != "\$INTERVAL" ]]; then log "检测到新的上报间隔: \${new_interval}秒 (当前: \${INTERVAL}秒)" INTERVAL="\$new_interval" # 更新配置文件 cat > "\$CONFIG_FILE" << EOL # VPS监控配置文件 WORKER_URL="\$WORKER_URL" SERVER_ID="\$SERVER_ID" API_KEY="\$API_KEY" INTERVAL="\$INTERVAL" EOL log "上报间隔已更新为: \${INTERVAL}秒" return 0 else log "配置无变化，当前间隔: \${INTERVAL}秒" return 0 fi else log "警告: 无法解析配置中的上报间隔，保持当前设置" return 1 fi else log "配置获取失败 (HTTP \$http_code): \$response_body" # 简化的错误处理 case "\$http_code" in "401") log "认证失败 - 请检查API密钥" ;; "404") log "服务器不存在 - 请检查服务器ID" ;; "429") log "请求过于频繁 - 将稍后重试" ;; "500"|"503") log "服务器错误 - 将稍后重试" ;; "000") log "网络连接失败" ;; *) log "未知错误 (HTTP \$http_code)" ;; esac return 1 fi } # 主循环 main() { log "VPS监控服务启动 (PID: \$\$)" echo \$\$ > "\$PID_FILE" # 信号处理 trap 'log "收到终止信号，正在停止..."; rm -f "\$PID_FILE"; exit 0' TERM INT # 启动时获取一次配置 log "启动时获取服务器配置..." get_config || log "启动时配置获取失败，使用当前配置" local config_check_counter=0 local config_check_interval=10 # 每10个周期检查一次配置（约10分钟） while true; do # 定期检查配置更新 if [[ \$config_check_counter -ge \$config_check_interval ]]; then log "定期检查配置更新..." get_config || log "配置检查失败，继续使用当前配置" config_check_counter=0 else config_check_counter=\$((config_check_counter + 1)) fi if ! report_metrics; then log "上报失败，将在下个周期重试" fi # 检查是否需要重启以应用新的间隔设置 if [[ -f "\$SCRIPT_DIR/restart_needed" ]]; then log "检测到间隔设置变更，正在重新加载配置..." rm -f "\$SCRIPT_DIR/restart_needed" # 重新加载配置 if [[ -f "\$CONFIG_FILE" ]]; then source "\$CONFIG_FILE" log "已重新加载配置，新的上报间隔: \${INTERVAL}秒" fi fi sleep "\$INTERVAL" done } # 启动主函数 main EOF chmod +x "$SERVICE_FILE" print_message "$GREEN" "监控服务脚本创建完成: $SERVICE_FILE" } # ==================== 用户类型检测和systemd命令选择机制 ==================== # 检测当前用户类型 detect_user_type() { if [[ $EUID -eq 0 ]]; then echo "root" else echo "user" fi } # 检查是否为root用户 is_root_user() { [[ $EUID -eq 0 ]] } # 检查systemd服务可用性（根据用户类型） check_systemd_availability() { if ! command_exists systemctl; then return 1 fi if is_root_user; then # root用户检查系统级systemd systemctl status >/dev/null 2>&1 else # 普通用户检查用户级systemd systemctl --user status >/dev/null 2>&1 fi } # 创建systemd服务（兼容版） create_systemd_service() { print_message "$BLUE" "配置systemd服务..." # 检查systemd可用性 if ! is_user_systemd_available; then print_message "$YELLOW" "systemd不可用，跳过systemd服务配置" return 1 fi # 确定服务文件路径 local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" mkdir -p "$(dirname "$service_path")" fi # 生成服务文件内容 print_message "$CYAN" " 生成systemd服务文件: $service_path" # 创建服务文件模板目录 mkdir -p "$SCRIPT_DIR/system/templates" # 生成服务文件内容 if is_root_user; then cat > "$SCRIPT_DIR/system/templates/systemd.service" << EOF [Unit] Description=cf-vps-monitor Service - VPS Monitoring Agent Documentation=https://github.com/kadidalax/cf-vps-monitor After=network.target network-online.target Wants=network-online.target [Service] Type=simple ExecStart=$SERVICE_FILE Restart=always RestartSec=10 User=root Group=root WorkingDirectory=$SCRIPT_DIR [Install] WantedBy=multi-user.target EOF else cat > "$SCRIPT_DIR/system/templates/systemd.service" << EOF [Unit] Description=cf-vps-monitor Service - VPS Monitoring Agent Documentation=https://github.com/kadidalax/cf-vps-monitor After=network.target network-online.target Wants=network-online.target [Service] Type=simple ExecStart=$SERVICE_FILE Restart=always RestartSec=10 WorkingDirectory=$SCRIPT_DIR [Install] WantedBy=default.target EOF fi # 复制模板到系统位置 cp "$SCRIPT_DIR/system/templates/systemd.service" "$service_path" # 记录到安装清单 record_installation "systemd" "$service_path" "create" "none" # 重新加载systemd配置 if is_root_user; then safe_systemctl daemon-reload safe_systemctl enable cf-vps-monitor.service else safe_systemctl --user daemon-reload safe_systemctl --user enable cf-vps-monitor.service fi print_message "$GREEN" "✓ systemd服务创建完成: $service_path" return 0 } # ==================== systemd lingering支持 ==================== # 简化的lingering启用 enable_lingering() { # root用户不需要lingering if is_root_user; then return 0 fi # 检查loginctl是否可用 if ! command_exists loginctl; then return 1 fi # 尝试启用lingering（静默处理） loginctl enable-linger "$USER" 2>/dev/null || true return 0 } # 启动监控服务 start_service() { local user_desc=$(get_user_type_description) print_message "$BLUE" "启动监控服务 ($user_desc)..." # 1. 检查是否已有进程在运行 if is_monitor_running; then local pids=$(find_monitor_processes) local first_pid=$(echo "$pids" | awk '{print $1}') print_message "$YELLOW" "监控服务已在运行 (PID: $first_pid)" return 0 fi # 2. 清理旧的PID文件 rm -f "$PID_FILE" 2>/dev/null || true # 3. 尝试使用systemd启动（简化版） local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi if [[ -f "$service_path" ]] && is_systemd_available; then local systemd_cmd="systemctl" [[ ! $(is_root_user) ]] && systemd_cmd="systemctl --user" if $systemd_cmd start cf-vps-monitor.service 2>/dev/null; then $systemd_cmd enable cf-vps-monitor.service 2>/dev/null || true print_message "$GREEN" "✓ 监控服务已启动 (systemd)" # 自动配置其他自启动设置 echo add_autostart_settings echo print_message "$CYAN" "提示: 已配置自启动设置，重启后监控服务会自动启动" return 0 fi fi # 4. 传统方式启动 print_message "$BLUE" "使用传统方式启动服务..." if [[ ! -f "$SERVICE_FILE" ]]; then print_message "$RED" "✗ 服务脚本不存在: $SERVICE_FILE" print_message "$CYAN" "请先运行安装命令" return 1 fi chmod +x "$SERVICE_FILE" 2>/dev/null || true if command_exists nohup; then nohup "$SERVICE_FILE" >> "$LOG_FILE" 2>&1 & else "$SERVICE_FILE" >> "$LOG_FILE" 2>&1 & fi local pid=$! echo "$pid" > "$PID_FILE" # 5. 验证启动成功 sleep 2 if kill -0 "$pid" 2>/dev/null; then print_message "$GREEN" "✓ 监控服务已启动 (PID: $pid)" print_message "$CYAN" "日志文件: $LOG_FILE" # 自动配置自启动设置 echo add_autostart_settings echo print_message "$CYAN" "提示: 已配置自启动设置，重启后监控服务会自动启动" return 0 else print_message "$RED" "✗ 监控服务启动失败" if [[ -f "$LOG_FILE" ]]; then print_message "$YELLOW" "查看日志: tail -f $LOG_FILE" fi rm -f "$PID_FILE" return 1 fi } # 渐进式停止单个进程（改进版） stop_single_process() { local pid="$1" # 首先验证PID if ! validate_pid "$pid"; then print_message "$YELLOW" " ⚠ PID $pid 无效或进程不存在" return 1 fi # 获取正确的进程信息 local cmd=$(get_process_command "$pid") print_message "$BLUE" "停止进程: $cmd (PID: $pid)" # 1. 温和停止（SIGTERM） if kill "$pid" 2>/dev/null; then sleep 2 # 2. 检查是否还在运行 if ! kill -0 "$pid" 2>/dev/null; then print_message "$GREEN" " ✓ 进程已正常停止" return 0 fi # 3. 强制停止（SIGKILL） if kill -9 "$pid" 2>/dev/null; then sleep 1 # 4. 最终确认 if ! kill -0 "$pid" 2>/dev/null; then print_message "$GREEN" " ✓ 进程已强制停止" return 0 else print_message "$RED" " ✗ 进程无法停止" return 1 fi fi fi print_message "$YELLOW" " ⚠ 无法发送信号" return 1 } # 停止监控服务 stop_service() { local user_desc=$(get_user_type_description) print_message "$BLUE" "停止监控服务 ($user_desc)..." local stopped=false # 1. 尝试使用systemd停止（简化版） local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi if [[ -f "$service_path" ]] && is_systemd_available; then local systemd_cmd="systemctl" [[ ! $(is_root_user) ]] && systemd_cmd="systemctl --user" if $systemd_cmd is-active cf-vps-monitor.service >/dev/null 2>&1; then $systemd_cmd stop cf-vps-monitor.service 2>/dev/null $systemd_cmd disable cf-vps-monitor.service 2>/dev/null || true stopped=true fi fi # 2. 查找并停止所有相关进程（使用精确检测） local pids=$(find_monitor_processes) if [[ -n "$pids" ]]; then local stopped_count=0 local total_count=0 for pid in $pids; do total_count=$((total_count + 1)) if stop_single_process "$pid"; then stopped_count=$((stopped_count + 1)) stopped=true fi done if [[ $stopped_count -gt 0 ]]; then print_message "$GREEN" "✓ 已停止 $stopped_count/$total_count 个监控进程" fi fi # 3. 清理PID文件 rm -f "$PID_FILE" 2>/dev/null || true # 4. 结果报告和自启动清理 if [[ "$stopped" == "true" ]]; then print_message "$GREEN" "✓ 监控服务已停止" # 自动移除自启动设置 echo remove_autostart_settings echo print_message "$CYAN" "提示: 已移除自启动设置，重启后监控服务不会自动启动" print_message "$CYAN" "如需重新启用监控，请使用启动功能" else print_message "$YELLOW" "没有发现运行中的监控服务" fi } # 检查服务状态 check_service_status() { local user_desc=$(get_user_type_description) print_message "$BLUE" "检查监控服务状态 ($user_desc)..." echo # 1. 使用精确检测逻辑 if is_monitor_running; then local pids=$(find_monitor_processes) local pid_count=$(echo "$pids" | wc -w) print_message "$GREEN" "✓ 监控服务正在运行" if [[ $pid_count -eq 1 ]]; then local pid=$(echo "$pids" | awk '{print $1}') local cmd=$(get_process_command "$pid") print_message "$CYAN" " 进程信息: PID $pid" print_message "$CYAN" " 命令行: $cmd" else print_message "$YELLOW" " 发现多个进程实例 ($pid_count 个):" for pid in $pids; do local cmd=$(get_process_command "$pid") print_message "$CYAN" " PID $pid: $cmd" done fi else print_message "$RED" "✗ 监控服务未运行" fi # 2. 检查systemd状态（如果可用） local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi if [[ -f "$service_path" ]] && is_systemd_available; then local systemd_cmd="systemctl" [[ ! $(is_root_user) ]] && systemd_cmd="systemctl --user" echo print_message "$BLUE" "systemd服务状态:" if $systemd_cmd is-active cf-vps-monitor.service >/dev/null 2>&1; then print_message "$GREEN" " ✓ systemd服务活跃" $systemd_cmd status cf-vps-monitor.service --no-pager -l 2>/dev/null || true else print_message "$YELLOW" " ✗ systemd服务未活跃" fi fi # 3. 检查PID文件状态 echo print_message "$BLUE" "PID文件状态:" if [[ -f "$PID_FILE" ]]; then local file_pid=$(cat "$PID_FILE" 2>/dev/null) if [[ -n "$file_pid" ]]; then if kill -0 "$file_pid" 2>/dev/null; then print_message "$GREEN" " ✓ PID文件有效 (PID: $file_pid)" else print_message "$YELLOW" " ⚠ PID文件存在但进程不存在 (清理中...)" rm -f "$PID_FILE" fi else print_message "$YELLOW" " ⚠ PID文件为空" fi else print_message "$YELLOW" " ✗ PID文件不存在" fi # 4. 显示配置信息 echo print_message "$BLUE" "配置信息:" if [[ -f "$CONFIG_FILE" ]]; then load_config print_message "$CYAN" " Worker URL: $WORKER_URL" print_message "$CYAN" " Server ID: $SERVER_ID" print_message "$CYAN" " API Key: ${API_KEY:0:8}..." print_message "$CYAN" " 上报间隔: ${INTERVAL}秒" else print_message "$YELLOW" " ✗ 配置文件不存在" fi # 5. 显示日志文件信息 echo print_message "$BLUE" "日志文件:" if [[ -f "$LOG_FILE" ]]; then local log_size=$(du -h "$LOG_FILE" 2>/dev/null | cut -f1) local log_lines=$(wc -l < "$LOG_FILE" 2>/dev/null || echo "0") print_message "$CYAN" " 文件: $LOG_FILE" print_message "$CYAN" " 大小: $log_size" print_message "$CYAN" " 行数: $log_lines" else print_message "$YELLOW" " ✗ 日志文件不存在" fi # 显示自启动状态 echo print_message "$CYAN" "自启动配置状态:" local active_count=0 # 检查systemd服务状态 local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" print_message "$CYAN" " systemd服务 (系统管理员):" if [[ -f "$service_path" ]] && command_exists systemctl; then if systemctl is-enabled cf-vps-monitor.service >/dev/null 2>&1; then print_message "$GREEN" " ✓ 服务已启用" active_count=$((active_count + 1)) print_message "$GREEN" " ✓ 系统级服务 (重启后自动运行)" else print_message "$YELLOW" " ✗ 服务未启用" fi else print_message "$YELLOW" " ✗ systemd服务文件不存在" fi else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" print_message "$CYAN" " systemd服务 (普通用户):" if [[ -f "$service_path" ]] && command_exists systemctl; then if systemctl --user is-enabled cf-vps-monitor.service >/dev/null 2>&1; then print_message "$GREEN" " ✓ 服务已启用" active_count=$((active_count + 1)) else print_message "$YELLOW" " ✗ 服务未启用" fi else print_message "$YELLOW" " ✗ systemd服务文件不存在" fi fi # 检查crontab状态 print_message "$CYAN" " crontab自启动:" if check_crontab_autostart; then print_message "$GREEN" " ✓ 已配置 (重启后自动运行)" active_count=$((active_count + 1)) else print_message "$YELLOW" " ✗ 未配置" fi # 检查profile状态 print_message "$CYAN" " shell profile自启动:" if [[ -f "$HOME/.bashrc" ]] && grep -q "cf-vps-monitor auto-start" "$HOME/.bashrc" 2>/dev/null; then print_message "$GREEN" " ✓ 已配置 (登录时自动运行)" active_count=$((active_count + 1)) else print_message "$YELLOW" " ✗ 未配置" fi # 自启动状态总结 echo print_message "$BLUE" "自启动保障总结:" echo " 活跃方案数: $active_count / 3" if [[ $active_count -eq 0 ]]; then print_message "$RED" " 状态: 无自启动保障" print_message "$YELLOW" " 建议: 重新安装服务以配置自启动" elif [[ $active_count -eq 1 ]]; then print_message "$YELLOW" " 状态: 基本保障" print_message "$CYAN" " 建议: 重新安装服务以配置完整保障" elif [[ $active_count -eq 2 ]]; then print_message "$GREEN" " 状态: 良好保障" else print_message "$GREEN" " 状态: 完全保障 (推荐)" fi # 如果检测有问题，提供诊断选项 if ! is_monitor_running && [[ $active_count -gt 0 ]]; then echo print_message "$YELLOW" "提示: 配置了自启动但服务未运行，输入 'd' 查看详细诊断" echo -n "是否查看诊断信息? (d/N): " read -r -t 10 diag_choice if [[ "$diag_choice" =~ ^[Dd]$ ]]; then echo diagnose_monitor_service fi fi } # ==================== crontab自启动方案 ==================== # 设置crontab自启动 setup_crontab_autostart() { print_message "$BLUE" "配置crontab自启动..." # 检查crontab可用性 if ! command_exists crontab; then return 1 fi # 获取当前crontab（减少fork操作） local current_crontab=$(crontab -l 2>/dev/null || echo "") # 检查是否已配置 if echo "$current_crontab" | grep -q "cf-vps-monitor"; then print_message "$GREEN" "✓ crontab自启动已存在" return 0 fi # 备份当前crontab local backup_file="$SCRIPT_DIR/system/backups/crontab_backup" echo "$current_crontab" > "$backup_file" # 优先级启动条目（简化进程检测） local crontab_entry="@reboot sleep 30 && pgrep -f 'cf-vps-monitor|vps-monitor-service' >/dev/null || $SERVICE_FILE" # 添加新条目（减少临时文件操作） if (echo "$current_crontab"; echo "$crontab_entry") | crontab - 2>/dev/null; then # 记录到安装清单 record_installation "crontab" "$USER" "add" "$backup_file" print_message "$GREEN" "✓ crontab自启动已配置" return 0 else return 1 fi } # 检查crontab自启动状态 check_crontab_autostart() { if ! command_exists crontab; then return 1 fi if crontab -l 2>/dev/null | grep -q "$SERVICE_FILE"; then return 0 else return 1 fi } # ==================== shell profile自启动方案 ==================== # 设置shell profile自启动 setup_profile_autostart() { print_message "$BLUE" "配置shell profile自启动..." local profile="$HOME/.bashrc" # 检查是否已配置 if grep -q "cf-vps-monitor auto-start" "$profile" 2>/dev/null; then return 0 fi # 备份原文件 local backup_file="$SCRIPT_DIR/system/backups/bashrc_backup" cp "$profile" "$backup_file" 2>/dev/null || touch "$backup_file" # 添加自启动代码 cat >> "$profile" << EOF # === cf-vps-monitor auto-start BEGIN === # VPS监控服务自启动检测 (最后保障) if [ -n "\$PS1" ] && [ "\$TERM" != "dumb" ]; then if ! pgrep -f 'cf-vps-monitor|vps-monitor-service' >/dev/null 2>&1; then (sleep 5 && nohup "$SERVICE_FILE" >/dev/null 2>&1 &) & fi fi # === cf-vps-monitor auto-start END === EOF # 记录到安装清单 record_installation "profile" "$profile" "modify" "$backup_file" print_message "$GREEN" "✓ shell profile自启动已配置" return 0 } # ==================== 多重自启动方案协调器 ==================== # 配置优先级自启动 setup_auto_start() { print_message "$BLUE" "配置优先级自启动机制..." echo # 检查服务脚本 if [[ ! -f "$SERVICE_FILE" ]]; then print_message "$RED" "✗ 服务脚本不存在，请先运行安装" return 1 fi local success_count=0 local total_attempts=3 # 优先级1: systemd服务 if is_user_systemd_available; then print_message "$CYAN" "优先级1: systemd服务" if create_systemd_service; then success_count=$((success_count + 1)) print_message "$GREEN" " ✓ systemd服务已配置" if ! is_root_user; then enable_lingering >/dev/null 2>&1 fi else print_message "$YELLOW" " ✗ systemd服务配置失败" fi else print_message "$YELLOW" " - systemd不可用，跳过" total_attempts=$((total_attempts - 1)) fi # 优先级2: crontab备用 print_message "$CYAN" "优先级2: crontab备用" if setup_crontab_autostart; then success_count=$((success_count + 1)) print_message "$GREEN" " ✓ crontab备用已配置" else print_message "$YELLOW" " ✗ crontab备用配置失败" fi # 优先级3: shell profile保障 print_message "$CYAN" "优先级3: shell profile保障" if setup_profile_autostart; then success_count=$((success_count + 1)) print_message "$GREEN" " ✓ profile保障已配置" else print_message "$YELLOW" " ✗ profile保障配置失败" fi echo if [[ $success_count -eq 0 ]]; then print_message "$RED" "✗ 所有自启动方案配置失败" return 1 else print_message "$GREEN" "✓ 配置了 $success_count/$total_attempts 种自启动方案" if [[ $success_count -eq $total_attempts ]]; then print_message "$GREEN" "完全保障" elif [[ $success_count -ge 2 ]]; then print_message "$GREEN" "良好保障" else print_message "$YELLOW" "基本保障" fi fi print_message "$CYAN" "优先级: systemd > crontab > profile" return 0 } # 查看日志 view_logs() { if [[ ! -f "$LOG_FILE" ]]; then print_message "$YELLOW" "日志文件不存在: $LOG_FILE" return fi print_message "$BLUE" "显示最近50行日志:" echo "----------------------------------------" tail -n 50 "$LOG_FILE" echo "----------------------------------------" print_message "$CYAN" "日志文件位置: $LOG_FILE" } # 测试连接 test_connection() { print_message "$BLUE" "测试连接到监控服务器..." load_config if [[ -z "$WORKER_URL" || -z "$SERVER_ID" || -z "$API_KEY" ]]; then print_message "$RED" "配置不完整，请先配置监控参数" return 1 fi print_message "$BLUE" "正在测试配置获取..." if get_config; then print_message "$GREEN" "✓ 配置获取测试成功" else print_message "$YELLOW" "⚠ 配置获取测试失败，但不影响基本功能" fi print_message "$BLUE" "正在测试数据上报..." if report_metrics; then print_message "$GREEN" "✓ 数据上报测试成功" else print_message "$RED" "✗ 数据上报测试失败，请检查配置和网络" return 1 fi print_message "$GREEN" "✓ 连接测试完成" } # 配置监控参数 configure_monitor() { print_message "$BLUE" "配置监控参数" echo load_config # Server ID echo -n "请输入Server ID" if [[ -n "$SERVER_ID" ]]; then echo -n " (当前: $SERVER_ID)" fi echo -n ": " read -r input_server_id if [[ -n "$input_server_id" ]]; then SERVER_ID="$input_server_id" fi # API Key echo -n "请输入API Key" if [[ -n "$API_KEY" ]]; then echo -n " (当前: ${API_KEY:0:8}...)" fi echo -n ": " read -r input_api_key if [[ -n "$input_api_key" ]]; then API_KEY="$input_api_key" fi # Worker URL echo -n "请输入Worker URL" if [[ -n "$WORKER_URL" ]]; then echo -n " (当前: $WORKER_URL)" fi echo -n ": " read -r input_url if [[ -n "$input_url" ]]; then WORKER_URL="$input_url" fi # 设置默认上报间隔为10秒，脚本会自动从服务器获取最新配置 if [[ -z "$INTERVAL" ]]; then INTERVAL="10" fi print_message "$CYAN" "上报间隔设置为: ${INTERVAL}秒 (脚本运行后会自动从服务器获取最新配置)" # 验证配置 if [[ -z "$WORKER_URL" || -z "$SERVER_ID" || -z "$API_KEY" ]]; then print_message "$RED" "配置不完整，请确保所有必需参数都已填写" return 1 fi # 保存配置 save_config print_message "$GREEN" "配置保存成功" # 询问是否测试连接 echo echo -n "是否测试连接? (y/N): " read -r test_choice if [[ "$test_choice" =~ ^[Yy]$ ]]; then test_connection fi } # 安装监控服务 install_monitor() { print_message "$BLUE" "开始安装VPS监控服务..." echo # 检查系统资源（防止fork错误） if ! check_system_resources; then print_message "$YELLOW" "系统资源紧张，启用简化模式" fi # 检测系统 detect_system detect_package_manager # 安装依赖 install_dependencies # 创建目录结构 create_directories # 配置监控参数 if ! configure_monitor; then error_exit "配置失败，安装中止" fi # 创建服务脚本 create_service_script # 创建systemd服务（如果可用） local systemd_available=false if create_systemd_service; then systemd_available=true fi # 配置多重自启动保障 echo print_message "$BLUE" "配置自启动机制..." if setup_auto_start; then print_message "$GREEN" "✓ 自启动机制配置完成" else print_message "$YELLOW" "⚠ 自启动配置部分失败，但不影响基本功能" fi # 启动服务 echo if start_service; then print_message "$GREEN" "✓ VPS监控服务安装并启动成功" echo print_message "$CYAN" "安装信息:" echo " 安装目录: $SCRIPT_DIR" echo " 配置文件: $CONFIG_FILE" echo " 日志文件: $LOG_FILE" echo " 服务脚本: $SERVICE_FILE" if [[ "$systemd_available" == "true" ]]; then local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi echo " systemd服务: $service_path" print_message "$GREEN" " 启动方式: systemd服务" else print_message "$GREEN" " 启动方式: 传统后台进程" fi echo print_message "$GREEN" "✓ 已配置多重自启动保障，VPS重启后将自动运行" echo print_message "$YELLOW" "提示: 使用 '$0 status' 检查服务状态和自启动状态" print_message "$YELLOW" "提示: 使用 '$0 logs' 查看运行日志" else error_exit "服务启动失败" fi } # 集中式彻底卸载监控服务 uninstall_monitor() { print_message "$YELLOW" "警告: 这将删除VPS监控服务及其数据" echo -n "确认卸载? (y/N): " read -r confirm if [[ ! "$confirm" =~ ^[Yy]$ ]]; then print_message "$BLUE" "取消卸载" return 0 fi print_message "$BLUE" "开始集中式卸载VPS监控服务..." # 1. 停止服务 stop_service # 2. 保护脚本本身 local script_path=$(realpath "$0") local need_backup=false if [[ "$script_path" == "$SCRIPT_DIR"/* ]]; then need_backup=true local backup_script="/tmp/cf-vps-monitor-backup.sh" cp "$script_path" "$backup_script" chmod +x "$backup_script" print_message "$CYAN" "已备份脚本到: $backup_script" fi # 3. 清理系统集成文件（兼容所有系统） if [[ -f "$INSTALL_MANIFEST" ]]; then print_message "$CYAN" "清理系统集成文件..." while IFS=':' read -r type path action backup; do case "$type" in "systemd") print_message "$CYAN" " 移除systemd服务: $path" rm -f "$path" 2>/dev/null || true # 只在有systemctl的系统上重载 if command_exists systemctl; then if is_root_user; then systemctl daemon-reload 2>/dev/null || true else systemctl --user daemon-reload 2>/dev/null || true fi fi ;; "crontab") print_message "$CYAN" " 清理crontab条目" (crontab -l 2>/dev/null || echo "") | grep -v "cf-vps-monitor" | crontab - 2>/dev/null || true ;; "profile") print_message "$CYAN" " 清理profile修改: $path" # 兼容FreeBSD的sed语法 if [[ "$OS" == "FreeBSD" ]] || [[ "$OS" == "Darwin" ]]; then sed -i '' '/# === cf-vps-monitor auto-start BEGIN ===/,/# === cf-vps-monitor auto-start END ===/d' "$path" 2>/dev/null || true else sed -i '/# === cf-vps-monitor auto-start BEGIN ===/,/# === cf-vps-monitor auto-start END ===/d' "$path" 2>/dev/null || true fi ;; esac done < "$INSTALL_MANIFEST" 2>/dev/null || true fi # 4. 强制删除安装目录（多重保障） print_message "$BLUE" "删除安装目录: $SCRIPT_DIR" # 确保不在目标目录内执行删除 cd / 2>/dev/null || cd "$HOME" 2>/dev/null || true # 尝试删除，如果失败提供详细信息 if ! rm -rf "$SCRIPT_DIR" 2>/dev/null; then print_message "$YELLOW" "标准删除失败，尝试强制删除..." # 尝试逐个删除文件 if [[ -d "$SCRIPT_DIR" ]]; then find "$SCRIPT_DIR" -type f -exec rm -f {} \; 2>/dev/null || true find "$SCRIPT_DIR" -type d -exec rmdir {} \; 2>/dev/null || true # 最后尝试删除主目录 rmdir "$SCRIPT_DIR" 2>/dev/null || rm -rf "$SCRIPT_DIR" 2>/dev/null || true fi fi # 5. 验证删除结果并提供反馈 if [[ -d "$SCRIPT_DIR" ]]; then print_message "$YELLOW" "⚠ 安装目录仍然存在: $SCRIPT_DIR" print_message "$CYAN" "可能原因: 文件被占用或权限不足" print_message "$CYAN" "手动删除: rm -rf '$SCRIPT_DIR'" # 显示目录内容帮助诊断 if [[ -r "$SCRIPT_DIR" ]]; then print_message "$CYAN" "目录内容:" ls -la "$SCRIPT_DIR" 2>/dev/null || true fi else print_message "$GREEN" "✓ VPS监控服务已彻底卸载" if [[ "$need_backup" == "true" ]]; then print_message "$YELLOW" "注意: 当前脚本已被删除，但备份在: $backup_script" else print_message "$CYAN" "当前脚本未被删除，可以重新安装" fi fi } # 显示帮助信息 show_help() { echo "VPS监控脚本 v2.0" echo echo "用法: $0 [选项] [参数]" echo echo "基本选项:" echo " install 安装监控服务" echo " uninstall 彻底卸载监控服务" echo " start 启动监控服务" echo " stop 停止监控服务" echo " restart 重启监控服务" echo " status 查看服务状态" echo " logs 查看运行日志" echo " config 配置监控参数" echo " test 测试连接" echo " menu 显示交互菜单" echo " help 显示此帮助信息" echo echo "一键安装参数:" echo " -i, --install 一键安装模式" echo " -s, --server-id ID 服务器ID" echo " -k, --api-key KEY API密钥" echo " -u, --worker-url URL Worker地址" echo echo "示例:" echo " $0 install # 交互式安装" echo " $0 status # 查看服务状态" echo " $0 logs # 查看日志" echo echo "一键安装示例:" echo " $0 -i -s server123 -k abc123 -u https://worker.example.com" echo echo "注意: 上报间隔会自动从服务器获取，无需手动设置" } # 显示交互菜单 show_menu() { while true; do clear print_message "$CYAN" "==================================" print_message "$CYAN" " VPS监控服务管理菜单" print_message "$CYAN" "==================================" echo echo "1. 安装监控服务" echo "2. 启动监控服务" echo echo "3. 停止监控服务" echo "4. 重启监控服务" echo echo "5. 查看服务状态" echo "6. 查看运行日志" echo echo "7. 配置监控参数" echo "8. 测试连接" echo print_message "$CYAN" "特殊操作:" echo "9. 彻底卸载服务" echo "0. 退出" echo print_message "$YELLOW" "请选择操作 (0-9): " read -r choice case $choice in 1) echo install_monitor echo print_message "$BLUE" "按任意键继续..." read -r ;; 2) echo start_service echo print_message "$BLUE" "按任意键继续..." read -r ;; 3) echo stop_service echo print_message "$BLUE" "按任意键继续..." read -r ;; 4) echo stop_service sleep 1 start_service echo print_message "$BLUE" "按任意键继续..." read -r ;; 5) echo check_service_status echo print_message "$BLUE" "按任意键继续..." read -r ;; 6) echo view_logs echo print_message "$BLUE" "按任意键继续..." read -r ;; 7) echo configure_monitor echo print_message "$BLUE" "按任意键继续..." read -r ;; 8) echo test_connection echo print_message "$BLUE" "按任意键继续..." read -r ;; 9) echo uninstall_monitor echo print_message "$BLUE" "按任意键继续..." read -r ;; 0) print_message "$GREEN" "感谢使用VPS监控服务！" exit 0 ;; *) print_message "$RED" "无效选择，请重新输入" sleep 1 ;; esac done } # 解析命令行参数 parse_arguments() { local install_mode=false local server_id="" local api_key="" local worker_url="" while [[ $# -gt 0 ]]; do case $1 in -i|--install) install_mode=true shift ;; -s|--server-id) server_id="$2" shift 2 ;; -k|--api-key) api_key="$2" shift 2 ;; -u|--worker-url) worker_url="$2" shift 2 ;; -h|--help) show_help exit 0 ;; *) # 如果是基本命令，返回处理 return 1 ;; esac done # 如果是一键安装模式 if [[ "$install_mode" == "true" ]]; then one_click_install "$server_id" "$api_key" "$worker_url" exit $? fi return 1 } # 一键安装函数 one_click_install() { local server_id="$1" local api_key="$2" local worker_url="$3" print_message "$BLUE" "开始一键安装VPS监控服务..." echo # 验证必需参数 if [[ -z "$server_id" || -z "$api_key" || -z "$worker_url" ]]; then print_message "$RED" "错误: 缺少必需参数" echo "必需参数: -s <服务器ID> -k -u " echo "使用 '$0 --help' 查看详细帮助" return 1 fi # 设置默认间隔为10秒（会自动从服务器获取最新配置） local interval="10" print_message "$CYAN" "安装参数:" echo " 服务器ID: $server_id" echo " API密钥: ${api_key:0:8}..." echo " Worker地址: $worker_url" echo " 初始上报间隔: ${interval}秒 (运行后会自动从服务器获取最新配置)" echo # 检测系统 detect_system detect_package_manager # 安装依赖 install_dependencies # 创建目录结构 create_directories # 设置配置参数 WORKER_URL="$worker_url" SERVER_ID="$server_id" API_KEY="$api_key" INTERVAL="$interval" # 保存配置 save_config print_message "$GREEN" "配置保存成功" # 测试连接 print_message "$BLUE" "测试连接..." if ! report_metrics; then print_message "$YELLOW" "警告: 连接测试失败，但将继续安装" print_message "$YELLOW" "请检查网络连接和配置参数" else print_message "$GREEN" "✓ 连接测试成功" fi # 创建服务脚本 create_service_script # 创建systemd服务（如果可用） local systemd_available=false if create_systemd_service; then systemd_available=true fi # 配置多重自启动保障 echo print_message "$BLUE" "配置自启动机制..." if setup_auto_start; then print_message "$GREEN" "✓ 自启动机制配置完成" else print_message "$YELLOW" "⚠ 自启动配置部分失败，但不影响基本功能" fi # 启动服务 echo if start_service; then print_message "$GREEN" "✓ VPS监控服务一键安装成功" echo print_message "$CYAN" "安装信息:" echo " 安装目录: $SCRIPT_DIR" echo " 配置文件: $CONFIG_FILE" echo " 日志文件: $LOG_FILE" echo " 服务脚本: $SERVICE_FILE" if [[ "$systemd_available" == "true" ]]; then local service_path if is_root_user; then service_path="/etc/systemd/system/cf-vps-monitor.service" else service_path="$HOME/.config/systemd/user/cf-vps-monitor.service" fi echo " systemd服务: $service_path" print_message "$GREEN" " 启动方式: systemd服务" else print_message "$GREEN" " 启动方式: 传统后台进程" fi echo print_message "$GREEN" "✓ 已配置多重自启动保障，VPS重启后将自动运行" echo print_message "$YELLOW" "提示: 使用 '$0 status' 检查服务状态和自启动状态" print_message "$YELLOW" "提示: 使用 '$0 logs' 查看运行日志" return 0 else print_message "$RED" "✗ 服务启动失败" return 1 fi } # 主函数 main() { # 首先尝试解析命令行参数 if parse_arguments "$@"; then return fi # 如果没有参数，显示菜单 if [[ $# -eq 0 ]]; then show_menu return fi # 处理命令行参数 case "$1" in install) install_monitor ;; uninstall) uninstall_monitor ;; start) start_service ;; stop) stop_service ;; restart) stop_service sleep 1 start_service ;; status) check_service_status ;; logs) view_logs ;; config) configure_monitor ;; test) test_connection ;; menu) show_menu ;; help|--help|-h) show_help ;; *) print_message "$RED" "未知选项: $1" echo show_help exit 1 ;; esac } # 函数加载模式支持（用于服务脚本） if [[ "${FUNCTIONS_ONLY:-false}" == "true" ]]; then # 只加载函数，不执行主程序 return 0 2>/dev/null || exit 0 fi # 脚本入口点 if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then main "$@" fi