feat(monitoring): 添加自动化服务监测Shell脚本

- 实现日志暴涨检测功能，按总行数窗口监控日志变化
- 添加ERROR上下文检测，提取最近错误日志的上下文信息
- 集成内存监控功能，记录内存使用峰值和平均值
- 实现MySQL连接数监控，支持容器和本机两种方式
- 支持新旧平台自动识别，适配不同部署环境
- 提供完整的日志记录和告警机制
- 实现定时循环监控，支持自定义监控间隔

AuxiliaryTool/ScriptTool/自动化服务监测/AutomatedServiceMonitoring.sh

+''''bash
+# filepath: /data/scripts/AutomatedServiceMonitoring.sh
+#!/usr/bin/env bash
+
+########################################
+# 自动化服务监测 Shell 脚本（单机版）
+# 在哪台服务器上运行，就监控哪台服务器（本机）
+########################################
+
+set -u  # 未定义变量直接报错
+# 如需调试可打开：set -x
+
+#################### 全局配置 ####################
+
+# 监控间隔（秒）
+INTERVAL_SECONDS=60
+
+# MySQL 账号（与 PRD 保持一致）
+MYSQL_USER="root"
+MYSQL_PASSWORD="dNrprU&2S"
+
+# 日志文件（默认写当前目录）
+LOG_FILE="./AutomatedServiceMonitoring.sh.log"
+
+# 当前主机标识
+HOST_NAME="$(hostname)"
+
+#################### 简单日志函数 ####################
+log() {
+    # $1: 级别 INFO/WARN/ERROR
+    # $2+: 消息
+    local level="$1"; shift
+    local ts
+    ts="$(date '+%Y-%m-%d %H:%M:%S')"
+    local msg="[$ts] [$level] $*"
+    echo "$msg"
+    echo "$msg" >>"$LOG_FILE"
+}
+
+#################### 平台识别 ####################
+detect_platform() {
+    # 输出两个全局变量：
+    #   PLATFORM_TYPE="new" 或 "legacy"
+    #   BASE_PATH="/data/services" 或 "/var/www"
+    if [ -d "/data/services" ]; then
+        PLATFORM_TYPE="new"
+        BASE_PATH="/data/services"
+    else
+        PLATFORM_TYPE="legacy"
+        BASE_PATH="/var/www"
+    fi
+}
+
+#################### 系统识别（ujava/upython/upython_voice） ####################
+detect_systems() {
+    HAS_UJAVA=0
+    HAS_UPYTHON=0
+    HAS_UPYTHON_VOICE=0
+    SYSTEMS=()    # bash 数组：meeting / ops / transcription
+
+    # 检查 docker 容器名
+    local names
+    names="$(docker ps --format '{{.Names}}' 2>/dev/null || true)"
+
+    if echo "$names" | grep -q "ujava"; then
+        HAS_UJAVA=1
+        SYSTEMS+=("meeting")
+    fi
+    if echo "$names" | grep -q "upython_voice"; then
+        HAS_UPYTHON_VOICE=1
+        SYSTEMS+=("transcription")
+    fi
+    if echo "$names" | grep -q "upython"; then
+        HAS_UPYTHON=1
+        SYSTEMS+=("ops")
+    fi
+}
+
+#################### 日志目标解析 ####################
+resolve_log_targets() {
+    # 根据 PLATFORM_TYPE / BASE_PATH / SYSTEMS 生成日志列表
+    # 结果放到全局数组：LOG_TARGET_SYS[] / LOG_TARGET_PATH[]
+    LOG_TARGET_SYS=()
+    LOG_TARGET_PATH=()
+
+    local has_meeting=0
+    for s in "${SYSTEMS[@]}"; do
+        if [[ "$s" == "meeting" ]]; then
+            has_meeting=1
+        fi
+    done
+
+    if [[ $has_meeting -eq 1 ]]; then
+        if [[ "$PLATFORM_TYPE" == "new" ]]; then
+            LOG_TARGET_SYS+=("meeting-2.0")
+            LOG_TARGET_PATH+=("$BASE_PATH/api/java-meeting/java-meeting2.0/logs/ubains-INFO-AND-ERROR.log")
+
+            LOG_TARGET_SYS+=("meeting-3.0")
+            LOG_TARGET_PATH+=("$BASE_PATH/api/java-meeting/java-meeting3.0/logs/ubains-INFO-AND-ERROR.log")
+        else
+            LOG_TARGET_SYS+=("meeting-2.0")
+            LOG_TARGET_PATH+=("/var/www/java/api-java-meeting2.0/logs/ubains-INFO-AND-ERROR.log")
+        fi
+    fi
+}
+
+#################### 日志暴涨检测（按总行数窗口） ####################
+# 使用关联数组记录上一次总行数和时间戳： key=日志路径
+declare -A BURST_LAST_TOTAL
+declare -A BURST_LAST_TS
+
+monitor_log_burst_once() {
+    local sys_name="$1"
+    local log_path="$2"
+    local window_seconds=300
+    local min_lines_threshold=1000
+    local rate_threshold_per_sec=5
+
+    # 检查文件是否存在
+    if [ ! -f "$log_path" ]; then
+        log INFO "[日志暴涨审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 文件不存在"
+        return
+    fi
+
+    local out
+    out="$(wc -l < "$log_path" 2>/dev/null || echo 0)"
+    local total_lines
+    total_lines=$(echo "$out" | tr -d ' ')
+
+    if ! [[ "$total_lines" =~ ^[0-9]+$ ]]; then
+        log INFO "[日志暴涨审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 无法获取总行数"
+        return
+    fi
+
+    local now_ts
+    now_ts="$(date +%s)"
+
+    local key="$log_path"
+    local last_total="${BURST_LAST_TOTAL[$key]-}"
+    local last_ts="${BURST_LAST_TS[$key]-}"
+
+    if [[ -z "${last_total:-}" || -z "${last_ts:-}" ]]; then
+        BURST_LAST_TOTAL["$key"]="$total_lines"
+        BURST_LAST_TS["$key"]="$now_ts"
+        log INFO "[日志暴涨审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 初始化窗口，总行数=$total_lines"
+        return
+    fi
+
+    local elapsed=$(( now_ts - last_ts ))
+    if (( elapsed <= 0 )); then
+        elapsed=1
+    fi
+
+    local delta_lines=$(( total_lines - last_total ))
+    if (( delta_lines < 0 )); then
+        delta_lines=0
+    fi
+
+    # 更新状态
+    BURST_LAST_TOTAL["$key"]="$total_lines"
+    BURST_LAST_TS["$key"]="$now_ts"
+
+    if (( elapsed < window_seconds )); then
+        log INFO "[日志暴涨审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 累积中：窗口=${elapsed}s 未达 ${window_seconds}s，新增行数=${delta_lines}"
+        return
+    fi
+
+    # shell 里做速率计算
+    local rate
+    rate=$(awk -v d="$delta_lines" -v e="$elapsed" 'BEGIN{ if(e<=0){e=1}; printf "%.2f", d/e }')
+
+    local start_ts_human end_ts_human
+    start_ts_human="$(date -d @"$last_ts" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date '+%Y-%m-%d %H:%M:%S')"
+    end_ts_human="$(date -d @"$now_ts" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date '+%Y-%m-%d %H:%M:%S')"
+
+    if (( delta_lines >= min_lines_threshold )) || awk -v r="$rate" -v th="$rate_threshold_per_sec" 'BEGIN{exit !(r>=th)}'; then
+        log WARN "[日志打印暴涨] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 窗口=${elapsed}s 新增行数=${delta_lines} 速率=${rate}行/秒 时间段=[${start_ts_human} ~ ${end_ts_human}]"
+    else
+        log INFO "[日志暴涨审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 窗口=${elapsed}s 新增行数=${delta_lines} 速率=${rate}行/秒 未发现暴涨"
+    fi
+}
+
+#################### ERROR 上下文检测 ####################
+monitor_log_errors_once() {
+    local sys_name="$1"
+    local log_path="$2"
+    local max_context=50
+    local max_sections=3
+
+    if [ ! -f "$log_path" ]; then
+        log INFO "[ERROR审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 文件不存在"
+        return
+    fi
+
+    # 最近 5000 行
+    local tail_out
+    tail_out="$(tail -n 5000 "$log_path" 2>/dev/null || true)"
+    if [[ -z "${tail_out//[[:space:]]/}" ]]; then
+        log INFO "[ERROR审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 最近5000行无内容"
+        return
+    fi
+
+    # ERROR 数量
+    local cnt
+    cnt="$(echo "$tail_out" | grep -c "ERROR" || true)"
+    if ! [[ "$cnt" =~ ^[0-9]+$ ]]; then
+        cnt=0
+    fi
+
+    if (( cnt == 0 )); then
+        log INFO "[ERROR审计] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 最近段落未发现 ERROR"
+        return
+    fi
+
+    # 上下文抓取：对最近 5000 行做 nl 标号，取最近几处 ERROR
+    local ctx
+    ctx="$(echo "$tail_out" | nl -ba | \
+        awk '{if($0 ~ /ERROR/) print $1}' | tail -n "$max_sections" | \
+        while read -r ln; do
+            start=$((ln - max_context)); end=$((ln + max_context))
+            if [ "$start" -lt 1 ]; then start=1; fi
+            echo "$tail_out" | sed -n "${start},${end}p"
+            echo
+            echo "---- 上下文分隔线 (行号: $ln) ----"
+            echo
+        done
+    )"
+
+    # 本地提取时间戳（YYYY-MM-DD HH:MM:SS）
+    local start_ts end_ts
+    start_ts="$(echo "$ctx" | grep -oE '[0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}:[0-9]{2}' | head -n1)"
+    end_ts="$(echo "$ctx" | grep -oE '[0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}:[0-9]{2}' | tail -n1)"
+
+    if [[ -n "$start_ts" || -n "$end_ts" ]]; then
+        log WARN "[ERROR出现] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 最近5000行 ERROR数量=$cnt 发生时间段=[$start_ts ~ $end_ts]"
+    else
+        log WARN "[ERROR出现] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path 最近5000行 ERROR数量=$cnt"
+    fi
+
+    log INFO "[ERROR上下文] 主机=$HOST_NAME 系统=$sys_name 日志=$log_path
+$ctx"
+}
+
+#################### 内存监控（/proc/meminfo） ####################
+# 按主机维度记录峰值和平均
+MEM_SAMPLES=0
+MEM_SUM_USED_MB=0
+MEM_PEAK_USED_MB=0
+MEM_PEAK_TS=0
+
+monitor_mem_once() {
+    local meminfo
+    meminfo="$(cat /proc/meminfo 2>/dev/null)" || {
+        log ERROR "[内存监测] $HOST_NAME => 获取 /proc/meminfo 失败"
+        return
+    }
+
+    local total_kb avail_kb
+    total_kb="$(echo "$meminfo" | awk '/^MemTotal:/ {print $2}')"
+    avail_kb="$(echo "$meminfo" | awk '/^MemAvailable:/ {print $2}')"
+
+    if [[ -z "$total_kb" || -z "$avail_kb" ]]; then
+        log ERROR "[内存监测] $HOST_NAME => 解析失败"
+        return
+    fi
+
+    local total_mb used_mb
+    total_mb=$(awk -v t="$total_kb" 'BEGIN{printf "%.0f",t/1024}')
+    used_mb=$(awk -v t="$total_kb" -v a="$avail_kb" 'BEGIN{u=(t-a)/1024; if(u<0)u=0; printf "%.0f",u}')
+
+    MEM_SAMPLES=$((MEM_SAMPLES + 1))
+    MEM_SUM_USED_MB=$((MEM_SUM_USED_MB + used_mb))
+    if (( used_mb > MEM_PEAK_USED_MB )); then
+        MEM_PEAK_USED_MB=$used_mb
+        MEM_PEAK_TS="$(date +%s)"
+    fi
+
+    local avg_used
+    avg_used=$(( MEM_SUM_USED_MB / MEM_SAMPLES ))
+
+    local peak_human="N/A"
+    if [[ "$MEM_PEAK_TS" != "0" && -n "$MEM_PEAK_TS" ]]; then
+        peak_human="$(date -d @"$MEM_PEAK_TS" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date '+%Y-%m-%d %H:%M:%S')"
+    fi
+
+    log INFO "[内存监测] $HOST_NAME 当前使用=${used_mb}MB 总=${total_mb}MB 平均=${avg_used}MB 峰值=${MEM_PEAK_USED_MB}MB@$peak_human"
+}
+
+#################### MySQL 连接数监控 ####################
+MYSQL_SAMPLES=0
+MYSQL_SUM_CONN=0
+MYSQL_PEAK_CONN=0
+MYSQL_PEAK_TS=0
+MYSQL_LAST_TOTAL=""
+MYSQL_LAST_TS=""
+
+find_mysql_container() {
+    # 回显容器名或空
+    docker ps --format '{{.Names}} {{.Image}}' 2>/dev/null | \
+        awk '{
+            name=$1; $1=""; image=$0;
+            low=tolower(name " " image);
+            if(index(low,"mysql")>0 || index(low,"mariadb")>0){
+                print name; exit 0;
+            }
+        }'
+}
+
+get_mysql_threads_connected_via_container() {
+    local container="$1"
+    local auth="-u${MYSQL_USER} -p${MYSQL_PASSWORD}"
+
+    local out
+    out="$(docker exec -i "${container}" mysql -ss ${auth} -e "SHOW STATUS LIKE 'Threads_connected';" 2>/dev/null | tail -n1 | cut -f2)"
+    if [[ "$out" =~ ^[0-9]+$ ]]; then
+        echo "$out"
+        return 0
+    fi
+
+    out="$(docker exec -i "${container}" mysqladmin ${auth} status 2>/dev/null)"
+    local val
+    val="$(echo "$out" | grep -oE 'Threads:[[:space:]]*[0-9]+' | awk '{print $2}')"
+    if [[ "$val" =~ ^[0-9]+$ ]]; then
+        echo "$val"
+        return 0
+    fi
+    return 1
+}
+
+monitor_mysql_once() {
+    local conn=""
+    local container
+    container="$(find_mysql_container || true)"
+
+    if [[ -n "$container" ]]; then
+        conn="$(get_mysql_threads_connected_via_container "$container" || true)"
+        if ! [[ "$conn" =~ ^[0-9]+$ ]]; then
+            log ERROR "[MySQL监测] $HOST_NAME => 在容器 $container 内获取连接数失败"
+            conn=""
+        fi
+    else
+        log INFO "[MySQL监测] $HOST_NAME => 未检测到 MySQL/MariaDB 容器，回退本机命令"
+    fi
+
+    if [[ -z "$conn" ]]; then
+        local auth="-u${MYSQL_USER} -p${MYSQL_PASSWORD}"
+        local out
+        out="$(mysqladmin ${auth} status 2>/dev/null || true)"
+        conn="$(echo "$out" | grep -oE 'Threads:[[:space:]]*[0-9]+' | awk '{print $2}')"
+    fi
+
+    if [[ -z "$conn" || ! "$conn" =~ ^[0-9]+$ ]]; then
+        local auth="-u${MYSQL_USER} -p${MYSQL_PASSWORD}"
+        local out2
+        out2="$(mysql -ss ${auth} -e "SHOW STATUS LIKE 'Threads_connected';" 2>/dev/null | tail -n1 | cut -f2)"
+        if [[ "$out2" =~ ^[0-9]+$ ]]; then
+            conn="$out2"
+        fi
+    fi
+
+    if [[ -z "$conn" || ! "$conn" =~ ^[0-9]+$ ]]; then
+        log ERROR "[MySQL监测] $HOST_NAME => 获取连接数失败（请确认 Docker/MySQL 客户端与权限）"
+        return
+    fi
+
+    local now_ts
+    now_ts="$(date +%s)"
+
+    MYSQL_SAMPLES=$((MYSQL_SAMPLES + 1))
+    MYSQL_SUM_CONN=$((MYSQL_SUM_CONN + conn))
+    if (( conn > MYSQL_PEAK_CONN )); then
+        MYSQL_PEAK_CONN=$conn
+        MYSQL_PEAK_TS="$now_ts"
+    fi
+
+    local avg_conn
+    avg_conn=$(( MYSQL_SUM_CONN / MYSQL_SAMPLES ))
+
+    local peak_human="N/A"
+    if [[ -n "$MYSQL_PEAK_TS" && "$MYSQL_PEAK_TS" != "0" ]]; then
+        peak_human="$(date -d @"$MYSQL_PEAK_TS" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date '+%Y-%m-%d %H:%M:%S')"
+    fi
+
+    log INFO "[MySQL监测] $HOST_NAME 当前连接数=$conn 平均=$avg_conn 峰值=${MYSQL_PEAK_CONN}@$peak_human"
+
+    # 暴涨判定
+    local window_seconds=300
+    local min_burst_conn=200
+    local rate_threshold_per_sec=1
+
+    if [[ -z "${MYSQL_LAST_TOTAL:-}" || -z "${MYSQL_LAST_TS:-}" ]]; then
+        MYSQL_LAST_TOTAL="$conn"
+        MYSQL_LAST_TS="$now_ts"
+        return
+    fi
+
+    local elapsed=$(( now_ts - MYSQL_LAST_TS ))
+    if (( elapsed <= 0 )); then elapsed=1; fi
+    local delta_conn=$(( conn - MYSQL_LAST_TOTAL ))
+    if (( delta_conn < 0 )); then delta_conn=0; fi
+
+    MYSQL_LAST_TOTAL="$conn"
+    MYSQL_LAST_TS="$now_ts"
+
+    if (( elapsed >= window_seconds )); then
+        local rate
+        rate=$(awk -v d="$delta_conn" -v e="$elapsed" 'BEGIN{ if(e<=0){e=1}; printf "%.2f", d/e }')
+        local start_ts_human end_ts_human
+        start_ts_human="$(date -d @"$((now_ts - elapsed))" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date '+%Y-%m-%d %H:%M:%S')"
+        end_ts_human="$(date -d @"$now_ts" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date '+%Y-%m-%d %H:%M:%S')"
+
+        if (( delta_conn >= min_burst_conn )) || awk -v r="$rate" -v th="$rate_threshold_per_sec" 'BEGIN{exit !(r>=th)}'; then
+            log WARN "[MySQL连接暴涨] $HOST_NAME 窗口=${elapsed}s 增量=${delta_conn} 速率=${rate}/s 时间段=[${start_ts_human} ~ ${end_ts_human}]"
+        else
+            log INFO "[MySQL暴涨审计] $HOST_NAME 窗口=${elapsed}s 增量=${delta_conn} 速率=${rate}/s 未发现暴涨"
+        fi
+    fi
+}
+
+#################### 主循环：监控本机 ####################
+main_loop() {
+    log INFO "[启动] 自动化服务监测 Shell 脚本，目标服务器(本机)=${HOST_NAME}"
+
+    detect_platform
+    detect_systems
+    log INFO "[平台识别] $HOST_NAME => 平台类型=$([[ "$PLATFORM_TYPE" == "new" ]] && echo 新统一平台(/data/services) || echo 传统平台(/var/www)) 基路径=${BASE_PATH}"
+    log INFO "[系统识别] $HOST_NAME => ujava=${HAS_UJAVA} upython=${HAS_UPYTHON} upython_voice=${HAS_UPYTHON_VOICE} 系统=(${SYSTEMS[*]:-无})"
+
+    while true; do
+        log INFO "[心跳] $HOST_NAME: 平台=$([[ "$PLATFORM_TYPE" == "new" ]] && echo 新 || echo 旧) 基路径=${BASE_PATH} 系统=${SYSTEMS[*]:-无}"
+
+        resolve_log_targets
+
+        # 日志审计
+        local i
+        for ((i=0; i<${#LOG_TARGET_SYS[@]}; i++)); do
+            local sys_name="${LOG_TARGET_SYS[$i]}"
+            local log_path="${LOG_TARGET_PATH[$i]}"
+
+            monitor_log_burst_once "$sys_name" "$log_path"
+            monitor_log_errors_once "$sys_name" "$log_path"
+        done
+
+        # 内存监测
+        monitor_mem_once
+
+        # MySQL 监测
+        monitor_mysql_once
+
+        sleep "$INTERVAL_SECONDS"
+    done
+}
+
+#################### 脚本入口 ####################
+main_loop
\ No newline at end of file