modify playwright

测试方案：

- lm2机器端口映射到本机：18080, 18081, 8082-8085
- 访问URL: http://localhost:18080/dashboard

![image.png](/attachments/30ed6e20-697a-4d3b-a6d3-6acccd2e9922)

![image.png](/attachments/38ef1751-0f3b-49c6-9100-f70d15617acc)

![image.png](/attachments/3be45005-9b9e-4165-8ef6-1d27405800f1)

![image.png](/attachments/eb916192-edc1-4096-8f9f-9769ab6d9039)

![image.png](/attachments/620e6efc-bd02-45ae-bba1-99a95a1b4c02)

![image.png](/attachments/986e77e7-c687-405f-a760-93282249f72f)

端到端测试通过：

![image.png](/attachments/c6e29875-4a16-4718-8b2f-368f64eb545e)

Co-authored-by: sundapeng.sdp <sundapeng@hashdata.cn>
Reviewed-on: #35
Reviewed-by: xuxt <xuxt@zgclab.edu.cn>
Reviewed-by: sundapeng <sundp@mail.zgclab.edu.cn>
Reviewed-by: huhy <husteryezi@163.com>

.gitattributes

@@ -0,0 +1 @@
+src/metric/client-plugins/all-in-one-full/plugins/*/bin/* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+.idea/

README.md

@@ -5,3 +5,10 @@
 项目文档：【腾讯文档】GPU集群运维系统
 https://docs.qq.com/doc/DQUxDdmhIZ1dpeERk
 
+## 构建账号配置
+
+镜像构建和运行账号的 UID/GID 可通过 `configs/build_user.conf` 配置，详细说明见 `doc/build-user-config.md`。
+
+## 本地端口占用提示
+
+如需运行 BIND 模块端到端测试且宿主机 53 端口已占用，可通过环境变量 `HOST_DNS_PORT`（默认 1053）指定对外映射端口，例如 `HOST_DNS_PORT=12053 ./scripts/00_e2e_test.sh`。

build/build_images.sh

@@ -10,18 +10,29 @@ Usage: $0 [OPTIONS]
 Options:
   --intranet        Use intranet mirror for log/bind builds
   --master-offline  Build master offline image (requires src/master/offline_wheels.tar.gz)
+  --metric          Build metric module images (ftp, prometheus, grafana, test nodes)
+  --no-cache        Build all images without using Docker layer cache
+  --only LIST       Comma-separated targets to build: core,master,metric,web,alert,sys,all
   -h, --help        Show this help message
 
 Examples:
   $0                             # Build with default sources
   $0 --intranet                  # Build with intranet mirror
   $0 --master-offline            # Additionally build argus-master:offline
-  $0 --intranet --master-offline
+  $0 --metric                    # Additionally build metric module images
+  $0 --intranet --master-offline --metric
 EOF
 }
 
 use_intranet=false
+build_core=true
+build_master=true
 build_master_offline=false
+build_metric=true
+build_web=true
+build_alert=true
+build_sys=true
+no_cache=false
 
 while [[ $# -gt 0 ]]; do
   case $1 in
@@ -29,10 +40,44 @@ while [[ $# -gt 0 ]]; do
       use_intranet=true
       shift
       ;;
+    --master)
+      build_master=true
+      shift
+      ;;
     --master-offline)
+      build_master=true
       build_master_offline=true
       shift
       ;;
+    --metric)
+      build_metric=true
+      shift
+      ;;
+    --no-cache)
+      no_cache=true
+      shift
+      ;;
+    --only)
+      if [[ -z ${2:-} ]]; then
+        echo "--only requires a target list" >&2; exit 1
+      fi
+      sel="$2"; shift 2
+      # reset all, then enable selected
+      build_core=false; build_master=false; build_metric=false; build_web=false; build_alert=false; build_sys=false
+      IFS=',' read -ra parts <<< "$sel"
+      for p in "${parts[@]}"; do
+        case "$p" in
+          core) build_core=true ;;
+          master) build_master=true ;;
+          metric) build_metric=true ;;
+          web) build_web=true ;;
+          alert) build_alert=true ;;
+          sys) build_sys=true ;;
+          all) build_core=true; build_master=true; build_metric=true; build_web=true; build_alert=true; build_sys=true ;;
+          *) echo "Unknown --only target: $p" >&2; exit 1 ;;
+        esac
+      done
+      ;;
     -h|--help)
       show_help
       exit 0
@@ -46,8 +91,23 @@ while [[ $# -gt 0 ]]; do
 done
 
 root="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+. "$root/scripts/common/build_user.sh"
+
+declare -a build_args=()
+
+if [[ "$use_intranet" == true ]]; then
+  build_args+=("--build-arg" "USE_INTRANET=true")
+fi
+
 cd "$root"
 
+load_build_user
+build_args+=("--build-arg" "ARGUS_BUILD_UID=${ARGUS_BUILD_UID}" "--build-arg" "ARGUS_BUILD_GID=${ARGUS_BUILD_GID}")
+
+if [[ "$no_cache" == true ]]; then
+  build_args+=("--no-cache")
+fi
+
 master_root="$root/src/master"
 master_offline_tar="$master_root/offline_wheels.tar.gz"
 master_offline_dir="$master_root/offline_wheels"
@@ -75,12 +135,12 @@ echo "======================================="
 
 if [[ "$use_intranet" == true ]]; then
   echo "🌐 Mode: Intranet (Using internal mirror: 10.68.64.1)"
-  build_args="--build-arg USE_INTRANET=true"
 else
   echo "🌐 Mode: Public (Using default package sources)"
-  build_args=""
 fi
 
+echo "👤 Build user UID:GID -> ${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}"
+
 echo "📁 Build context: $root"
 echo ""
 
@@ -88,13 +148,22 @@ build_image() {
   local image_name=$1
   local dockerfile_path=$2
   local tag=$3
-  local extra_args=$4
+  local context="."
+  shift 3
+
+  if [[ $# -gt 0 ]]; then
+    context=$1
+    shift
+  fi
+
+  local extra_args=("$@")
 
   echo "🔄 Building $image_name image..."
   echo "   Dockerfile: $dockerfile_path"
   echo "   Tag: $tag"
+  echo "   Context: $context"
 
-  if docker build $build_args $extra_args -f "$dockerfile_path" -t "$tag" .; then
+  if docker build "${build_args[@]}" "${extra_args[@]}" -f "$dockerfile_path" -t "$tag" "$context"; then
     echo "✅ $image_name image built successfully"
     return 0
   else
@@ -103,44 +172,210 @@ build_image() {
   fi
 }
 
+pull_base_image() {
+  local image_ref=$1
+  local attempts=${2:-3}
+  local delay=${3:-5}
+
+  # If the image already exists locally, skip pulling.
+  if docker image inspect "$image_ref" >/dev/null 2>&1; then
+    echo "   Local image present; skip pull: $image_ref"
+    return 0
+  fi
+
+  for ((i=1; i<=attempts; i++)); do
+    echo "   Pulling base image ($i/$attempts): $image_ref"
+    if docker pull "$image_ref" >/dev/null; then
+      echo "   Base image ready: $image_ref"
+      return 0
+    fi
+    echo "   Pull failed: $image_ref"
+    if (( i < attempts )); then
+      echo "   Retrying in ${delay}s..."
+      sleep "$delay"
+    fi
+  done
+
+  echo "❌ Unable to pull base image after ${attempts} attempts: $image_ref"
+  return 1
+}
+
 images_built=()
 build_failed=false
 
-if build_image "Elasticsearch" "src/log/elasticsearch/build/Dockerfile" "argus-elasticsearch:latest" ""; then
+if [[ "$build_core" == true ]]; then
-  images_built+=("argus-elasticsearch:latest")
+  if build_image "Elasticsearch" "src/log/elasticsearch/build/Dockerfile" "argus-elasticsearch:latest"; then
-else
+    images_built+=("argus-elasticsearch:latest")
-  build_failed=true
+  else
+    build_failed=true
+  fi
+
+  echo ""
+
+  if build_image "Kibana" "src/log/kibana/build/Dockerfile" "argus-kibana:latest"; then
+    images_built+=("argus-kibana:latest")
+  else
+    build_failed=true
+  fi
+
+  echo ""
+
+  if build_image "BIND9" "src/bind/build/Dockerfile" "argus-bind9:latest"; then
+    images_built+=("argus-bind9:latest")
+  else
+    build_failed=true
+  fi
 fi
 
 echo ""
 
-if build_image "Kibana" "src/log/kibana/build/Dockerfile" "argus-kibana:latest" ""; then
+if [[ "$build_master" == true ]]; then
-  images_built+=("argus-kibana:latest")
+  echo ""
-else
+  echo "🔄 Building Master image..."
-  build_failed=true
-fi
-
-echo ""
-
-if build_image "BIND9" "src/bind/build/Dockerfile" "argus-bind9:latest" ""; then
-  images_built+=("argus-bind9:latest")
-else
-  build_failed=true
-fi
-
-echo ""
-
-if [[ "$build_master_offline" == true ]]; then
-  echo "🏗️  Building master offline image"
   pushd "$master_root" >/dev/null
-  if ./scripts/build_images.sh --offline --tag argus-master:offline; then
+  master_args=("--tag" "argus-master:latest")
-    images_built+=("argus-master:offline")
+  if [[ "$use_intranet" == true ]]; then
+    master_args+=("--intranet")
+  fi
+  if [[ "$build_master_offline" == true ]]; then
+    master_args+=("--offline")
+  fi
+  if [[ "$no_cache" == true ]]; then
+    master_args+=("--no-cache")
+  fi
+  if ./scripts/build_images.sh "${master_args[@]}"; then
+    if [[ "$build_master_offline" == true ]]; then
+      images_built+=("argus-master:offline")
+    else
+      images_built+=("argus-master:latest")
+    fi
   else
     build_failed=true
   fi
   popd >/dev/null
 fi
 
+if [[ "$build_metric" == true ]]; then
+  echo ""
+  echo "Building Metric module images..."
+
+  metric_base_images=(
+    "ubuntu:22.04"
+    "ubuntu/prometheus:3-24.04_stable"
+    "grafana/grafana:11.1.0"
+  )
+
+  for base_image in "${metric_base_images[@]}"; do
+    if ! pull_base_image "$base_image"; then
+      build_failed=true
+    fi
+  done
+
+  metric_builds=(
+    "Metric FTP|src/metric/ftp/build/Dockerfile|argus-metric-ftp:latest|src/metric/ftp/build"
+    "Metric Prometheus|src/metric/prometheus/build/Dockerfile|argus-metric-prometheus:latest|src/metric/prometheus/build"
+    "Metric Grafana|src/metric/grafana/build/Dockerfile|argus-metric-grafana:latest|src/metric/grafana/build"
+  )
+
+  for build_spec in "${metric_builds[@]}"; do
+    IFS='|' read -r image_label dockerfile_path image_tag build_context <<< "$build_spec"
+    if build_image "$image_label" "$dockerfile_path" "$image_tag" "$build_context"; then
+      images_built+=("$image_tag")
+    else
+      build_failed=true
+    fi
+    echo ""
+  done
+fi
+
+# =======================================
+# Sys (system tests) node images
+# =======================================
+
+if [[ "$build_sys" == true ]]; then
+  echo ""
+  echo "Building Sys node images..."
+
+  sys_base_images=(
+    "ubuntu:22.04"
+    "nvidia/cuda:12.2.2-runtime-ubuntu22.04"
+  )
+
+  for base_image in "${sys_base_images[@]}"; do
+    if ! pull_base_image "$base_image"; then
+      build_failed=true
+    fi
+  done
+
+  sys_builds=(
+    "Sys Node|src/sys/build/node/Dockerfile|argus-sys-node:latest|."
+    "Sys Metric Test Node|src/sys/build/test-node/Dockerfile|argus-sys-metric-test-node:latest|."
+    "Sys Metric Test GPU Node|src/sys/build/test-gpu-node/Dockerfile|argus-sys-metric-test-gpu-node:latest|."
+  )
+
+  for build_spec in "${sys_builds[@]}"; do
+    IFS='|' read -r image_label dockerfile_path image_tag build_context <<< "$build_spec"
+    if build_image "$image_label" "$dockerfile_path" "$image_tag" "$build_context"; then
+      images_built+=("$image_tag")
+    else
+      build_failed=true
+    fi
+    echo ""
+  done
+fi
+
+# =======================================
+# Web & Alert module images
+# =======================================
+
+if [[ "$build_web" == true || "$build_alert" == true ]]; then
+  echo ""
+  echo "Building Web and Alert module images..."
+
+  # Pre-pull commonly used base images for stability
+  web_alert_base_images=(
+    "node:20"
+    "ubuntu:24.04"
+  )
+
+  for base_image in "${web_alert_base_images[@]}"; do
+    if ! pull_base_image "$base_image"; then
+      build_failed=true
+    fi
+  done
+
+  if [[ "$build_web" == true ]]; then
+    web_builds=(
+      "Web Frontend|src/web/build_tools/frontend/Dockerfile|argus-web-frontend:latest|."
+      "Web Proxy|src/web/build_tools/proxy/Dockerfile|argus-web-proxy:latest|."
+    )
+    for build_spec in "${web_builds[@]}"; do
+      IFS='|' read -r image_label dockerfile_path image_tag build_context <<< "$build_spec"
+      if build_image "$image_label" "$dockerfile_path" "$image_tag" "$build_context"; then
+        images_built+=("$image_tag")
+      else
+        build_failed=true
+      fi
+      echo ""
+    done
+  fi
+
+  if [[ "$build_alert" == true ]]; then
+    alert_builds=(
+      "Alertmanager|src/alert/alertmanager/build/Dockerfile|argus-alertmanager:latest|."
+    )
+    for build_spec in "${alert_builds[@]}"; do
+      IFS='|' read -r image_label dockerfile_path image_tag build_context <<< "$build_spec"
+      if build_image "$image_label" "$dockerfile_path" "$image_tag" "$build_context"; then
+        images_built+=("$image_tag")
+      else
+        build_failed=true
+      fi
+      echo ""
+    done
+  fi
+fi
+
 echo "======================================="
 echo "📦 Build Summary"
 echo "======================================="
@@ -167,7 +402,6 @@ if [[ "$build_master_offline" == true ]]; then
   echo ""
   echo "🧳 Master offline wheels 已解压到 $master_offline_dir"
 fi
-
 echo ""
 echo "🚀 Next steps:"
 echo "   ./build/save_images.sh --compress          # 导出镜像"

build/save_images.sh

@@ -68,6 +68,12 @@ declare -A images=(
     ["argus-kibana:latest"]="argus-kibana-latest.tar"
     ["argus-bind9:latest"]="argus-bind9-latest.tar"
     ["argus-master:offline"]="argus-master-offline.tar"
+    ["argus-metric-ftp:latest"]="argus-metric-ftp-latest.tar"
+    ["argus-metric-prometheus:latest"]="argus-metric-prometheus-latest.tar"
+    ["argus-metric-grafana:latest"]="argus-metric-grafana-latest.tar"
+    ["argus-web-frontend:latest"]="argus-web-frontend-latest.tar"
+    ["argus-web-proxy:latest"]="argus-web-proxy-latest.tar"
+    ["argus-alertmanager:latest"]="argus-alertmanager-latest.tar"
 )
 
 # 函数：检查镜像是否存在

configs/.gitignore

@@ -0,0 +1,2 @@
+# Local overrides for build user/group settings
+build_user.local.conf

configs/build_user.conf

@@ -0,0 +1,6 @@
+# Default build-time UID/GID for Argus images
+# Override by creating configs/build_user.local.conf with the same format.
+# Syntax: KEY=VALUE, supports UID/GID only. Whitespace and lines starting with # are ignored.
+
+UID=2133
+GID=2015

doc/build-user-config.md

@@ -0,0 +1,38 @@
+# Argus 镜像构建 UID/GID 配置说明
+
+通过统一配置文件可以为 Kibana、Elasticsearch、Bind、Master 等容器指定运行账号，解决跨机器部署时 UID/GID 不一致导致的权限问题。
+
+## 配置入口
+
+- 默认配置存放在 `configs/build_user.conf`，内容示例：
+
+  ```bash
+  UID=2133
+  GID=2015
+  ```
+
+- 如果需要本地覆盖，可在 `configs/` 下新建 `build_user.local.conf`，字段与默认文件一致。该文件已列入 `.gitignore`，不会被意外提交。
+- 亦可在执行脚本前通过环境变量 `ARGUS_BUILD_UID` / `ARGUS_BUILD_GID` 强制指定值，优先级最高。
+
+## 作用范围
+
+- `build/build_images.sh` 在构建 log/bind/master 镜像时读取配置，并传递 `--build-arg ARGUS_BUILD_UID/GID`；控制台会输出当前使用的 UID/GID。
+- `src/master/scripts/build_images.sh` 同步使用配置，确保单独构建 master 镜像时行为一致。
+- 各镜像 Dockerfile 会根据传入的 UID/GID 调整容器内账号（如 `elasticsearch`、`kibana`、`bind`、`argus`），并以环境变量形式暴露运行时可见值。
+- Master 启动脚本会在执行 DNS 逻辑后，降权到配置的账号运行 `gunicorn`，确保写入 `/private/argus/**` 的文件具备正确属主。
+- Log 模块测试脚本 `01_bootstrap.sh` 会根据配置修正挂载目录属主，方便端到端测试在任意用户下运行。
+
+## 使用建议
+
+1. 初次克隆仓库后无需修改，默认 UID/GID 保持向后兼容。
+2. 如果在目标环境中使用新的账号（例如 `uid=4001,gid=4001`）：
+   - 编辑 `configs/build_user.local.conf` 填入新值；
+   - 使用新账号登录，并确保其加入宿主机的 `docker` 组；
+   - 重新执行 `build/build_images.sh` 或相关模块的构建脚本。
+3. 切换配置后建议重新运行目标模块的端到端脚本（如 `src/log/tests/scripts/01_bootstrap.sh`、`src/master/tests/scripts/00_e2e_test.sh`、`src/agent/tests/scripts/00_e2e_test.sh`），验证 `/private/argus` 下文件属主是否为期望账号。
+
+## 故障排查
+
+- **镜像构建报错 `groupmod: GID already in use`**：说明所选 GID 已存在于基础镜像，建议换用未占用的值，或在自定义基础镜像中先移除冲突。
+- **容器内运行时报写权限不足**：检查宿主机挂载目录是否已经由目标 UID/GID 创建；必要时重新执行模块的 `01_bootstrap.sh` 之类的准备脚本。
+- **仍看到旧 UID/GID**：确认脚本执行时未继承旧缓存，可运行 `ARGUS_BUILD_UID=... ARGUS_BUILD_GID=... ./build/build_images.sh` 强制覆盖。

scripts/common/build_user.sh

@@ -0,0 +1,115 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# Shared helper to load Argus build user/group configuration.
+# Usage:
+#   source "${PROJECT_ROOT}/scripts/common/build_user.sh"
+#   load_build_user
+#   echo "$ARGUS_BUILD_UID:$ARGUS_BUILD_GID"
+
+ARGUS_BUILD_UID_DEFAULT=2133
+ARGUS_BUILD_GID_DEFAULT=2015
+
+shopt -s extglob
+
+_ARGUS_BUILD_USER_LOADED="${_ARGUS_BUILD_USER_LOADED:-0}"
+
+_argus_build_user_script_dir() {
+  local dir
+  dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+  echo "$dir"
+}
+
+argus_project_root() {
+  local script_dir
+  script_dir="$(_argus_build_user_script_dir)"
+  (cd "$script_dir/../.." >/dev/null && pwd)
+}
+
+_argus_trim() {
+  local value="$1"
+  value="${value##+([[:space:]])}"
+  value="${value%%+([[:space:]])}"
+  printf '%s' "$value"
+}
+
+_argus_is_number() {
+  [[ "$1" =~ ^[0-9]+$ ]]
+}
+
+load_build_user() {
+  if [[ "$_ARGUS_BUILD_USER_LOADED" == "1" ]]; then
+    return 0
+  fi
+
+  local project_root config_files config uid gid
+  project_root="$(argus_project_root)"
+  config_files=(
+    "$project_root/configs/build_user.local.conf"
+    "$project_root/configs/build_user.conf"
+  )
+
+  uid="$ARGUS_BUILD_UID_DEFAULT"
+  gid="$ARGUS_BUILD_GID_DEFAULT"
+
+  for config in "${config_files[@]}"; do
+    if [[ -f "$config" ]]; then
+      while IFS= read -r raw_line || [[ -n "$raw_line" ]]; do
+        local line key value
+        line="${raw_line%%#*}"
+        line="$(_argus_trim "${line}")"
+        [[ -z "$line" ]] && continue
+        if [[ "$line" != *=* ]]; then
+          echo "[ARGUS build_user] Ignoring malformed line in $config: $raw_line" >&2
+          continue
+        fi
+        key="${line%%=*}"
+        value="${line#*=}"
+        key="$(_argus_trim "$key")"
+        value="$(_argus_trim "$value")"
+        case "$key" in
+          UID)
+            uid="$value"
+            ;;
+          GID)
+            gid="$value"
+            ;;
+          *)
+            echo "[ARGUS build_user] Unknown key '$key' in $config" >&2
+            ;;
+        esac
+      done < "$config"
+      break
+    fi
+  done
+
+  if [[ -n "${ARGUS_BUILD_UID:-}" ]]; then
+    uid="$ARGUS_BUILD_UID"
+  fi
+  if [[ -n "${ARGUS_BUILD_GID:-}" ]]; then
+    gid="$ARGUS_BUILD_GID"
+  fi
+
+  if ! _argus_is_number "$uid"; then
+    echo "[ARGUS build_user] Invalid UID '$uid'" >&2
+    return 1
+  fi
+  if ! _argus_is_number "$gid"; then
+    echo "[ARGUS build_user] Invalid GID '$gid'" >&2
+    return 1
+  fi
+
+  export ARGUS_BUILD_UID="$uid"
+  export ARGUS_BUILD_GID="$gid"
+  _ARGUS_BUILD_USER_LOADED=1
+}
+
+argus_build_user_args() {
+  load_build_user
+  printf '%s' "--build-arg ARGUS_BUILD_UID=${ARGUS_BUILD_UID} --build-arg ARGUS_BUILD_GID=${ARGUS_BUILD_GID}"
+}
+
+print_build_user() {
+  load_build_user
+  echo "ARGUS build user: UID=${ARGUS_BUILD_UID} GID=${ARGUS_BUILD_GID}"
+}

src/agent/.gitignore

@@ -1,3 +1,5 @@
 build/
 *.egg-info/
 __pycache__/
+
+.env

src/agent/README.md

@@ -9,7 +9,21 @@ cd src/agent
 ./scripts/build_binary.sh  # 生成 dist/argus-agent
 ```
 
-脚本会在本地创建虚拟环境、通过 `pyproject.toml` 安装依赖，并调用 PyInstaller 产出单文件可执行程序。构建过程中生成的 `build/`、`dist/` 会被 `.gitignore` 忽略。
+脚本默认会在 Docker 容器 (`python:3.11-slim-bullseye`) 内执行 PyInstaller，确保产物运行时兼容 glibc 2.31+（覆盖 2.35 环境）。构建流程注意事项：
+
+- 每次构建前会清理 `build/`、`dist/` 并在容器内重新创建虚拟环境。
+- 需要使用内网 Python 镜像时，可通过 `PIP_INDEX_URL`、`PIP_EXTRA_INDEX_URL`、`PIP_TRUSTED_HOST` 等环境变量传入，脚本会自动透传给容器。
+- 如果宿主机无法运行 Docker，可设置 `AGENT_BUILD_USE_DOCKER=0` 回退到本地构建；此时代码必须在 glibc ≤ 2.35 的机器上执行。
+
+构建结束后脚本会在 `build/compat_check/` 下解包关键动态库并输出最高 `GLIBC_x.y` 版本，便于快速核对兼容性。如果结果中缺少 `libssl.so.3` / `libcrypto.so.3`，表示系统会在目标宿主机上使用本地 OpenSSL 库，无需额外处理。
+
+例如：
+
+```bash
+strings build/compat_check/libpython*.so.1.0 | grep -Eo 'GLIBC_[0-9]+\.[0-9]+' | sort -Vu | tail -n1
+```
+
+如遇构建失败，常见原因是 Docker 不可用（请改用 `AGENT_BUILD_USE_DOCKER=0`）或无法访问 Python 包镜像（先设置上述镜像环境变量后重试）。
 
 ## 运行时配置
 
@@ -20,11 +34,23 @@ Agent 不再依赖配置文件；所有参数均由环境变量与主机名推
 | `MASTER_ENDPOINT` | 是 | N/A | Master 基础地址，可写 `http://host:3000` 或 `host:3000`（自动补全 `http://`）。 |
 | `REPORT_INTERVAL_SECONDS` | 否 | `60` | 状态上报间隔（秒）。必须为正整数。 |
 | `AGENT_HOSTNAME` | 否 | `$(hostname)` | 覆盖容器内主机名，便于测试或特殊命名需求。 |
+| `AGENT_ENV` | 否 | 来源于主机名 | 运行环境标识（如 `dev`、`prod`）。与 `AGENT_USER`、`AGENT_INSTANCE` 必须同时设置。 |
+| `AGENT_USER` | 否 | 来源于主机名 | 归属用户或团队标识。与 `AGENT_ENV`、`AGENT_INSTANCE` 必须同时设置。 |
+| `AGENT_INSTANCE` | 否 | 来源于主机名 | 实例编号或别名。与 `AGENT_ENV`、`AGENT_USER` 必须同时设置。 |
+
+主机名与元数据的解析优先级：
+
+1. 若设置 `AGENT_ENV` / `AGENT_USER` / `AGENT_INSTANCE` 且全部存在，则直接使用这些值。
+2. 否则检查历史 `node.json`（注册成功后由 Master 返回的信息），若包含 `env` / `user` / `instance` 则沿用。
+3. 若以上均不可用，则按历史约定从主机名解析 `env-user-instance` 前缀。
+4. 如果仍无法得到完整结果，Agent 启动会失败并提示需要提供上述环境变量。
+
+> 提示：在首次部署时需确保环境变量或主机名能够提供完整信息。完成注册后，Agent 会把 Master 返回的元数据写入 `node.json`，后续重启无需再次提供环境变量就能保持一致性。
 
 派生路径：
 
 - 节点信息：`/private/argus/agent/<hostname>/node.json`
-- 子模块健康目录：`/private/argus/agent/health/<hostname>/`
+- 子模块健康目录：`/private/argus/agent/<hostname>/health/`
 
 健康目录中的文件需遵循 `<模块前缀>-*.json` 命名（例如 `log-fluentbit.json`、`metric-node-exporter.json`），文件内容会原样并入上报的 `health` 字段。
 
@@ -50,4 +76,3 @@ cd src/agent/tests
 4. 清理 `tests/private/` 与临时容器网络。
 
 如需在真实环境部署，只需将 `dist/argus-agent` 连同健康目录挂载到目标主机，并按上表设置环境变量即可。
-

src/agent/app/collector.py

@@ -18,13 +18,12 @@ _HOSTNAME_PATTERN = re.compile(r"^([^-]+)-([^-]+)-([^-]+)-.*$")
 def collect_metadata(config: AgentConfig) -> Dict[str, Any]:
     """汇总节点注册需要的静态信息。"""
     hostname = config.hostname
-    env, user, instance = _parse_hostname(hostname)
     meta = {
         "hostname": hostname,
         "ip": _detect_ip_address(),
-        "env": env,
+        "env": config.environment,
-        "user": user,
+        "user": config.user,
-        "instance": instance,
+        "instance": config.instance,
         "cpu_number": _detect_cpu_count(),
         "memory_in_bytes": _detect_memory_bytes(),
         "gpu_number": _detect_gpu_count(),

src/agent/app/config.py

@@ -6,14 +6,21 @@ from dataclasses import dataclass
 from pathlib import Path
 from typing import Final
 
+from .state import load_node_state
 from .version import VERSION
+from .log import get_logger
 
 DEFAULT_REPORT_INTERVAL_SECONDS: Final[int] = 60
 
+LOGGER = get_logger("argus.agent.config")
+
 
 @dataclass(frozen=True)
 class AgentConfig:
     hostname: str
+    environment: str
+    user: str
+    instance: str
     node_file: str
     version: str
     master_endpoint: str
@@ -47,12 +54,69 @@ def _resolve_hostname() -> str:
     return os.environ.get("AGENT_HOSTNAME") or socket.gethostname()
 
 
+def _load_metadata_from_state(node_file: str) -> tuple[str, str, str] | None:
+    state = load_node_state(node_file)
+    if not state:
+        return None
+
+    meta = state.get("meta_data") or {}
+    env = meta.get("env") or state.get("env")
+    user = meta.get("user") or state.get("user")
+    instance = meta.get("instance") or state.get("instance")
+
+    if env and user and instance:
+        LOGGER.debug("Metadata resolved from node state", extra={"node_file": node_file})
+        return env, user, instance
+
+    LOGGER.warning(
+        "node.json missing metadata fields; ignoring",
+        extra={"node_file": node_file, "meta_data": meta},
+    )
+    return None
+
+
+def _resolve_metadata_fields(hostname: str, node_file: str) -> tuple[str, str, str]:
+    env = os.environ.get("AGENT_ENV")
+    user = os.environ.get("AGENT_USER")
+    instance = os.environ.get("AGENT_INSTANCE")
+
+    if env and user and instance:
+        return env, user, instance
+
+    if any([env, user, instance]):
+        LOGGER.warning(
+            "Incomplete metadata environment variables; falling back to persisted metadata",
+            extra={
+                "has_env": bool(env),
+                "has_user": bool(user),
+                "has_instance": bool(instance),
+            },
+        )
+
+    state_metadata = _load_metadata_from_state(node_file)
+    if state_metadata is not None:
+        return state_metadata
+
+    from .collector import _parse_hostname  # Local import to avoid circular dependency
+
+    env, user, instance = _parse_hostname(hostname)
+
+    if not all([env, user, instance]):
+        raise ValueError(
+            "Failed to determine metadata fields; set AGENT_ENV/USER/INSTANCE or use supported hostname pattern"
+        )
+
+    return env, user, instance
+
+
 def load_config() -> AgentConfig:
     """从环境变量推导配置，移除了外部配置文件依赖。"""
 
     hostname = _resolve_hostname()
     node_file = f"/private/argus/agent/{hostname}/node.json"
-    health_dir = f"/private/argus/agent/health/{hostname}/"
+    environment, user, instance = _resolve_metadata_fields(hostname, node_file)
+
+    health_dir = f"/private/argus/agent/{hostname}/health/"
 
     master_endpoint_env = os.environ.get("MASTER_ENDPOINT")
     if master_endpoint_env is None:
@@ -66,6 +130,9 @@ def load_config() -> AgentConfig:
 
     return AgentConfig(
         hostname=hostname,
+        environment=environment,
+        user=user,
+        instance=instance,
         node_file=node_file,
         version=VERSION,
         master_endpoint=master_endpoint,

src/agent/scripts/agent_deployment_verify.sh

@@ -0,0 +1,690 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+LOG_PREFIX="[AGENT-VERIFY]"
+MASTER_ENDPOINT_DEFAULT=""
+AGENT_DATA_ROOT_DEFAULT="/private/argus/agent"
+AGENT_ETC_ROOT_DEFAULT="/private/argus/etc"
+REPORT_INTERVAL_DEFAULT="2"
+
+ALLOW_CONFIG_TOUCH="false"
+KEEP_TEST_HEALTH="false"
+
+log_info() {
+  echo "${LOG_PREFIX} INFO  $*"
+}
+
+log_warn() {
+  echo "${LOG_PREFIX} WARN  $*" >&2
+}
+
+log_error() {
+  echo "${LOG_PREFIX} ERROR $*" >&2
+}
+
+usage() {
+  cat <<'USAGE'
+Usage: agent_deployment_verify.sh [options]
+
+Options:
+  --allow-config-touch   Enable optional config PUT dry-run check.
+  --keep-test-health     Keep the temporary verify health file after checks.
+  -h, --help             Show this help message.
+
+Environment variables:
+  MASTER_ENDPOINT        (required) Master API base endpoint, e.g. http://master:3000
+  AGENT_DATA_ROOT        (default: /private/argus/agent)
+  AGENT_ETC_ROOT         (default: /private/argus/etc)
+  VERIFY_HOSTNAME        (default: output of hostname)
+  REPORT_INTERVAL_SECONDS (default: 2) Agent report interval in seconds
+USAGE
+}
+
+while [[ $# -gt 0 ]]; do
+  case "$1" in
+    --allow-config-touch)
+      ALLOW_CONFIG_TOUCH="true"
+      shift
+      ;;
+    --keep-test-health)
+      KEEP_TEST_HEALTH="true"
+      shift
+      ;;
+    -h|--help)
+      usage
+      exit 0
+      ;;
+    *)
+      log_error "Unknown option: $1"
+      usage >&2
+      exit 2
+      ;;
+  esac
+done
+
+MASTER_ENDPOINT="${MASTER_ENDPOINT:-$MASTER_ENDPOINT_DEFAULT}"
+AGENT_DATA_ROOT="${AGENT_DATA_ROOT:-$AGENT_DATA_ROOT_DEFAULT}"
+AGENT_ETC_ROOT="${AGENT_ETC_ROOT:-$AGENT_ETC_ROOT_DEFAULT}"
+VERIFY_HOSTNAME="${VERIFY_HOSTNAME:-$(hostname)}"
+REPORT_INTERVAL_SECONDS="${REPORT_INTERVAL_SECONDS:-$REPORT_INTERVAL_DEFAULT}"
+
+if [[ -z "$MASTER_ENDPOINT" ]]; then
+  log_error "MASTER_ENDPOINT is required"
+  exit 2
+fi
+
+if ! [[ "$REPORT_INTERVAL_SECONDS" =~ ^[0-9]+$ ]] || [[ "$REPORT_INTERVAL_SECONDS" -le 0 ]]; then
+  log_warn "Invalid REPORT_INTERVAL_SECONDS='$REPORT_INTERVAL_SECONDS', fallback to $REPORT_INTERVAL_DEFAULT"
+  REPORT_INTERVAL_SECONDS="$REPORT_INTERVAL_DEFAULT"
+fi
+
+normalize_endpoint() {
+  local endpoint="$1"
+  if [[ "$endpoint" != http://* && "$endpoint" != https://* ]]; then
+    endpoint="http://$endpoint"
+  fi
+  endpoint="${endpoint%/}"
+  echo "$endpoint"
+}
+
+MASTER_BASE="$(normalize_endpoint "$MASTER_ENDPOINT")"
+
+NODE_DIR="$AGENT_DATA_ROOT/$VERIFY_HOSTNAME"
+NODE_JSON="$NODE_DIR/node.json"
+HEALTH_DIR="$NODE_DIR/health"
+DNS_CONF="$AGENT_ETC_ROOT/dns.conf"
+UPDATE_SCRIPT="$AGENT_ETC_ROOT/update-dns.sh"
+
+declare -a RESULTS_PASS=()
+declare -a RESULTS_WARN=()
+declare -a RESULTS_FAIL=()
+
+add_result() {
+  local level="$1" message="$2"
+  case "$level" in
+    PASS)
+      RESULTS_PASS+=("$message")
+      log_info "$message"
+      ;;
+    WARN)
+      RESULTS_WARN+=("$message")
+      log_warn "$message"
+      ;;
+    FAIL)
+      RESULTS_FAIL+=("$message")
+      log_error "$message"
+      ;;
+  esac
+}
+
+HAS_JQ="0"
+if command -v jq >/dev/null 2>&1; then
+  HAS_JQ="1"
+fi
+
+if ! command -v curl >/dev/null 2>&1; then
+  log_error "curl command not found; please install curl (e.g. apt-get install -y curl)"
+  exit 2
+fi
+
+if [[ "$HAS_JQ" == "0" ]] && ! command -v python3 >/dev/null 2>&1; then
+  log_error "Neither jq nor python3 is available for JSON processing"
+  exit 2
+fi
+
+CURL_OPTS=(--fail --show-error --silent --max-time 10)
+
+curl_json() {
+  local url="$1"
+  if ! curl "${CURL_OPTS[@]}" "$url"; then
+    return 1
+  fi
+}
+
+json_query() {
+  local json="$1" jq_expr="$2" py_expr="$3"
+  if [[ "$HAS_JQ" == "1" ]]; then
+    if ! output=$(printf '%s' "$json" | jq -e -r "$jq_expr" 2>/dev/null); then
+      return 1
+    fi
+    printf '%s' "$output"
+    return 0
+  fi
+
+  python3 - "$py_expr" <<'PY'
+import json
+import sys
+
+expr = sys.argv[1]
+try:
+    data = json.load(sys.stdin)
+    value = eval(expr, {}, {"data": data})
+except Exception:
+    sys.exit(1)
+if value is None:
+    sys.exit(1)
+if isinstance(value, (dict, list)):
+    print(json.dumps(value))
+else:
+    print(value)
+PY
+}
+
+json_length() {
+  local json="$1" jq_expr="$2" py_expr="$3"
+  if [[ "$HAS_JQ" == "1" ]]; then
+    if ! output=$(printf '%s' "$json" | jq -e "$jq_expr" 2>/dev/null); then
+      return 1
+    fi
+    printf '%s' "$output"
+    return 0
+  fi
+
+  python3 - "$py_expr" <<'PY'
+import json
+import sys
+
+expr = sys.argv[1]
+try:
+    data = json.load(sys.stdin)
+    value = eval(expr, {}, {"data": data})
+except Exception:
+    sys.exit(1)
+try:
+    print(len(value))
+except Exception:
+    sys.exit(1)
+PY
+}
+
+json_has_key() {
+  local json="$1" jq_expr="$2" py_expr="$3"
+  if [[ "$HAS_JQ" == "1" ]]; then
+    if printf '%s' "$json" | jq -e "$jq_expr" >/dev/null 2>&1; then
+      return 0
+    fi
+    return 1
+  fi
+
+  python3 - "$py_expr" <<'PY'
+import json
+import sys
+
+expr = sys.argv[1]
+try:
+    data = json.load(sys.stdin)
+    value = eval(expr, {}, {"data": data})
+except Exception:
+    sys.exit(1)
+if value:
+    sys.exit(0)
+sys.exit(1)
+PY
+}
+
+iso_to_epoch() {
+  local value="$1"
+  if command -v date >/dev/null 2>&1; then
+    date -d "$value" +%s 2>/dev/null && return 0
+  fi
+  if command -v python3 >/dev/null 2>&1; then
+    python3 - "$value" <<'PY'
+import sys
+from datetime import datetime
+
+value = sys.argv[1]
+if value is None or value == "":
+    sys.exit(1)
+if value.endswith('Z'):
+    value = value[:-1] + '+00:00'
+try:
+    dt = datetime.fromisoformat(value)
+except ValueError:
+    sys.exit(1)
+print(int(dt.timestamp()))
+PY
+    return $?
+  fi
+  return 1
+}
+
+validate_json_file() {
+  local path="$1"
+  if [[ "$HAS_JQ" == "1" ]]; then
+    jq empty "$path" >/dev/null 2>&1 && return 0
+    return 1
+  fi
+  if command -v python3 >/dev/null 2>&1; then
+    python3 - "$path" <<'PY'
+import json
+import sys
+path = sys.argv[1]
+with open(path, 'r', encoding='utf-8') as handle:
+    json.load(handle)
+PY
+    return $?
+  fi
+  return 0
+}
+
+ensure_directory() {
+  local dir="$1"
+  if [[ ! -d "$dir" ]]; then
+    log_warn "Creating missing directory $dir"
+    mkdir -p "$dir"
+  fi
+}
+
+TEST_HEALTH_FILE=""
+TEST_HEALTH_BACKUP=""
+TEST_HEALTH_EXISTED="false"
+
+cleanup() {
+  if [[ -n "$TEST_HEALTH_FILE" ]]; then
+    if [[ "$TEST_HEALTH_EXISTED" == "true" ]]; then
+      printf '%s' "$TEST_HEALTH_BACKUP" > "$TEST_HEALTH_FILE"
+    elif [[ "$KEEP_TEST_HEALTH" == "true" ]]; then
+      :
+    else
+      rm -f "$TEST_HEALTH_FILE"
+    fi
+  fi
+}
+
+trap cleanup EXIT
+
+log_info "Starting agent deployment verification for hostname '$VERIFY_HOSTNAME'"
+
+# 4.2 Master health checks
+health_resp=""
+if ! health_resp=$(curl "${CURL_OPTS[@]}" -w '\n%{http_code} %{time_total}' "$MASTER_BASE/healthz" 2>/tmp/agent_verify_healthz.err); then
+  error_detail=$(cat /tmp/agent_verify_healthz.err || true)
+  add_result FAIL "GET /healthz failed: $error_detail"
+else
+  http_meta=$(tail -n1 <<<"$health_resp")
+  payload=$(head -n -1 <<<"$health_resp" || true)
+  status_code=${http_meta%% *}
+  elapsed=${http_meta##* }
+  add_result PASS "GET /healthz status=$status_code elapsed=${elapsed}s payload=$payload"
+fi
+rm -f /tmp/agent_verify_healthz.err
+
+if ! readyz_resp=$(curl "${CURL_OPTS[@]}" -w '\n%{http_code} %{time_total}' "$MASTER_BASE/readyz" 2>/tmp/agent_verify_readyz.err); then
+  error_detail=$(cat /tmp/agent_verify_readyz.err || true)
+  add_result FAIL "GET /readyz failed: $error_detail"
+  readyz_payload=""
+else
+  readyz_meta=$(tail -n1 <<<"$readyz_resp")
+  readyz_payload=$(head -n -1 <<<"$readyz_resp" || true)
+  readyz_status=${readyz_meta%% *}
+  readyz_elapsed=${readyz_meta##* }
+  add_result PASS "GET /readyz status=$readyz_status elapsed=${readyz_elapsed}s"
+fi
+rm -f /tmp/agent_verify_readyz.err
+
+# 4.3 Nodes list and detail
+if ! nodes_json=$(curl_json "$MASTER_BASE/api/v1/master/nodes" 2>/tmp/agent_verify_nodes.err); then
+  error_detail=$(cat /tmp/agent_verify_nodes.err || true)
+  add_result FAIL "GET /api/v1/master/nodes failed: $error_detail"
+  nodes_json=""
+fi
+rm -f /tmp/agent_verify_nodes.err
+
+NODE_ENTRY=""
+NODE_ID=""
+NODE_IP=""
+if [[ -n "$nodes_json" ]]; then
+  if [[ "$HAS_JQ" == "1" ]]; then
+    NODE_ENTRY=$(printf '%s' "$nodes_json" | jq -e --arg name "$VERIFY_HOSTNAME" '.[] | select(.name == $name)') || NODE_ENTRY=""
+  else
+    NODE_ENTRY=$(python3 - "$VERIFY_HOSTNAME" <<'PY'
+import json
+import sys
+
+hostname = sys.argv[1]
+nodes = json.load(sys.stdin)
+for node in nodes:
+    if node.get("name") == hostname:
+        import json as _json
+        print(_json.dumps(node))
+        sys.exit(0)
+sys.exit(1)
+PY
+    ) || NODE_ENTRY=""
+  fi
+
+  if [[ -z "$NODE_ENTRY" ]]; then
+    add_result FAIL "Current node '$VERIFY_HOSTNAME' not found in master nodes list"
+  else
+    if NODE_ID=$(json_query "$NODE_ENTRY" '.id' 'data["id"]'); then
+      add_result PASS "Discovered node id '$NODE_ID' for hostname '$VERIFY_HOSTNAME'"
+    else
+      add_result FAIL "Failed to extract node id from master response"
+    fi
+  fi
+
+  if [[ -n "$NODE_ENTRY" ]] && NODE_DETAIL=$(curl_json "$MASTER_BASE/api/v1/master/nodes/$NODE_ID" 2>/tmp/agent_verify_node_detail.err); then
+    NODE_DETAIL_JSON="$NODE_DETAIL"
+    add_result PASS "Fetched node detail for $NODE_ID"
+    if NODE_IP=$(json_query "$NODE_DETAIL_JSON" '.meta_data.ip // .meta_data.host_ip // empty' 'data.get("meta_data", {}).get("ip") or data.get("meta_data", {}).get("host_ip") or ""'); then
+      if [[ -n "$NODE_IP" ]]; then
+        add_result PASS "Registered node IP=$NODE_IP"
+      else
+        add_result INFO "Node detail does not expose IP fields"
+      fi
+    fi
+  else
+    error_detail=$(cat /tmp/agent_verify_node_detail.err 2>/dev/null || true)
+    add_result FAIL "Failed to fetch node detail for $NODE_ID: $error_detail"
+    NODE_DETAIL_JSON=""
+  fi
+  rm -f /tmp/agent_verify_node_detail.err
+
+  if stats_json=$(curl_json "$MASTER_BASE/api/v1/master/nodes/statistics" 2>/tmp/agent_verify_stats.err); then
+    if total_nodes=$(json_query "$stats_json" '.total // .total_nodes' 'data.get("total") or data.get("total_nodes")'); then
+      if [[ "$total_nodes" =~ ^[0-9]+$ ]] && [[ "$total_nodes" -ge 1 ]]; then
+        add_result PASS "Statistics total=$total_nodes"
+      else
+        add_result WARN "Statistics total field not numeric: $total_nodes"
+      fi
+    else
+      add_result WARN "Unable to read total field from statistics"
+    fi
+
+    active_nodes=""
+    if [[ "$HAS_JQ" == "1" ]]; then
+      active_nodes=$(printf '%s' "$stats_json" | jq -e 'if .status_statistics then (.status_statistics[] | select(.status == "online") | .count) else empty end' 2>/dev/null | head -n1 || true)
+    elif command -v python3 >/dev/null 2>&1; then
+      active_nodes=$(printf '%s' "$stats_json" | python3 -c 'import json,sys; data=json.load(sys.stdin); print(next((row.get("count") for row in data.get("status_statistics", []) if row.get("status")=="online"), ""))' 2>/dev/null)
+    fi
+    if [[ -n "$active_nodes" ]]; then
+      add_result PASS "Online nodes reported by master: $active_nodes"
+    fi
+
+    if [[ "$HAS_JQ" == "1" ]]; then
+      node_count=$(printf '%s' "$nodes_json" | jq 'length')
+    else
+      node_count=$(json_length "$nodes_json" 'length' 'len(data)')
+    fi
+    if [[ "$total_nodes" =~ ^[0-9]+$ ]] && [[ "$node_count" =~ ^[0-9]+$ ]] && [[ "$total_nodes" -lt "$node_count" ]]; then
+      add_result WARN "Statistics total=$total_nodes less than nodes list count=$node_count"
+    fi
+  else
+    error_detail=$(cat /tmp/agent_verify_stats.err 2>/dev/null || true)
+    add_result FAIL "Failed to fetch node statistics: $error_detail"
+  fi
+  rm -f /tmp/agent_verify_stats.err
+else
+  NODE_DETAIL_JSON=""
+fi
+
+# 4.4 Agent persistence checks
+if [[ -f "$NODE_JSON" ]]; then
+  node_file_content="$(cat "$NODE_JSON")"
+  if node_id_local=$(json_query "$node_file_content" '.id' 'data["id"]'); then
+    if [[ "$NODE_ID" != "" && "$node_id_local" == "$NODE_ID" ]]; then
+      add_result PASS "node.json id matches master ($NODE_ID)"
+    else
+      add_result FAIL "node.json id '$node_id_local' differs from master id '$NODE_ID'"
+    fi
+  else
+    add_result FAIL "Unable to extract id from node.json"
+  fi
+  if node_name_local=$(json_query "$node_file_content" '.name' 'data["name"]'); then
+    if [[ "$node_name_local" == "$VERIFY_HOSTNAME" ]]; then
+      add_result PASS "node.json name matches $VERIFY_HOSTNAME"
+    else
+      add_result FAIL "node.json name '$node_name_local' differs from hostname '$VERIFY_HOSTNAME'"
+    fi
+  else
+    add_result FAIL "Unable to extract name from node.json"
+  fi
+
+  if register_time=$(json_query "$node_file_content" '.register_time' 'data.get("register_time")'); then
+    if iso_to_epoch "$register_time" >/dev/null 2>&1; then
+      add_result PASS "node.json register_time valid ISO timestamp"
+    else
+      add_result WARN "node.json register_time invalid: $register_time"
+    fi
+  else
+    add_result WARN "node.json missing register_time"
+  fi
+
+  if last_updated=$(json_query "$node_file_content" '.last_updated' 'data.get("last_updated")'); then
+    if iso_to_epoch "$last_updated" >/dev/null 2>&1; then
+      add_result PASS "node.json last_updated valid ISO timestamp"
+    else
+      add_result WARN "node.json last_updated invalid: $last_updated"
+    fi
+  else
+    add_result WARN "node.json missing last_updated"
+  fi
+else
+  add_result FAIL "node.json not found at $NODE_JSON"
+  node_file_content=""
+fi
+
+ensure_directory "$HEALTH_DIR"
+
+if [[ -d "$HEALTH_DIR" ]]; then
+  shopt -s nullglob
+  health_files=("$HEALTH_DIR"/*.json)
+  shopt -u nullglob
+  if [[ ${#health_files[@]} -eq 0 ]]; then
+    add_result WARN "Health directory $HEALTH_DIR is empty"
+  else
+    for hf in "${health_files[@]}"; do
+      base=$(basename "$hf")
+      if [[ "$base" != *-* ]]; then
+        add_result WARN "Health file $base does not follow <module>-*.json"
+        continue
+      fi
+      if ! validate_json_file "$hf" >/dev/null 2>&1; then
+        add_result WARN "Health file $base is not valid JSON"
+      fi
+    done
+  fi
+else
+  add_result WARN "Health directory $HEALTH_DIR missing"
+fi
+
+if getent hosts master.argus.com >/dev/null 2>&1; then
+  resolved_ips=$(getent hosts master.argus.com | awk '{print $1}' | xargs)
+  add_result PASS "master.argus.com resolves to $resolved_ips"
+else
+  add_result FAIL "Failed to resolve master.argus.com"
+fi
+
+# 4.5 Master-Node status consistency
+sleep_interval=$((REPORT_INTERVAL_SECONDS + 2))
+
+if [[ -n "$NODE_DETAIL_JSON" ]]; then
+  detail_pre="$NODE_DETAIL_JSON"
+else
+  detail_pre=""
+fi
+
+if [[ -z "$detail_pre" && -n "$NODE_ID" ]]; then
+  if detail_pre=$(curl_json "$MASTER_BASE/api/v1/master/nodes/$NODE_ID" 2>/tmp/agent_verify_detail_pre.err); then
+    add_result PASS "Fetched node detail pre-check"
+  else
+    error_detail=$(cat /tmp/agent_verify_detail_pre.err 2>/dev/null || true)
+    add_result FAIL "Unable to fetch node detail for status check: $error_detail"
+  fi
+  rm -f /tmp/agent_verify_detail_pre.err
+fi
+
+server_ts_pre=""
+agent_ts_pre=""
+server_ts_post=""
+agent_ts_post=""
+
+if [[ -n "$detail_pre" ]]; then
+  server_ts_pre=$(json_query "$detail_pre" '.last_report' 'data.get("last_report")' || echo "")
+  agent_ts_pre=$(json_query "$detail_pre" '.agent_last_report' 'data.get("agent_last_report")' || echo "")
+  log_info "Captured initial last_report timestamps server='$server_ts_pre' agent='$agent_ts_pre'"
+
+  sleep "$sleep_interval"
+
+  if detail_post=$(curl_json "$MASTER_BASE/api/v1/master/nodes/$NODE_ID" 2>/tmp/agent_verify_detail_post.err); then
+    server_ts_post=$(json_query "$detail_post" '.last_report' 'data.get("last_report")' || echo "")
+    agent_ts_post=$(json_query "$detail_post" '.agent_last_report' 'data.get("agent_last_report")' || echo "")
+    if [[ "$server_ts_post" != "$server_ts_pre" ]]; then
+      add_result PASS "last_report.server_timestamp advanced (pre=$server_ts_pre post=$server_ts_post)"
+    else
+      add_result FAIL "last_report.server_timestamp did not change after ${sleep_interval}s"
+    fi
+    if [[ "$agent_ts_post" != "$agent_ts_pre" ]]; then
+      add_result PASS "last_report.agent_timestamp advanced"
+    else
+      add_result FAIL "last_report.agent_timestamp did not change"
+    fi
+
+    if [[ -n "$node_file_content" ]]; then
+      if node_last_updated=$(json_query "$node_file_content" '.last_updated' 'data.get("last_updated")'); then
+        if epoch_post=$(iso_to_epoch "$server_ts_post" 2>/dev/null); then
+          if node_epoch=$(iso_to_epoch "$node_last_updated" 2>/dev/null); then
+            diff=$((epoch_post - node_epoch))
+            [[ $diff -lt 0 ]] && diff=$((-diff))
+            tolerance=$((REPORT_INTERVAL_SECONDS * 2))
+            if [[ $diff -le $tolerance ]]; then
+              add_result PASS "last_report.server_timestamp and node.json last_updated within tolerance ($diff s)"
+            else
+              add_result WARN "Timestamp gap between master ($server_ts_post) and node.json ($node_last_updated) is ${diff}s"
+            fi
+          fi
+        fi
+      fi
+    fi
+
+    NODE_DETAIL_JSON="$detail_post"
+  else
+    error_detail=$(cat /tmp/agent_verify_detail_post.err 2>/dev/null || true)
+    add_result FAIL "Failed to fetch node detail post-check: $error_detail"
+  fi
+  rm -f /tmp/agent_verify_detail_post.err
+fi
+
+# 4.6 Health simulation
+TEST_HEALTH_FILE="$HEALTH_DIR/verify-master.json"
+ensure_directory "$HEALTH_DIR"
+
+if [[ -f "$TEST_HEALTH_FILE" ]]; then
+  TEST_HEALTH_EXISTED="true"
+  TEST_HEALTH_BACKUP="$(cat "$TEST_HEALTH_FILE")"
+else
+  TEST_HEALTH_EXISTED="false"
+fi
+
+create_health_file() {
+  local message="$1"
+  cat > "$TEST_HEALTH_FILE" <<HEALTHJSON
+{"status":"ok","message":"$message"}
+HEALTHJSON
+}
+
+validate_health_in_master() {
+  local expected_message="$1"
+  local detail_json="$2"
+  local message
+  if message=$(json_query "$detail_json" '.health["verify-master"].message' 'data.get("health", {}).get("verify-master", {}).get("message")'); then
+    if [[ "$message" == "$expected_message" ]]; then
+      return 0
+    fi
+  fi
+  return 1
+}
+
+remove_health_from_master() {
+  local detail_json="$1"
+  if json_has_key "$detail_json" '(.health | has("verify-master"))' '"verify-master" in data.get("health", {})'; then
+    return 1
+  fi
+  return 0
+}
+
+health_message_one="verify $(date +%s)"
+create_health_file "$health_message_one"
+add_result PASS "Created test health file $TEST_HEALTH_FILE"
+
+sleep "$sleep_interval"
+if detail_health_one=$(curl_json "$MASTER_BASE/api/v1/master/nodes/$NODE_ID" 2>/tmp/agent_verify_health1.err); then
+  if validate_health_in_master "$health_message_one" "$detail_health_one"; then
+    add_result PASS "Master reflects verify-master health message"
+  else
+    add_result FAIL "Master health payload does not match test message"
+  fi
+else
+  error_detail=$(cat /tmp/agent_verify_health1.err 2>/dev/null || true)
+  add_result FAIL "Failed to fetch node detail during health validation: $error_detail"
+  detail_health_one=""
+fi
+rm -f /tmp/agent_verify_health1.err
+
+health_message_two="verify $(date +%s)-update"
+create_health_file "$health_message_two"
+sleep "$sleep_interval"
+if detail_health_two=$(curl_json "$MASTER_BASE/api/v1/master/nodes/$NODE_ID" 2>/tmp/agent_verify_health2.err); then
+  if validate_health_in_master "$health_message_two" "$detail_health_two"; then
+    add_result PASS "Master health updated to new message"
+  else
+    add_result FAIL "Master health message did not update"
+  fi
+else
+  error_detail=$(cat /tmp/agent_verify_health2.err 2>/dev/null || true)
+  add_result FAIL "Failed to fetch node detail after health update: $error_detail"
+  detail_health_two=""
+fi
+rm -f /tmp/agent_verify_health2.err
+
+rm -f "$TEST_HEALTH_FILE"
+sleep "$sleep_interval"
+if detail_health_three=$(curl_json "$MASTER_BASE/api/v1/master/nodes/$NODE_ID" 2>/tmp/agent_verify_health3.err); then
+  if remove_health_from_master "$detail_health_three"; then
+    add_result PASS "Master health no longer lists verify-master after removal"
+  else
+    add_result FAIL "Master health still contains verify-master after file deletion"
+  fi
+else
+  error_detail=$(cat /tmp/agent_verify_health3.err 2>/dev/null || true)
+  add_result FAIL "Failed to fetch node detail after health removal: $error_detail"
+fi
+rm -f /tmp/agent_verify_health3.err
+
+if [[ "$TEST_HEALTH_EXISTED" == "true" ]]; then
+  printf '%s' "$TEST_HEALTH_BACKUP" > "$TEST_HEALTH_FILE"
+fi
+
+# Optional config touch
+if [[ "$ALLOW_CONFIG_TOUCH" == "true" ]]; then
+  if [[ -n "$NODE_ID" ]]; then
+    payload='{"label": {"verify": "true"}}'
+    if curl "${CURL_OPTS[@]}" -X PUT -H 'Content-Type: application/json' -d "$payload" "$MASTER_BASE/api/v1/master/nodes/$NODE_ID/config" >/tmp/agent_verify_config.log 2>&1; then
+      add_result PASS "Config PUT dry-run succeeded"
+    else
+      add_result WARN "Config PUT dry-run failed: $(cat /tmp/agent_verify_config.log)"
+    fi
+    rm -f /tmp/agent_verify_config.log
+  fi
+else
+  add_result WARN "Config PUT dry-run skipped (enable with --allow-config-touch)"
+fi
+
+# Result summary
+echo
+echo "==== Verification Summary ===="
+for entry in "${RESULTS_PASS[@]}"; do
+  printf 'PASS: %s\n' "$entry"
+done
+for entry in "${RESULTS_WARN[@]}"; do
+  printf 'WARN: %s\n' "$entry"
+done
+for entry in "${RESULTS_FAIL[@]}"; do
+  printf 'FAIL: %s\n' "$entry"
+done
+
+if [[ ${#RESULTS_FAIL[@]} -gt 0 ]]; then
+  exit 1
+fi
+
+exit 0

src/agent/scripts/build_binary.sh

@@ -6,37 +6,271 @@ MODULE_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 BUILD_ROOT="$MODULE_ROOT/build"
 DIST_DIR="$MODULE_ROOT/dist"
 PYINSTALLER_BUILD="$BUILD_ROOT/pyinstaller"
+PYINSTALLER_SPEC="$PYINSTALLER_BUILD/spec"
+PYINSTALLER_WORK="$PYINSTALLER_BUILD/work"
 VENV_DIR="$BUILD_ROOT/venv"
 
-mkdir -p "$PYINSTALLER_BUILD"
+AGENT_BUILD_IMAGE="${AGENT_BUILD_IMAGE:-python:3.11-slim-bullseye}"
-mkdir -p "$DIST_DIR"
+AGENT_BUILD_USE_DOCKER="${AGENT_BUILD_USE_DOCKER:-1}"
+# 默认在容器内忽略代理以避免公司内网代理在 Docker 网络不可达导致 pip 失败（可用 0 关闭）
+AGENT_BUILD_IGNORE_PROXY="${AGENT_BUILD_IGNORE_PROXY:-1}"
+USED_DOCKER=0
 
-if [[ ! -d "$VENV_DIR" ]]; then
+run_host_build() {
-  python3 -m venv "$VENV_DIR"
+  echo "[INFO] Using host Python environment for build" >&2
+  rm -rf "$BUILD_ROOT" "$DIST_DIR"
+  mkdir -p "$PYINSTALLER_BUILD" "$DIST_DIR"
+  python3 -m venv --copies "$VENV_DIR"
+  # shellcheck disable=SC1091
+  source "$VENV_DIR/bin/activate"
+
+  pip install --upgrade pip
+  pip install .
+  pip install "pyinstaller==6.6.0"
+
+  pyinstaller \
+    --clean \
+    --onefile \
+    --name argus-agent \
+    --distpath "$DIST_DIR" \
+    --workpath "$PYINSTALLER_WORK" \
+    --specpath "$PYINSTALLER_SPEC" \
+    --add-data "$MODULE_ROOT/pyproject.toml:." \
+    "$MODULE_ROOT/entry.py"
+
+  chmod +x "$DIST_DIR/argus-agent"
+  deactivate
+}
+
+run_docker_build() {
+  if ! command -v docker >/dev/null 2>&1; then
+    echo "[ERROR] docker 命令不存在，无法在容器内构建。请安装 Docker 或设置 AGENT_BUILD_USE_DOCKER=0" >&2
+    exit 1
+  fi
+
+  USED_DOCKER=1
+  echo "[INFO] Building agent binary inside $AGENT_BUILD_IMAGE" >&2
+
+  local host_uid host_gid
+  host_uid="$(id -u)"
+  host_gid="$(id -g)"
+  docker_env=("--rm" "-v" "$MODULE_ROOT:/workspace" "-w" "/workspace" "--env" "TARGET_UID=${host_uid}" "--env" "TARGET_GID=${host_gid}")
+
+  pass_env_if_set() {
+    local var="$1"
+    local value="${!var:-}"
+    if [[ -n "$value" ]]; then
+      docker_env+=("--env" "$var=$value")
+    fi
+  }
+
+  pass_env_if_set PIP_INDEX_URL
+  pass_env_if_set PIP_EXTRA_INDEX_URL
+  pass_env_if_set PIP_TRUSTED_HOST
+  pass_env_if_set HTTP_PROXY
+  pass_env_if_set HTTPS_PROXY
+  pass_env_if_set NO_PROXY
+  pass_env_if_set http_proxy
+  pass_env_if_set https_proxy
+  pass_env_if_set no_proxy
+  pass_env_if_set AGENT_BUILD_IGNORE_PROXY
+
+build_script=$(cat <<'INNER'
+set -euo pipefail
+cd /workspace
+apt-get update >/dev/null
+apt-get install -y --no-install-recommends binutils >/dev/null
+rm -rf /var/lib/apt/lists/*
+rm -rf build dist
+mkdir -p build/pyinstaller dist
+python3 -m venv --copies build/venv
+source build/venv/bin/activate
+# 若指定忽略代理，则清空常见代理与 pip 镜像环境变量，避免容器内代理不可达
+if [ "${AGENT_BUILD_IGNORE_PROXY:-1}" = "1" ]; then
+  unset http_proxy https_proxy HTTP_PROXY HTTPS_PROXY PIP_INDEX_URL PIP_EXTRA_INDEX_URL PIP_TRUSTED_HOST
 fi
-
-# shellcheck disable=SC1091
-source "$VENV_DIR/bin/activate"
-
 pip install --upgrade pip
-pip install "$MODULE_ROOT"
+pip install .
-pip install "pyinstaller==6.6.0"
+pip install pyinstaller==6.6.0
-
-rm -rf "$PYINSTALLER_BUILD"/*
-rm -f "$DIST_DIR/argus-agent"
-
 pyinstaller \
   --clean \
   --onefile \
   --name argus-agent \
-  --distpath "$DIST_DIR" \
+  --distpath dist \
-  --workpath "$PYINSTALLER_BUILD/work" \
+  --workpath build/pyinstaller/work \
-  --specpath "$PYINSTALLER_BUILD/spec" \
+  --specpath build/pyinstaller/spec \
-  --add-data "$MODULE_ROOT/pyproject.toml:." \
+  --add-data /workspace/pyproject.toml:. \
-  "$MODULE_ROOT/entry.py"
+  entry.py
+chmod +x dist/argus-agent
 
-chmod +x "$DIST_DIR/argus-agent"
+TARGET_UID="${TARGET_UID:-0}"
+TARGET_GID="${TARGET_GID:-0}"
+chown -R "$TARGET_UID:$TARGET_GID" dist build 2>/dev/null || true
+
+python3 - <<'PY'
+from pathlib import Path
+from PyInstaller.archive.readers import CArchiveReader
+import sys
+
+archive = Path('dist/argus-agent')
+out_dir = Path('build/compat_check')
+out_dir.mkdir(parents=True, exist_ok=True)
+
+major, minor = sys.version_info[:2]
+libpython = f'libpython{major}.{minor}.so.1.0'
+expected_libs = [
+    libpython,
+    'libssl.so.3',
+    'libcrypto.so.3',
+]
+reader = CArchiveReader(str(archive))
+extracted = []
+missing = []
+for name in expected_libs:
+    try:
+        data = reader.extract(name)
+    except KeyError:
+        missing.append(name)
+        continue
+    (out_dir / name).write_bytes(data)
+    extracted.append(name)
+(out_dir / 'manifest').write_text('\n'.join(extracted))
+if extracted:
+    print('[INFO] Extracted libraries: ' + ', '.join(extracted))
+if missing:
+    print('[WARN] Missing expected libraries in bundle: ' + ', '.join(missing))
+PY
+
+compat_check() {
+  local lib_path="$1"
+  if [[ ! -f "$lib_path" ]]; then
+    echo "[WARN] Missing $lib_path for GLIBC check"
+    return
+  fi
+  local max_glibc
+  max_glibc=$(strings -a "$lib_path" | grep -Eo 'GLIBC_[0-9]+\.[0-9]+' | sort -Vu | tail -n 1 || true)
+  if [[ -n "$max_glibc" ]]; then
+    echo "[INFO] $lib_path references up to $max_glibc"
+  else
+    echo "[INFO] $lib_path does not expose GLIBC version strings"
+  fi
+}
+
+compat_libs=()
+if [[ -f build/compat_check/manifest ]]; then
+  mapfile -t compat_libs < build/compat_check/manifest
+fi
+
+if [[ ${#compat_libs[@]} -eq 0 ]]; then
+  echo "[WARN] No libraries captured for GLIBC inspection"
+else
+  for lib in "${compat_libs[@]}"; do
+    compat_check "build/compat_check/$lib"
+  done
+fi
 
 deactivate
+INNER
+  )
+
+  if ! docker run "${docker_env[@]}" "$AGENT_BUILD_IMAGE" bash -lc "$build_script"; then
+    echo "[ERROR] Docker 构建失败，请检查 Docker 权限或设置 AGENT_BUILD_USE_DOCKER=0 在兼容主机上构建" >&2
+    exit 1
+  fi
+}
+
+if [[ "$AGENT_BUILD_USE_DOCKER" == "1" ]]; then
+  run_docker_build
+else
+  run_host_build
+fi
+
+if [[ ! -f "$DIST_DIR/argus-agent" ]]; then
+  echo "[ERROR] Agent binary was not produced" >&2
+  exit 1
+fi
+
+if [[ "$USED_DOCKER" != "1" ]]; then
+  if [[ ! -x "$VENV_DIR/bin/python" ]]; then
+    echo "[WARN] PyInstaller virtualenv missing at $VENV_DIR; skipping compatibility check" >&2
+  else
+    COMPAT_DIR="$BUILD_ROOT/compat_check"
+    rm -rf "$COMPAT_DIR"
+    mkdir -p "$COMPAT_DIR"
+
+    EXTRACT_SCRIPT=$(cat <<'PY'
+from pathlib import Path
+from PyInstaller.archive.readers import CArchiveReader
+import sys
+
+archive = Path('dist/argus-agent')
+out_dir = Path('build/compat_check')
+out_dir.mkdir(parents=True, exist_ok=True)
+
+major, minor = sys.version_info[:2]
+libpython = f'libpython{major}.{minor}.so.1.0'
+expected_libs = [
+    libpython,
+    'libssl.so.3',
+    'libcrypto.so.3',
+]
+reader = CArchiveReader(str(archive))
+extracted = []
+missing = []
+for name in expected_libs:
+    try:
+        data = reader.extract(name)
+    except KeyError:
+        missing.append(name)
+        continue
+    (out_dir / name).write_bytes(data)
+    extracted.append(name)
+(out_dir / 'manifest').write_text('\n'.join(extracted))
+if extracted:
+    print('[INFO] Extracted libraries: ' + ', '.join(extracted))
+if missing:
+    print('[WARN] Missing expected libraries in bundle: ' + ', '.join(missing))
+PY
+)
+
+    "$VENV_DIR/bin/python" - <<PY
+$EXTRACT_SCRIPT
+PY
+
+    compat_libs=()
+    if [[ -f "$COMPAT_DIR/manifest" ]]; then
+      mapfile -t compat_libs < "$COMPAT_DIR/manifest"
+    fi
+
+    check_glibc_version() {
+      local lib_path="$1"
+      if [[ ! -f "$lib_path" ]]; then
+        echo "[WARN] Skipping GLIBC check; file not found: $lib_path" >&2
+        return
+      fi
+      if command -v strings >/dev/null 2>&1; then
+        local max_glibc
+        max_glibc=$(strings -a "$lib_path" | grep -Eo 'GLIBC_[0-9]+\.[0-9]+' | sort -Vu | tail -n 1 || true)
+        if [[ -n "$max_glibc" ]]; then
+          echo "[INFO] $lib_path references up to $max_glibc"
+        else
+          echo "[INFO] $lib_path does not expose GLIBC version strings"
+        fi
+      else
+        echo "[WARN] strings command unavailable; cannot inspect $lib_path" >&2
+      fi
+    }
+
+    if [[ ${#compat_libs[@]} -eq 0 ]]; then
+      echo "[WARN] No libraries captured for GLIBC inspection" >&2
+    else
+      for lib in "${compat_libs[@]}"; do
+        check_glibc_version "$COMPAT_DIR/$lib"
+      done
+    fi
+  fi
+else
+  echo "[INFO] Compatibility check executed inside container"
+fi
 
 echo "[INFO] Agent binary generated at $DIST_DIR/argus-agent"

src/agent/tests/docker-compose.yml

@@ -1,16 +1,19 @@
 services:
   bind:
-    image: ${BIND_IMAGE_TAG:-argus-bind9:e2e}
+    image: ${BIND_IMAGE_TAG:-argus-bind9:latest}
     container_name: argus-bind-agent-e2e
     volumes:
       - ./private:/private
-    restart: unless-stopped
     networks:
       default:
         ipv4_address: 172.28.0.2
+    environment:
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
+    restart: always
 
   master:
-    image: argus-master:dev
+    image: argus-master:latest
     container_name: argus-master-agent-e2e
     depends_on:
       - bind
@@ -18,6 +21,8 @@ services:
       - OFFLINE_THRESHOLD_SECONDS=6
       - ONLINE_THRESHOLD_SECONDS=2
       - SCHEDULER_INTERVAL_SECONDS=1
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
     ports:
       - "32300:3000"
     volumes:
@@ -27,9 +32,10 @@ services:
     networks:
       default:
         ipv4_address: 172.28.0.10
+    restart: always
 
   agent:
-    image: ubuntu:24.04
+    image: ubuntu:22.04
     container_name: argus-agent-e2e
     hostname: dev-e2euser-e2einst-pod-0
     depends_on:
@@ -38,17 +44,51 @@ services:
     environment:
       - MASTER_ENDPOINT=http://master.argus.com:3000
       - REPORT_INTERVAL_SECONDS=2
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
     volumes:
       - ./private/argus/agent/dev-e2euser-e2einst-pod-0:/private/argus/agent/dev-e2euser-e2einst-pod-0
-      - ./private/argus/agent/health/dev-e2euser-e2einst-pod-0:/private/argus/agent/health/dev-e2euser-e2einst-pod-0
+      - ./private/argus/agent/dev-e2euser-e2einst-pod-0/health:/private/argus/agent/dev-e2euser-e2einst-pod-0/health
       - ./private/argus/etc:/private/argus/etc
       - ../dist/argus-agent:/usr/local/bin/argus-agent:ro
       - ./scripts/agent_entrypoint.sh:/usr/local/bin/agent-entrypoint.sh:ro
+      - ../scripts/agent_deployment_verify.sh:/usr/local/bin/agent_deployment_verify.sh:ro
     entrypoint:
       - /usr/local/bin/agent-entrypoint.sh
     networks:
       default:
         ipv4_address: 172.28.0.20
+    restart: always
+
+  agent_env:
+    image: ubuntu:22.04
+    container_name: argus-agent-env-e2e
+    hostname: host_abc
+    depends_on:
+      - master
+      - bind
+    environment:
+      - MASTER_ENDPOINT=http://master.argus.com:3000
+      - REPORT_INTERVAL_SECONDS=2
+      - AGENT_ENV=prod
+      - AGENT_USER=ml
+      - AGENT_INSTANCE=node-3
+      - AGENT_HOSTNAME=host_abc
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
+    volumes:
+      - ./private/argus/agent/host_abc:/private/argus/agent/host_abc
+      - ./private/argus/agent/host_abc/health:/private/argus/agent/host_abc/health
+      - ./private/argus/etc:/private/argus/etc
+      - ../dist/argus-agent:/usr/local/bin/argus-agent:ro
+      - ./scripts/agent_entrypoint.sh:/usr/local/bin/agent-entrypoint.sh:ro
+      - ../scripts/agent_deployment_verify.sh:/usr/local/bin/agent_deployment_verify.sh:ro
+    entrypoint:
+      - /usr/local/bin/agent-entrypoint.sh
+    networks:
+      default:
+        ipv4_address: 172.28.0.21
+    restart: always
 
 networks:
   default:

src/agent/tests/scripts/00_e2e_test.sh

@@ -7,9 +7,10 @@ SCRIPTS=(
   "02_up.sh"
   "03_wait_and_assert_registration.sh"
   "04_write_health_files.sh"
-  "05_assert_status_on_master.sh"
+  "05_verify_agent.sh"
-  "06_restart_agent_and_reregister.sh"
+  "06_assert_status_on_master.sh"
-  "07_down.sh"
+  "07_restart_agent_and_reregister.sh"
+  "08_down.sh"
 )
 
 for script in "${SCRIPTS[@]}"; do

src/agent/tests/scripts/01_bootstrap.sh

@@ -11,13 +11,21 @@ TMP_ROOT="$TEST_ROOT/tmp"
 
 AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
 AGENT_CONFIG_DIR="$PRIVATE_ROOT/argus/agent/$AGENT_HOSTNAME"
-AGENT_HEALTH_DIR="$PRIVATE_ROOT/argus/agent/health/$AGENT_HOSTNAME"
+AGENT_HEALTH_DIR="$PRIVATE_ROOT/argus/agent/$AGENT_HOSTNAME/health"
 MASTER_PRIVATE_DIR="$PRIVATE_ROOT/argus/master"
 METRIC_PRIVATE_DIR="$PRIVATE_ROOT/argus/metric/prometheus"
 DNS_DIR="$PRIVATE_ROOT/argus/etc"
-BIND_IMAGE_TAG="${BIND_IMAGE_TAG:-argus-bind9:e2e}"
+BIND_IMAGE_TAG="${BIND_IMAGE_TAG:-argus-bind9:latest}"
 BIND_ROOT="$(cd "$MASTER_ROOT/../bind" && pwd)"
 
+ensure_image() {
+  local image="$1"
+  if ! docker image inspect "$image" >/dev/null 2>&1; then
+    echo "[ERROR] Docker image '$image' 未找到，请先运行统一构建脚本 (例如 ./build/build_images.sh) 生成所需镜像" >&2
+    exit 1
+  fi
+}
+
 mkdir -p "$AGENT_CONFIG_DIR"
 mkdir -p "$AGENT_HEALTH_DIR"
 mkdir -p "$MASTER_PRIVATE_DIR"
@@ -35,9 +43,8 @@ else
   echo "[WARN] bind update script missing at $BIND_ROOT/build/update-dns.sh"
 fi
 
-pushd "$MASTER_ROOT" >/dev/null
+ensure_image "argus-master:latest"
-./scripts/build_images.sh --tag argus-master:dev
+ensure_image "$BIND_IMAGE_TAG"
-popd >/dev/null
 
 AGENT_BINARY="$AGENT_ROOT/dist/argus-agent"
 
@@ -50,11 +57,6 @@ if [[ ! -x "$AGENT_BINARY" ]]; then
   exit 1
 fi
 
-# 中文提示：构建测试专用 bind9 镜像，确保解析服务可用
-pushd "$REPO_ROOT" >/dev/null
-docker build -f src/bind/build/Dockerfile -t "$BIND_IMAGE_TAG" .
-popd >/dev/null
-
 echo "$AGENT_BINARY" > "$TMP_ROOT/agent_binary_path"
 echo "$BIND_IMAGE_TAG" > "$TMP_ROOT/bind_image_tag"
 

src/agent/tests/scripts/02_up.sh

@@ -3,8 +3,19 @@ set -euo pipefail
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+REPO_ROOT="$(cd "$TEST_ROOT/../../.." && pwd)"
 
 TMP_ROOT="$TEST_ROOT/tmp"
+ENV_FILE="$TEST_ROOT/.env"
+
+source "$REPO_ROOT/scripts/common/build_user.sh"
+load_build_user
+export ARGUS_BUILD_UID ARGUS_BUILD_GID
+
+cat > "$ENV_FILE" <<EOF
+ARGUS_BUILD_UID=$ARGUS_BUILD_UID
+ARGUS_BUILD_GID=$ARGUS_BUILD_GID
+EOF
 
 if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
   echo "[ERROR] Agent binary path missing; run 01_bootstrap.sh first" >&2
@@ -17,7 +28,7 @@ if [[ ! -x "$AGENT_BINARY" ]]; then
   exit 1
 fi
 
-BIND_IMAGE_TAG_VALUE="argus-bind9:e2e"
+BIND_IMAGE_TAG_VALUE="argus-bind9:latest"
 if [[ -f "$TMP_ROOT/bind_image_tag" ]]; then
   BIND_IMAGE_TAG_VALUE="$(cat "$TMP_ROOT/bind_image_tag")"
 fi
@@ -30,7 +41,7 @@ compose() {
   fi
 }
 
-docker container rm -f argus-agent-e2e argus-master-agent-e2e argus-bind-agent-e2e >/dev/null 2>&1 || true
+docker container rm -f argus-agent-e2e argus-agent-env-e2e argus-master-agent-e2e argus-bind-agent-e2e >/dev/null 2>&1 || true
 
 docker network rm tests_default >/dev/null 2>&1 || true
 

src/agent/tests/scripts/03_wait_and_assert_registration.sh

@@ -6,11 +6,14 @@ TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 TMP_ROOT="$TEST_ROOT/tmp"
 API_BASE="http://localhost:32300/api/v1/master"
 AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
+ENV_AGENT_HOSTNAME="host_abc"
 NODE_FILE="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME/node.json"
+ENV_NODE_FILE="$TEST_ROOT/private/argus/agent/$ENV_AGENT_HOSTNAME/node.json"
 
 mkdir -p "$TMP_ROOT"
 
-node_id=""
+primary_node_id=""
+env_node_id=""
 for _ in {1..30}; do
   sleep 2
   response=$(curl -sS "$API_BASE/nodes" || true)
@@ -19,24 +22,49 @@ for _ in {1..30}; do
   fi
   list_file="$TMP_ROOT/nodes_list.json"
   echo "$response" > "$list_file"
-  node_id=$(python3 - "$list_file" <<'PY'
+  readarray -t node_ids < <(python3 - "$list_file" "$AGENT_HOSTNAME" "$ENV_AGENT_HOSTNAME" <<'PY'
 import json, sys
+
 with open(sys.argv[1]) as handle:
     nodes = json.load(handle)
-print(nodes[0]["id"] if nodes else "")
+
+target_primary = sys.argv[2]
+target_env = sys.argv[3]
+
+primary_id = ""
+env_id = ""
+
+for node in nodes:
+    if node.get("name") == target_primary:
+        primary_id = node.get("id", "")
+    if node.get("name") == target_env:
+        env_id = node.get("id", "")
+
+print(primary_id)
+print(env_id)
 PY
-)
+  )
-  if [[ -n "$node_id" ]]; then
+
+  primary_node_id="${node_ids[0]}"
+  env_node_id="${node_ids[1]}"
+
+  if [[ -n "$primary_node_id" && -n "$env_node_id" ]]; then
     break
   fi
  done
 
-if [[ -z "$node_id" ]]; then
+if [[ -z "$primary_node_id" ]]; then
-  echo "[ERROR] Agent did not register within timeout" >&2
+  echo "[ERROR] Primary agent did not register within timeout" >&2
   exit 1
 fi
 
-echo "$node_id" > "$TMP_ROOT/node_id"
+if [[ -z "$env_node_id" ]]; then
+  echo "[ERROR] Env-variable agent did not register within timeout" >&2
+  exit 1
+fi
+
+echo "$primary_node_id" > "$TMP_ROOT/node_id"
+echo "$env_node_id" > "$TMP_ROOT/node_id_host_abc"
 
 if [[ ! -f "$NODE_FILE" ]]; then
   echo "[ERROR] node.json not created at $NODE_FILE" >&2
@@ -50,8 +78,20 @@ with open(sys.argv[1]) as handle:
 assert "id" in node and node["id"], "node.json missing id"
 PY
 
+if [[ ! -f "$ENV_NODE_FILE" ]]; then
+  echo "[ERROR] node.json not created at $ENV_NODE_FILE" >&2
+  exit 1
+fi
+
+python3 - "$ENV_NODE_FILE" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+assert "id" in node and node["id"], "env agent node.json missing id"
+PY
+
 detail_file="$TMP_ROOT/initial_detail.json"
-curl -sS "$API_BASE/nodes/$node_id" -o "$detail_file"
+curl -sS "$API_BASE/nodes/$primary_node_id" -o "$detail_file"
 python3 - "$detail_file" "$TMP_ROOT/initial_ip" <<'PY'
 import json, sys, pathlib
 with open(sys.argv[1]) as handle:
@@ -62,4 +102,5 @@ if not ip:
 pathlib.Path(sys.argv[2]).write_text(ip)
 PY
 
-echo "[INFO] Agent registered with node id $node_id"
+echo "[INFO] Agent registered with node id $primary_node_id"
+echo "[INFO] Env-variable agent registered with node id $env_node_id"

src/agent/tests/scripts/04_write_health_files.sh

@@ -3,7 +3,7 @@ set -euo pipefail
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
-HEALTH_DIR="$TEST_ROOT/private/argus/agent/health/dev-e2euser-e2einst-pod-0"
+HEALTH_DIR="$TEST_ROOT/private/argus/agent/dev-e2euser-e2einst-pod-0/health"
 
 cat > "$HEALTH_DIR/log-fluentbit.json" <<JSON
 {

src/agent/tests/scripts/05_verify_agent.sh

@@ -0,0 +1,60 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+REPO_ROOT="$(cd "$TEST_ROOT/.." && pwd)"
+VERIFY_SCRIPT="$REPO_ROOT/scripts/agent_deployment_verify.sh"
+ENV_NODE_ID_FILE="$TEST_ROOT/tmp/node_id_host_abc"
+PRIMARY_CONTAINER="argus-agent-e2e"
+ENV_CONTAINER="argus-agent-env-e2e"
+PRIMARY_HOSTNAME="dev-e2euser-e2einst-pod-0"
+ENV_HOSTNAME="host_abc"
+
+if ! docker ps --format '{{.Names}}' | grep -q "^${PRIMARY_CONTAINER}$"; then
+  echo "[WARN] agent container not running; skip verification"
+  exit 0
+fi
+
+if docker exec -i "$PRIMARY_CONTAINER" bash -lc 'command -v curl >/dev/null && command -v jq >/dev/null'; then
+  echo "[INFO] curl/jq already installed in agent container"
+else
+  echo "[INFO] Installing curl/jq in agent container"
+  docker exec -i "$PRIMARY_CONTAINER" bash -lc 'apt-get update >/dev/null 2>&1 && apt-get install -y curl jq >/dev/null 2>&1' || true
+fi
+
+if [[ ! -f "$VERIFY_SCRIPT" ]]; then
+  echo "[ERROR] Verification script missing at $VERIFY_SCRIPT" >&2
+  exit 1
+fi
+
+run_verifier() {
+  local container="$1" hostname="$2"
+
+  if ! docker ps --format '{{.Names}}' | grep -q "^${container}$"; then
+    echo "[WARN] container $container not running; skip"
+    return
+  fi
+
+  if ! docker exec -i "$container" bash -lc 'command -v /usr/local/bin/agent_deployment_verify.sh >/dev/null'; then
+    echo "[ERROR] /usr/local/bin/agent_deployment_verify.sh missing in $container" >&2
+    exit 1
+  fi
+
+  echo "[INFO] Running verification for $hostname in $container"
+  docker exec -i "$container" env VERIFY_HOSTNAME="$hostname" /usr/local/bin/agent_deployment_verify.sh
+}
+
+run_verifier "$PRIMARY_CONTAINER" "$PRIMARY_HOSTNAME"
+
+if docker ps --format '{{.Names}}' | grep -q "^${ENV_CONTAINER}$"; then
+  if docker exec -i "$ENV_CONTAINER" bash -lc 'command -v curl >/dev/null && command -v jq >/dev/null'; then
+    echo "[INFO] curl/jq already installed in env agent container"
+  else
+    echo "[INFO] Installing curl/jq in env agent container"
+    docker exec -i "$ENV_CONTAINER" bash -lc 'apt-get update >/dev/null 2>&1 && apt-get install -y curl jq >/dev/null 2>&1' || true
+  fi
+  run_verifier "$ENV_CONTAINER" "$ENV_HOSTNAME"
+else
+  echo "[WARN] env-driven agent container not running; skip secondary verification"
+fi

src/agent/tests/scripts/06_assert_status_on_master.sh

@@ -6,6 +6,8 @@ TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 TMP_ROOT="$TEST_ROOT/tmp"
 API_BASE="http://localhost:32300/api/v1/master"
 NODE_ID="$(cat "$TMP_ROOT/node_id")"
+ENV_NODE_ID="$(cat "$TMP_ROOT/node_id_host_abc")"
+ENV_HOSTNAME="host_abc"
 NODES_JSON="$TEST_ROOT/private/argus/metric/prometheus/nodes.json"
 
 success=false
@@ -41,13 +43,36 @@ if [[ ! -f "$NODES_JSON" ]]; then
   exit 1
 fi
 
-python3 - "$NODES_JSON" <<'PY'
+python3 - "$NODES_JSON" "$NODE_ID" "$ENV_NODE_ID" <<'PY'
 import json, sys
 with open(sys.argv[1]) as handle:
     nodes = json.load(handle)
-assert len(nodes) == 1, nodes
+
-entry = nodes[0]
+expected_primary = sys.argv[2]
-assert entry["node_id"], entry
+expected_env = sys.argv[3]
+
+ids = {entry.get("node_id") for entry in nodes}
+assert expected_primary in ids, nodes
+assert expected_env in ids, nodes
+assert len(nodes) >= 2, nodes
 PY
 
 echo "[INFO] Master reflects agent health and nodes.json entries"
+
+env_detail_file="$TMP_ROOT/env_agent_detail.json"
+curl -sS "$API_BASE/nodes/$ENV_NODE_ID" -o "$env_detail_file"
+python3 - "$env_detail_file" "$ENV_HOSTNAME" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+
+expected_name = sys.argv[2]
+
+assert node.get("name") == expected_name, node
+meta = node.get("meta_data", {})
+assert meta.get("env") == "prod", meta
+assert meta.get("user") == "ml", meta
+assert meta.get("instance") == "node-3", meta
+PY
+
+echo "[INFO] Env-variable agent reports expected metadata"

src/agent/tests/scripts/06_restart_agent_and_reregister.sh

@@ -1,125 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
-TMP_ROOT="$TEST_ROOT/tmp"
-API_BASE="http://localhost:32300/api/v1/master"
-NODE_ID="$(cat "$TMP_ROOT/node_id")"
-AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
-NETWORK_NAME="tests_default"
-NEW_AGENT_IP="172.28.0.200"
-ENTRYPOINT_SCRIPT="$SCRIPT_DIR/agent_entrypoint.sh"
-
-# 中文提示：重启场景也需要同样的入口脚本，确保 DNS 注册逻辑一致
-if [[ ! -f "$ENTRYPOINT_SCRIPT" ]]; then
-  echo "[ERROR] agent entrypoint script missing at $ENTRYPOINT_SCRIPT" >&2
-  exit 1
-fi
-
-if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
-  echo "[ERROR] Agent binary path missing; rerun bootstrap" >&2
-  exit 1
-fi
-
-AGENT_BINARY="$(cat "$TMP_ROOT/agent_binary_path")"
-if [[ ! -x "$AGENT_BINARY" ]]; then
-  echo "[ERROR] Agent binary not executable: $AGENT_BINARY" >&2
-  exit 1
-fi
-
-compose() {
-  if docker compose version >/dev/null 2>&1; then
-    docker compose "$@"
-  else
-    docker-compose "$@"
-  fi
-}
-
-before_file="$TMP_ROOT/before_restart.json"
-curl -sS "$API_BASE/nodes/$NODE_ID" -o "$before_file"
-prev_last_updated=$(python3 - "$before_file" <<'PY'
-import json, sys
-with open(sys.argv[1]) as handle:
-    node = json.load(handle)
-print(node.get("last_updated", ""))
-PY
-)
-prev_ip=$(python3 - "$before_file" <<'PY'
-import json, sys
-with open(sys.argv[1]) as handle:
-    node = json.load(handle)
-print(node["meta_data"].get("ip", ""))
-PY
-)
-initial_ip=$(cat "$TMP_ROOT/initial_ip")
-if [[ "$prev_ip" != "$initial_ip" ]]; then
-  echo "[ERROR] Expected initial IP $initial_ip, got $prev_ip" >&2
-  exit 1
-fi
-
-pushd "$TEST_ROOT" >/dev/null
-compose rm -sf agent
-popd >/dev/null
-
-docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
-
-AGENT_DIR="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME"
-HEALTH_DIR="$TEST_ROOT/private/argus/agent/health/$AGENT_HOSTNAME"
-
-# 先以 sleep 方式启动容器，确保我们掌握注册时的网络状态
-if ! docker run -d \
-  --name argus-agent-e2e \
-  --hostname "$AGENT_HOSTNAME" \
-  --network "$NETWORK_NAME" \
-  --ip "$NEW_AGENT_IP" \
-  -v "$AGENT_DIR:/private/argus/agent/$AGENT_HOSTNAME" \
-  -v "$HEALTH_DIR:/private/argus/agent/health/$AGENT_HOSTNAME" \
-  -v "$TEST_ROOT/private/argus/etc:/private/argus/etc" \
-  -v "$AGENT_BINARY:/usr/local/bin/argus-agent:ro" \
-  -v "$ENTRYPOINT_SCRIPT:/usr/local/bin/agent-entrypoint.sh:ro" \
-  -e MASTER_ENDPOINT=http://master.argus.com:3000 \
-  -e REPORT_INTERVAL_SECONDS=2 \
-  --entrypoint /usr/local/bin/agent-entrypoint.sh \
-  ubuntu:24.04 >/dev/null; then
-  echo "[ERROR] Failed to start agent container with custom IP" >&2
-  exit 1
-fi
-
-success=false
-detail_file="$TMP_ROOT/post_restart.json"
-for _ in {1..20}; do
-  sleep 3
-  if ! curl -sS "$API_BASE/nodes/$NODE_ID" -o "$detail_file"; then
-    continue
-  fi
-  if python3 - "$detail_file" "$prev_last_updated" "$NODE_ID" "$prev_ip" "$NEW_AGENT_IP" <<'PY'
-import json, sys
-with open(sys.argv[1]) as handle:
-    node = json.load(handle)
-prev_last_updated = sys.argv[2]
-expected_id = sys.argv[3]
-old_ip = sys.argv[4]
-expected_ip = sys.argv[5]
-last_updated = node.get("last_updated")
-current_ip = node["meta_data"].get("ip")
-assert node["id"] == expected_id
-if current_ip != expected_ip:
-    raise SystemExit(1)
-if current_ip == old_ip:
-    raise SystemExit(1)
-if not last_updated or last_updated == prev_last_updated:
-    raise SystemExit(1)
-PY
-  then
-    success=true
-    break
-  fi
-done
-
-if [[ "$success" != true ]]; then
-  echo "[ERROR] Agent did not report expected new IP $NEW_AGENT_IP after restart" >&2
-  exit 1
-fi
-
-echo "[INFO] Agent restart produced successful re-registration with IP change"

src/agent/tests/scripts/07_restart_agent_and_reregister.sh

@@ -0,0 +1,254 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+TMP_ROOT="$TEST_ROOT/tmp"
+API_BASE="http://localhost:32300/api/v1/master"
+NODE_ID="$(cat "$TMP_ROOT/node_id")"
+ENV_NODE_ID_FILE="$TMP_ROOT/node_id_host_abc"
+if [[ ! -f "$ENV_NODE_ID_FILE" ]]; then
+  echo "[ERROR] Env agent node id file missing at $ENV_NODE_ID_FILE" >&2
+  exit 1
+fi
+
+ENV_NODE_ID="$(cat "$ENV_NODE_ID_FILE")"
+AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
+ENV_AGENT_HOSTNAME="host_abc"
+NETWORK_NAME="tests_default"
+NEW_AGENT_IP="172.28.0.200"
+NEW_ENV_AGENT_IP="172.28.0.210"
+ENTRYPOINT_SCRIPT="$SCRIPT_DIR/agent_entrypoint.sh"
+VERIFY_SCRIPT="$TEST_ROOT/../scripts/agent_deployment_verify.sh"
+ENV_FILE="$TEST_ROOT/.env"
+
+# 中文提示：重启场景也需要同样的入口脚本，确保 DNS 注册逻辑一致
+if [[ ! -f "$ENTRYPOINT_SCRIPT" ]]; then
+  echo "[ERROR] agent entrypoint script missing at $ENTRYPOINT_SCRIPT" >&2
+  exit 1
+fi
+
+if [[ ! -f "$VERIFY_SCRIPT" ]]; then
+  echo "[ERROR] agent verification script missing at $VERIFY_SCRIPT" >&2
+  exit 1
+fi
+
+if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
+  echo "[ERROR] Agent binary path missing; rerun bootstrap" >&2
+  exit 1
+fi
+
+AGENT_BINARY="$(cat "$TMP_ROOT/agent_binary_path")"
+if [[ ! -x "$AGENT_BINARY" ]]; then
+  echo "[ERROR] Agent binary not executable: $AGENT_BINARY" >&2
+  exit 1
+fi
+
+if [[ -f "$ENV_FILE" ]]; then
+  set -a
+  # shellcheck disable=SC1090
+  source "$ENV_FILE"
+  set +a
+else
+  REPO_ROOT="$(cd "$TEST_ROOT/../../.." && pwd)"
+  # shellcheck disable=SC1090
+  source "$REPO_ROOT/scripts/common/build_user.sh"
+  load_build_user
+fi
+
+AGENT_UID="${ARGUS_BUILD_UID:-2133}"
+AGENT_GID="${ARGUS_BUILD_GID:-2015}"
+
+compose() {
+  if docker compose version >/dev/null 2>&1; then
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+before_file="$TMP_ROOT/before_restart.json"
+curl -sS "$API_BASE/nodes/$NODE_ID" -o "$before_file"
+prev_last_updated=$(python3 - "$before_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+print(node.get("last_updated", ""))
+PY
+)
+prev_ip=$(python3 - "$before_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+print(node["meta_data"].get("ip", ""))
+PY
+)
+initial_ip=$(cat "$TMP_ROOT/initial_ip")
+if [[ "$prev_ip" != "$initial_ip" ]]; then
+  echo "[ERROR] Expected initial IP $initial_ip, got $prev_ip" >&2
+  exit 1
+fi
+
+env_before_file="$TMP_ROOT/env_before_restart.json"
+curl -sS "$API_BASE/nodes/$ENV_NODE_ID" -o "$env_before_file"
+env_prev_last_updated=$(python3 - "$env_before_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+print(node.get("last_updated", ""))
+PY
+)
+env_prev_ip=$(python3 - "$env_before_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+print(node["meta_data"].get("ip", ""))
+PY
+)
+
+pushd "$TEST_ROOT" >/dev/null
+compose rm -sf agent
+compose rm -sf agent_env
+popd >/dev/null
+
+docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
+docker container rm -f argus-agent-env-e2e >/dev/null 2>&1 || true
+
+AGENT_DIR="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME"
+HEALTH_DIR="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME/health"
+
+ENV_AGENT_DIR="$TEST_ROOT/private/argus/agent/$ENV_AGENT_HOSTNAME"
+ENV_HEALTH_DIR="$TEST_ROOT/private/argus/agent/$ENV_AGENT_HOSTNAME/health"
+
+# 先以 sleep 方式启动容器，确保我们掌握注册时的网络状态
+if ! docker run -d \
+  --name argus-agent-e2e \
+  --hostname "$AGENT_HOSTNAME" \
+  --network "$NETWORK_NAME" \
+  --ip "$NEW_AGENT_IP" \
+  -v "$AGENT_DIR:/private/argus/agent/$AGENT_HOSTNAME" \
+  -v "$HEALTH_DIR:/private/argus/agent/$AGENT_HOSTNAME/health" \
+  -v "$TEST_ROOT/private/argus/etc:/private/argus/etc" \
+  -v "$AGENT_BINARY:/usr/local/bin/argus-agent:ro" \
+  -v "$ENTRYPOINT_SCRIPT:/usr/local/bin/agent-entrypoint.sh:ro" \
+  -v "$VERIFY_SCRIPT:/usr/local/bin/agent_deployment_verify.sh:ro" \
+  -e MASTER_ENDPOINT=http://master.argus.com:3000 \
+  -e REPORT_INTERVAL_SECONDS=2 \
+  -e ARGUS_BUILD_UID="$AGENT_UID" \
+  -e ARGUS_BUILD_GID="$AGENT_GID" \
+  --entrypoint /usr/local/bin/agent-entrypoint.sh \
+  ubuntu:22.04 >/dev/null; then
+  echo "[ERROR] Failed to start agent container with custom IP" >&2
+  exit 1
+fi
+
+success=false
+detail_file="$TMP_ROOT/post_restart.json"
+for _ in {1..20}; do
+  sleep 3
+  if ! curl -sS "$API_BASE/nodes/$NODE_ID" -o "$detail_file"; then
+    continue
+  fi
+  if python3 - "$detail_file" "$prev_last_updated" "$NODE_ID" "$prev_ip" "$NEW_AGENT_IP" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+prev_last_updated = sys.argv[2]
+expected_id = sys.argv[3]
+old_ip = sys.argv[4]
+expected_ip = sys.argv[5]
+last_updated = node.get("last_updated")
+current_ip = node["meta_data"].get("ip")
+assert node["id"] == expected_id
+if current_ip != expected_ip:
+    raise SystemExit(1)
+if current_ip == old_ip:
+    raise SystemExit(1)
+if not last_updated or last_updated == prev_last_updated:
+    raise SystemExit(1)
+PY
+  then
+    success=true
+    break
+  fi
+done
+
+if [[ "$success" != true ]]; then
+  echo "[ERROR] Agent did not report expected new IP $NEW_AGENT_IP after restart" >&2
+  exit 1
+fi
+
+echo "[INFO] Agent restart produced successful re-registration with IP change"
+
+# ---- Restart env-driven agent without metadata environment variables ----
+
+if [[ ! -d "$ENV_AGENT_DIR" ]]; then
+  echo "[ERROR] Env agent data dir missing at $ENV_AGENT_DIR" >&2
+  exit 1
+fi
+
+if [[ ! -d "$ENV_HEALTH_DIR" ]]; then
+  mkdir -p "$ENV_HEALTH_DIR"
+fi
+
+if ! docker run -d \
+  --name argus-agent-env-e2e \
+  --hostname "$ENV_AGENT_HOSTNAME" \
+  --network "$NETWORK_NAME" \
+  --ip "$NEW_ENV_AGENT_IP" \
+  -v "$ENV_AGENT_DIR:/private/argus/agent/$ENV_AGENT_HOSTNAME" \
+  -v "$ENV_HEALTH_DIR:/private/argus/agent/$ENV_AGENT_HOSTNAME/health" \
+  -v "$TEST_ROOT/private/argus/etc:/private/argus/etc" \
+  -v "$AGENT_BINARY:/usr/local/bin/argus-agent:ro" \
+  -v "$ENTRYPOINT_SCRIPT:/usr/local/bin/agent-entrypoint.sh:ro" \
+  -v "$VERIFY_SCRIPT:/usr/local/bin/agent_deployment_verify.sh:ro" \
+  -e MASTER_ENDPOINT=http://master.argus.com:3000 \
+  -e REPORT_INTERVAL_SECONDS=2 \
+  -e ARGUS_BUILD_UID="$AGENT_UID" \
+  -e ARGUS_BUILD_GID="$AGENT_GID" \
+  --entrypoint /usr/local/bin/agent-entrypoint.sh \
+  ubuntu:22.04 >/dev/null; then
+  echo "[ERROR] Failed to start env-driven agent container without metadata env" >&2
+  exit 1
+fi
+
+env_success=false
+env_detail_file="$TMP_ROOT/env_post_restart.json"
+for _ in {1..20}; do
+  sleep 3
+  if ! curl -sS "$API_BASE/nodes/$ENV_NODE_ID" -o "$env_detail_file"; then
+    continue
+  fi
+  if python3 - "$env_detail_file" "$env_prev_last_updated" "$ENV_NODE_ID" "$env_prev_ip" "$NEW_ENV_AGENT_IP" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+prev_last_updated = sys.argv[2]
+expected_id = sys.argv[3]
+old_ip = sys.argv[4]
+expected_ip = sys.argv[5]
+last_updated = node.get("last_updated")
+current_ip = node["meta_data"].get("ip")
+meta = node.get("meta_data", {})
+assert node["id"] == expected_id
+if current_ip != expected_ip:
+    raise SystemExit(1)
+if current_ip == old_ip:
+    raise SystemExit(1)
+if not last_updated or last_updated == prev_last_updated:
+    raise SystemExit(1)
+if meta.get("env") != "prod" or meta.get("user") != "ml" or meta.get("instance") != "node-3":
+    raise SystemExit(1)
+PY
+  then
+    env_success=true
+    break
+  fi
+done
+
+if [[ "$env_success" != true ]]; then
+  echo "[ERROR] Env-driven agent did not reuse persisted metadata after restart" >&2
+  exit 1
+fi
+
+echo "[INFO] Env-driven agent restart succeeded with persisted metadata"

src/agent/tests/scripts/08_down.sh

@@ -3,6 +3,7 @@ set -euo pipefail
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+ENV_FILE="$TEST_ROOT/.env"
 
 compose() {
   if docker compose version >/dev/null 2>&1; then
@@ -12,13 +13,24 @@ compose() {
   fi
 }
 
-docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
+docker container rm -f argus-agent-e2e argus-agent-env-e2e >/dev/null 2>&1 || true
 
 pushd "$TEST_ROOT" >/dev/null
 compose down --remove-orphans
 popd >/dev/null
 
-rm -rf "$TEST_ROOT/private"
+if [[ -d "$TEST_ROOT/private" ]]; then
+  docker run --rm \
+    -v "$TEST_ROOT/private:/target" \
+    ubuntu:24.04 \
+    chown -R "$(id -u):$(id -g)" /target >/dev/null 2>&1 || true
+  rm -rf "$TEST_ROOT/private"
+fi
+
 rm -rf "$TEST_ROOT/tmp"
 
+if [[ -f "$ENV_FILE" ]]; then
+  rm -f "$ENV_FILE"
+fi
+
 echo "[INFO] Agent E2E environment cleaned up"

src/agent/tests/scripts/agent_entrypoint.sh

@@ -5,11 +5,36 @@ LOG_PREFIX="[AGENT-ENTRYPOINT]"
 DNS_SCRIPT="/private/argus/etc/update-dns.sh"
 DNS_CONF="/private/argus/etc/dns.conf"
 TARGET_DOMAIN="master.argus.com"
+AGENT_UID="${ARGUS_BUILD_UID:-2133}"
+AGENT_GID="${ARGUS_BUILD_GID:-2015}"
+AGENT_HOSTNAME="${HOSTNAME:-unknown}"
+AGENT_DATA_DIR="/private/argus/agent/${AGENT_HOSTNAME}"
+AGENT_HEALTH_DIR="${AGENT_DATA_DIR}/health"
+RUNTIME_GROUP="argusagent"
+RUNTIME_USER="argusagent"
 
 log() {
   echo "${LOG_PREFIX} $*"
 }
 
+mkdir -p "$AGENT_DATA_DIR" "$AGENT_HEALTH_DIR"
+chown -R "$AGENT_UID:$AGENT_GID" "$AGENT_DATA_DIR" "$AGENT_HEALTH_DIR" 2>/dev/null || true
+chown -R "$AGENT_UID:$AGENT_GID" "/private/argus/etc" 2>/dev/null || true
+
+if ! getent group "$AGENT_GID" >/dev/null 2>&1; then
+  groupadd -g "$AGENT_GID" "$RUNTIME_GROUP"
+else
+  RUNTIME_GROUP="$(getent group "$AGENT_GID" | cut -d: -f1)"
+fi
+
+if ! getent passwd "$AGENT_UID" >/dev/null 2>&1; then
+  useradd -u "$AGENT_UID" -g "$AGENT_GID" -M -s /bin/bash "$RUNTIME_USER"
+else
+  RUNTIME_USER="$(getent passwd "$AGENT_UID" | cut -d: -f1)"
+fi
+
+log "运行用户: $RUNTIME_USER ($AGENT_UID:$AGENT_GID)"
+
 # 中文提示：等待 bind 下发的 update-dns.sh 脚本
 for _ in {1..30}; do
   if [[ -x "$DNS_SCRIPT" ]]; then
@@ -21,9 +46,14 @@ done
 
 if [[ -x "$DNS_SCRIPT" ]]; then
   log "执行 update-dns.sh 更新容器 DNS"
-  if ! "$DNS_SCRIPT"; then
+  while true; do
-    log "update-dns.sh 执行失败，继续尝试默认 DNS"
+    if "$DNS_SCRIPT"; then
-  fi
+      log "update-dns.sh 执行成功"
+      break
+    fi
+    log "update-dns.sh 执行失败，3 秒后重试"
+    sleep 3
+  done
 else
   log "未获取到 update-dns.sh，使用镜像默认 DNS"
 fi
@@ -46,4 +76,4 @@ for _ in {1..30}; do
 done
 
 log "启动 argus-agent"
-exec /usr/local/bin/argus-agent
+exec su -s /bin/bash -c /usr/local/bin/argus-agent "$RUNTIME_USER"

src/agent/tests/test_config_metadata.py

@@ -0,0 +1,151 @@
+from __future__ import annotations
+
+import os
+import unittest
+from contextlib import contextmanager
+from unittest.mock import patch
+
+from app.config import AgentConfig, load_config
+
+
+@contextmanager
+def temp_env(**overrides: str | None):
+    originals: dict[str, str | None] = {}
+    try:
+        for key, value in overrides.items():
+            originals[key] = os.environ.get(key)
+            if value is None:
+                os.environ.pop(key, None)
+            else:
+                os.environ[key] = value
+        yield
+    finally:
+        for key, original in originals.items():
+            if original is None:
+                os.environ.pop(key, None)
+            else:
+                os.environ[key] = original
+
+
+class LoadConfigMetadataTests(unittest.TestCase):
+    @patch("app.config.Path.mkdir")
+    def test_metadata_from_environment_variables(self, mock_mkdir):
+        with temp_env(
+            MASTER_ENDPOINT="http://master.local",
+            AGENT_HOSTNAME="dev-user-one-pod",
+            AGENT_ENV="prod",
+            AGENT_USER="ops",
+            AGENT_INSTANCE="node-1",
+        ):
+            config = load_config()
+
+        self.assertEqual(config.environment, "prod")
+        self.assertEqual(config.user, "ops")
+        self.assertEqual(config.instance, "node-1")
+        mock_mkdir.assert_called()
+
+    @patch("app.config.Path.mkdir")
+    def test_metadata_falls_back_to_hostname(self, mock_mkdir):
+        with temp_env(
+            MASTER_ENDPOINT="http://master.local",
+            AGENT_HOSTNAME="qa-team-abc-pod-2",
+            AGENT_ENV=None,
+            AGENT_USER=None,
+            AGENT_INSTANCE=None,
+        ):
+            config = load_config()
+
+        self.assertEqual(config.environment, "qa")
+        self.assertEqual(config.user, "team")
+        self.assertEqual(config.instance, "abc")
+        mock_mkdir.assert_called()
+
+    @patch("app.config._load_metadata_from_state", return_value=("prod", "ops", "node-1"))
+    @patch("app.config.Path.mkdir")
+    def test_metadata_from_node_state(self, mock_mkdir, mock_state):
+        with temp_env(
+            MASTER_ENDPOINT="http://master.local",
+            AGENT_HOSTNAME="host_abc",
+            AGENT_ENV=None,
+            AGENT_USER=None,
+            AGENT_INSTANCE=None,
+        ):
+            config = load_config()
+
+        self.assertEqual(config.environment, "prod")
+        self.assertEqual(config.user, "ops")
+        self.assertEqual(config.instance, "node-1")
+        mock_state.assert_called_once()
+        mock_mkdir.assert_called()
+
+    @patch("app.config.Path.mkdir")
+    def test_partial_environment_variables_fallback(self, mock_mkdir):
+        with temp_env(
+            MASTER_ENDPOINT="http://master.local",
+            AGENT_HOSTNAME="stage-ml-001-node",
+            AGENT_ENV="prod",
+            AGENT_USER=None,
+            AGENT_INSTANCE=None,
+        ):
+            config = load_config()
+
+        self.assertEqual(config.environment, "stage")
+        self.assertEqual(config.user, "ml")
+        self.assertEqual(config.instance, "001")
+        mock_mkdir.assert_called()
+
+    @patch("app.config.Path.mkdir")
+    def test_invalid_hostname_raises_error(self, mock_mkdir):
+        with temp_env(
+            MASTER_ENDPOINT="http://master.local",
+            AGENT_HOSTNAME="invalidhostname",
+            AGENT_ENV=None,
+            AGENT_USER=None,
+            AGENT_INSTANCE=None,
+        ):
+            with self.assertRaises(ValueError):
+                load_config()
+
+        mock_mkdir.assert_not_called()
+
+
+class CollectMetadataTests(unittest.TestCase):
+    @patch("app.collector._detect_ip_address", return_value="127.0.0.1")
+    @patch("app.collector._detect_gpu_count", return_value=0)
+    @patch("app.collector._detect_memory_bytes", return_value=1024)
+    @patch("app.collector._detect_cpu_count", return_value=8)
+    def test_collect_metadata_uses_config_fields(
+        self,
+        mock_cpu,
+        mock_memory,
+        mock_gpu,
+        mock_ip,
+    ):
+        config = AgentConfig(
+            hostname="dev-user-001-pod",
+            environment="prod",
+            user="ops",
+            instance="node-1",
+            node_file="/tmp/node.json",
+            version="1.0.0",
+            master_endpoint="http://master.local",
+            report_interval_seconds=60,
+            health_dir="/tmp/health",
+        )
+
+        from app.collector import collect_metadata
+
+        metadata = collect_metadata(config)
+
+        self.assertEqual(metadata["env"], "prod")
+        self.assertEqual(metadata["user"], "ops")
+        self.assertEqual(metadata["instance"], "node-1")
+        self.assertEqual(metadata["hostname"], "dev-user-001-pod")
+        self.assertEqual(metadata["ip"], "127.0.0.1")
+        self.assertEqual(metadata["cpu_number"], 8)
+        self.assertEqual(metadata["memory_in_bytes"], 1024)
+        self.assertEqual(metadata["gpu_number"], 0)
+
+
+if __name__ == "__main__":
+    unittest.main()

src/alert/README.md

@@ -0,0 +1,31 @@
+# Alertmanager
+
+## 构建
+1. 首先设置构建和部署的环境变量, 在项目根目录下执行：
+```bash
+cp src/alert/tests/.env.example src/alert/tests/.env
+```
+
+然后找到复制出来的.env文件，修改环境变量。
+
+2. 使用脚本构建，在项目根目录下执行：
+
+```bash
+bash src/alert/alertmanager/build/build.sh
+```
+
+构建成功后，会在项目根目录下生成argus-alertmanager-latest.tar
+
+## 部署
+
+提供docker-compose部署。在src/alert/tests目录下
+```bash
+docker-compose up -d
+```
+
+## 动态配置
+配置文件放在`/private/argus/alert/alertmanager/alertmanager.yml`下，修改alertmanager.yml后，调用`http://alertmanager.alert.argus.com:9093/-/reload`接口(POST)可以重新加载配置.
+
+```bash
+curl -X POST http://localhost:9093/-/reload
+```

src/alert/alertmanager/build/Dockerfile

@@ -0,0 +1,96 @@
+# 基于 Ubuntu 24.04
+FROM ubuntu:24.04
+
+# 切换到 root 用户
+USER root
+
+# 安装必要依赖
+RUN apt-get update && \
+    apt-get install -y wget supervisor net-tools inetutils-ping vim ca-certificates passwd && \
+    apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# 设置 Alertmanager 版本（与本地离线包保持一致）
+ARG ALERTMANAGER_VERSION=0.28.1
+
+# 使用仓库内预置的离线包构建（无需联网）
+COPY src/alert/alertmanager/build/alertmanager-${ALERTMANAGER_VERSION}.linux-amd64.tar.gz /tmp/
+RUN tar xvf /tmp/alertmanager-${ALERTMANAGER_VERSION}.linux-amd64.tar.gz -C /tmp && \
+    mv /tmp/alertmanager-${ALERTMANAGER_VERSION}.linux-amd64 /usr/local/alertmanager && \
+    rm -f /tmp/alertmanager-${ALERTMANAGER_VERSION}.linux-amd64.tar.gz
+
+ENV ALERTMANAGER_BASE_PATH=/private/argus/alert/alertmanager
+
+ARG ARGUS_BUILD_UID=2133
+ARG ARGUS_BUILD_GID=2015
+ENV ARGUS_BUILD_UID=${ARGUS_BUILD_UID}
+ENV ARGUS_BUILD_GID=${ARGUS_BUILD_GID}
+
+RUN mkdir -p /usr/share/alertmanager && \
+    mkdir -p ${ALERTMANAGER_BASE_PATH} && \
+    mkdir -p /private/argus/etc && \
+    rm -rf /alertmanager && \
+    ln -s ${ALERTMANAGER_BASE_PATH} /alertmanager
+
+# 创建 alertmanager 用户（可自定义 UID/GID）
+# 创建 alertmanager 用户组
+RUN set -eux; \
+    # 确保目标 GID 存在；若已被占用，直接使用该 GID（组名不限）\
+    if ! getent group "${ARGUS_BUILD_GID}" >/dev/null; then \
+        groupadd -g "${ARGUS_BUILD_GID}" alertmanager || true; \
+    fi; \
+    # 确保存在 alertmanager 用户；若 UID 已被占用，跳过并继续使用现有 UID 的用户
+    if ! id alertmanager >/dev/null 2>&1; then \
+        if getent passwd "${ARGUS_BUILD_UID}" >/dev/null; then \
+            # UID 已占用，则创建同名用户但不指定 UID（避免冲突），仅保证 user 存在
+            useradd -M -s /usr/sbin/nologin -g "${ARGUS_BUILD_GID}" alertmanager || true; \
+        else \
+            useradd -M -s /usr/sbin/nologin -u "${ARGUS_BUILD_UID}" -g "${ARGUS_BUILD_GID}" alertmanager || true; \
+        fi; \
+    else \
+        usermod -g "${ARGUS_BUILD_GID}" alertmanager || true; \
+    fi
+
+RUN chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" /usr/share/alertmanager /alertmanager ${ALERTMANAGER_BASE_PATH} /private/argus/etc /usr/local/bin || true
+
+# 配置内网 apt 源 (如果指定了内网选项)
+RUN if [ "$USE_INTRANET" = "true" ]; then \
+        echo "Configuring intranet apt sources..." && \
+        cp /etc/apt/sources.list /etc/apt/sources.list.bak && \
+        echo "deb [trusted=yes] http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
+        echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
+    fi
+
+
+# 配置部署时使用的 apt 源
+RUN if [ "$USE_INTRANET" = "true" ]; then \
+    echo "deb [trusted=yes] https://10.92.132.52/mirrors/ubuntu2204/ jammy main" > /etc/apt/sources.list; \
+    fi
+
+# 创建 supervisor 日志目录
+RUN mkdir -p /var/log/supervisor
+
+# 复制 supervisor 配置文件
+COPY src/alert/alertmanager/build/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+
+# 复制启动脚本
+COPY src/alert/alertmanager/build/start-am-supervised.sh /usr/local/bin/start-am-supervised.sh
+RUN chmod +x /usr/local/bin/start-am-supervised.sh
+
+# 复制 Alertmanager 配置文件
+COPY src/alert/alertmanager/build/alertmanager.yml /etc/alertmanager/alertmanager.yml
+RUN chmod +x /etc/alertmanager/alertmanager.yml
+# COPY src/alert/alertmanager/build/alertmanager.yml ${ALERTMANAGER_BASE_PATH}/alertmanager.yml
+
+# 复制 DNS 监控脚本
+COPY src/alert/alertmanager/build/dns-monitor.sh /usr/local/bin/dns-monitor.sh
+RUN chmod +x /usr/local/bin/dns-monitor.sh
+
+# 保持 root 用户，由 supervisor 控制 user 切换
+USER root
+
+# 暴露端口（Alertmanager 默认端口 9093）
+EXPOSE 9093
+
+# 使用 supervisor 作为入口点
+CMD ["/usr/bin/supervisord", "-c", "/etc/supervisor/conf.d/supervisord.conf"]

src/alert/alertmanager/build/alertmanager.yml

@@ -0,0 +1,19 @@
+global:
+  resolve_timeout: 5m
+
+route:
+  group_by: ['alertname', 'instance']   # 分组：相同 alertname + instance 的告警合并
+  group_wait: 30s        # 第一个告警后，等 30s 看是否有同组告警一起发
+  group_interval: 5m     # 同组告警变化后，至少 5 分钟再发一次
+  repeat_interval: 3h    # 相同告警，3 小时重复提醒一次
+  receiver: 'null'
+
+receivers:
+  - name: 'null'
+
+inhibit_rules:
+  - source_match:
+      severity: 'critical'     # critical 告警存在时
+    target_match:
+      severity: 'warning'      # 抑制相同 instance 的 warning 告警
+    equal: ['instance']

src/alert/alertmanager/build/build.sh

@@ -0,0 +1,13 @@
+#!/bin/bash
+set -euo pipefail
+docker pull ubuntu:24.04
+
+source src/alert/tests/.env
+
+docker build \
+  --build-arg ARGUS_BUILD_UID=${ARGUS_BUILD_UID} \
+  --build-arg ARGUS_BUILD_GID=${ARGUS_BUILD_GID} \
+  -f src/alert/alertmanager/build/Dockerfile \
+  -t argus-alertmanager:latest .
+
+docker save -o argus-alertmanager-latest.tar argus-alertmanager:latest

src/alert/alertmanager/build/dns-monitor.sh

@@ -0,0 +1,68 @@
+#!/bin/bash
+
+# DNS监控脚本 - 每10秒检查dns.conf是否有变化
+# 如果有变化则执行update-dns.sh脚本
+
+DNS_CONF="/private/argus/etc/dns.conf"
+DNS_BACKUP="/tmp/dns.conf.backup"
+UPDATE_SCRIPT="/private/argus/etc/update-dns.sh"
+LOG_FILE="/var/log/supervisor/dns-monitor.log"
+
+# 确保日志文件存在
+touch "$LOG_FILE"
+
+log_message() {
+    echo "$(date '+%Y-%m-%d %H:%M:%S') [DNS-Monitor] $1" >> "$LOG_FILE"
+}
+
+log_message "DNS监控脚本启动"
+
+while true; do
+    if [ -f "$DNS_CONF" ]; then
+        if [ -f "$DNS_BACKUP" ]; then
+            # 比较文件内容
+            if ! cmp -s "$DNS_CONF" "$DNS_BACKUP"; then
+                log_message "检测到DNS配置变化"
+
+                # 更新备份文件
+                cp "$DNS_CONF" "$DNS_BACKUP"
+
+                # 执行更新脚本
+                if [ -x "$UPDATE_SCRIPT" ]; then
+                    log_message "执行DNS更新脚本: $UPDATE_SCRIPT"
+                    "$UPDATE_SCRIPT" >> "$LOG_FILE" 2>&1
+                    if [ $? -eq 0 ]; then
+                        log_message "DNS更新脚本执行成功"
+                    else
+                        log_message "DNS更新脚本执行失败"
+                    fi
+                else
+                    log_message "警告: 更新脚本不存在或不可执行: $UPDATE_SCRIPT"
+                fi
+            fi
+        else
+
+            # 第一次检测到配置文件，执行更新脚本
+            if [ -x "$UPDATE_SCRIPT" ]; then
+                log_message "执行DNS更新脚本: $UPDATE_SCRIPT"
+                "$UPDATE_SCRIPT" >> "$LOG_FILE" 2>&1
+                if [ $? -eq 0 ]; then
+                    log_message "DNS更新脚本执行成功"
+
+		    # 第一次运行，创建备份并执行更新
+		    cp "$DNS_CONF" "$DNS_BACKUP"
+		    log_message "创建DNS配置备份文件"
+
+                else
+                    log_message "DNS更新脚本执行失败"
+                fi
+            else
+                log_message "警告: 更新脚本不存在或不可执行: $UPDATE_SCRIPT"
+            fi
+        fi
+    else
+        log_message "警告: DNS配置文件不存在: $DNS_CONF"
+    fi
+
+    sleep 10
+done

src/alert/alertmanager/build/fetch-dist.sh

@@ -0,0 +1,22 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 下载 Alertmanager 离线安装包到本目录，用于 Docker 构建时 COPY
+# 用法：
+#   ./fetch-dist.sh [version]
+# 示例：
+#   ./fetch-dist.sh 0.28.1
+
+VER="${1:-0.28.1}"
+OUT="alertmanager-${VER}.linux-amd64.tar.gz"
+URL="https://github.com/prometheus/alertmanager/releases/download/v${VER}/${OUT}"
+
+if [[ -f "$OUT" ]]; then
+  echo "[INFO] $OUT already exists, skip download"
+  exit 0
+fi
+
+echo "[INFO] Downloading $URL"
+curl -fL --retry 3 --connect-timeout 10 -o "$OUT" "$URL"
+echo "[OK] Saved to $(pwd)/$OUT"
+

src/alert/alertmanager/build/start-am-supervised.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Alertmanager under supervisor..."
+
+ALERTMANAGER_BASE_PATH=${ALERTMANAGER_BASE_PATH:-/private/argus/alert/alertmanager}
+
+echo "[INFO] Alertmanager base path: ${ALERTMANAGER_BASE_PATH}"
+
+# 使用容器内的 /etc/alertmanager/alertmanager.yml 作为配置文件，避免写入挂载卷导致的权限问题
+echo "[INFO] Using /etc/alertmanager/alertmanager.yml as configuration"
+
+
+# 记录容器 IP 地址
+DOMAIN=alertmanager.alert.argus.com
+IP=$(ifconfig | grep -A 1 eth0 | grep inet | awk '{print $2}')
+echo "current IP: ${IP}"
+echo "${IP}" > /private/argus/etc/${DOMAIN}
+chmod 755 /private/argus/etc/${DOMAIN}
+
+
+echo "[INFO] Starting Alertmanager process..."
+
+# 启动 Alertmanager 主进程
+exec /usr/local/alertmanager/alertmanager --config.file=/etc/alertmanager/alertmanager.yml --storage.path=/alertmanager --cluster.listen-address=""

src/alert/alertmanager/build/supervisord.conf

@@ -0,0 +1,39 @@
+[supervisord]
+nodaemon=true
+logfile=/var/log/supervisor/supervisord.log
+pidfile=/var/run/supervisord.pid
+user=root
+
+[program:alertmanager]
+command=/usr/local/bin/start-am-supervised.sh
+user=ubuntu
+stdout_logfile=/var/log/supervisor/alertmanager.log
+stderr_logfile=/var/log/supervisor/alertmanager_error.log
+autorestart=true
+startretries=3
+startsecs=10
+stopwaitsecs=20
+killasgroup=true
+stopasgroup=true
+
+[program:dns-monitor]
+command=/usr/local/bin/dns-monitor.sh
+user=root
+stdout_logfile=/var/log/supervisor/dns-monitor.log
+stderr_logfile=/var/log/supervisor/dns-monitor_error.log
+autorestart=true
+startretries=3
+startsecs=5
+stopwaitsecs=10
+killasgroup=true
+stopasgroup=true
+
+[unix_http_server]
+file=/var/run/supervisor.sock
+chmod=0700
+
+[supervisorctl]
+serverurl=unix:///var/run/supervisor.sock
+
+[rpcinterface:supervisor]
+supervisor.rpcinterface_factory = supervisor.rpcinterface:make_main_rpcinterface

src/alert/alertmanager/config/rule_files/README.md

@@ -0,0 +1,60 @@
+# 告警配置
+
+> 参考：[自定义Prometheus告警规则](https://yunlzheng.gitbook.io/prometheus-book/parti-prometheus-ji-chu/alert/prometheus-alert-rule)
+
+在Prometheus中配置告警的有两个步骤：
+
+1. 写告警规则文件（rules文件）
+2. 在promethues.yml里加载规则，并配置Alertmanager
+
+## 1. 编写告警规则文件
+告警规则如下：
+```yml
+groups:
+  - name: example-rules
+    interval: 30s  # 每30秒评估一次
+    rules:
+      - alert: InstanceDown
+        expr: up == 0
+        for: 1m
+        labels:
+          severity: critical
+        annotations:
+          summary: "实例 {{ $labels.instance }} 已宕机"
+          description: "{{ $labels.instance }} 在 {{ $labels.job }} 中无响应超过 1 分钟。"
+
+      - alert: HighCpuUsage
+        expr: 100 - (avg by (instance) (irate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 80
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          summary: "CPU 使用率过高"
+          description: "实例 {{ $labels.instance }} CPU 使用率超过 80% 持续 5 分钟。"
+```
+
+其中：
+
+- `alert`：告警规则的名称。
+- `expr`：基于PromQL表达式告警触发条件，用于计算是否有时间序列满足该条件。
+- `for`：评估等待时间，可选参数。用于表示只有当触发条件持续一段时间后才发送告警。在等待期间新产生告警的状态为pending。
+- `labels`：自定义标签，允许用户指定要附加到告警上的一组附加标签，可以在Alertmanager中做路由和分组。
+- `annotations`：用于指定一组附加信息，比如用于描述告警详细信息的文字等，annotations的内容在告警产生时会一同作为参数发送到Alertmanager。可以提供告警摘要和详细信息。
+
+## 2. promothues.yml里引用
+在prometheus.yml中加上`rule_files`和`alerting`:
+
+```yml
+global:
+  [ evaluation_interval: <duration> | default = 1m ]
+
+rule_files:
+  [ - <filepath_glob> ... ]
+
+alerting:
+  alertmanagers:
+    - static_configs:
+        - targets:
+            - "alertmanager.alert.argus.com:9093"   # Alertmanager 地址
+
+```

src/alert/alertmanager/config/rule_files/example_rules.yml

@@ -0,0 +1,37 @@
+groups:
+  - name: example-rules
+    interval: 30s  # 每30秒评估一次
+    rules:
+      - alert: InstanceDown
+        expr: up == 0
+        for: 1m
+        labels:
+          severity: critical
+        annotations:
+          summary: "实例 {{ $labels.instance }} 已宕机"
+          description: "{{ $labels.instance }} 在 {{ $labels.job }} 中无响应超过 1 分钟。"
+
+      - alert: HighCpuUsage
+        expr: 100 - (avg by (instance) (irate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 80
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          summary: "CPU 使用率过高"
+          description: "实例 {{ $labels.instance }} CPU 使用率超过 80% 持续 5 分钟。"
+      - alert: HighMemoryUsage
+        expr: (node_memory_MemTotal_bytes - node_memory_MemAvailable_bytes) / node_memory_MemTotal_bytes * 100 > 80
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          summary: "内存使用率过高"
+          description: "实例 {{ $labels.instance }} 内存使用率超过 80% 持续 5 分钟。"
+      - alert: DiskSpaceLow
+        expr: (node_filesystem_size_bytes{fstype!~"tmpfs|overlay"} - node_filesystem_free_bytes{fstype!~"tmpfs|overlay"}) / node_filesystem_size_bytes{fstype!~"tmpfs|overlay"} * 100 > 90
+        for: 10m
+        labels:
+          severity: warning
+        annotations:
+          summary: "磁盘空间不足"
+          description: "实例 {{ $labels.instance }} 磁盘空间不足超过 90% 持续 10 分钟。"

src/alert/tests/.env

@@ -0,0 +1,5 @@
+DATA_ROOT=/home/argus/tmp/private/argus
+ARGUS_BUILD_UID=1048
+ARGUS_BUILD_GID=1048
+
+USE_INTRANET=false

src/alert/tests/.env.example

@@ -0,0 +1,5 @@
+DATA_ROOT=/home/argus/tmp/private/argus
+ARGUS_BUILD_UID=1048
+ARGUS_BUILD_GID=1048
+
+USE_INTRANET=false

src/alert/tests/docker-compose.yml

@@ -0,0 +1,37 @@
+services:
+  alertmanager:
+    build:
+      context: ../../../
+      dockerfile: src/alert/alertmanager/build/Dockerfile
+      args:
+        ARGUS_BUILD_UID: ${ARGUS_BUILD_UID:-2133}
+        ARGUS_BUILD_GID: ${ARGUS_BUILD_GID:-2015}
+        USE_INTRANET: ${USE_INTRANET:-false}
+    image: argus-alertmanager:latest
+    container_name: argus-alertmanager
+    environment:
+      - ALERTMANAGER_BASE_PATH=/private/argus/alert/alertmanager
+      - ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}
+      - ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}
+    ports:
+      - "${ARGUS_PORT:-9093}:9093"
+    volumes:
+      - ${DATA_ROOT:-./data}/alert/alertmanager:/private/argus/alert/alertmanager
+      - ${DATA_ROOT:-./data}/etc:/private/argus/etc
+    networks:
+      - argus-debug-net
+    restart: unless-stopped
+    logging:
+      driver: "json-file"
+      options:
+        max-size: "10m"
+        max-file: "3"
+
+networks:
+  argus-debug-net:
+    driver: bridge
+    name: argus-debug-net
+
+volumes:
+  alertmanager_data:
+    driver: local

src/alert/tests/scripts/verify_alertmanager.sh

@@ -0,0 +1,113 @@
+#!/bin/bash
+# verify_alertmanager.sh
+# 用于部署后验证 Prometheus 与 Alertmanager 通信链路是否正常
+
+set -euo pipefail
+
+#=============================
+# 基础配置
+#=============================
+PROM_URL="${PROM_URL:-http://prom.metric.argus.com:9090}"
+ALERT_URL="${ALERT_URL:-http://alertmanager.alert.argus.com:9093}"
+# TODO: 根据实际部署环境调整规则目录
+DATA_ROOT="${DATA_ROOT:-/private/argus}"
+RULE_DIR = "$DATA_ROOT/metric/prometheus/rules"
+TMP_RULE="/tmp/test_rule.yml"
+
+#=============================
+# 辅助函数
+#=============================
+GREEN="\033[32m"; RED="\033[31m"; YELLOW="\033[33m"; RESET="\033[0m"
+
+log_info()    { echo -e "${YELLOW}[INFO]${RESET} $1"; }
+log_success() { echo -e "${GREEN}[OK]${RESET} $1"; }
+log_error()   { echo -e "${RED}[ERROR]${RESET} $1"; }
+
+fail_exit() { log_error "$1"; exit 1; }
+
+#=============================
+# Step 1: 检查 Alertmanager 是否可访问
+#=============================
+log_info "检查 Alertmanager 状态..."
+if curl -sSf "${ALERT_URL}/api/v2/status" >/dev/null 2>&1; then
+  log_success "Alertmanager 服务正常 (${ALERT_URL})"
+else
+  fail_exit "无法访问 Alertmanager，请检查端口映射与容器状态。"
+fi
+
+#=============================
+# Step 2: 手动发送测试告警
+#=============================
+log_info "发送手动测试告警..."
+curl -s -XPOST "${ALERT_URL}/api/v2/alerts" -H "Content-Type: application/json" -d '[
+  {
+    "labels": {
+      "alertname": "ManualTestAlert",
+      "severity": "info"
+    },
+    "annotations": {
+      "summary": "This is a test alert from deploy verification"
+    },
+    "startsAt": "'$(date -Iseconds)'"
+  }
+]' >/dev/null && log_success "测试告警已成功发送到 Alertmanager"
+
+#=============================
+# Step 3: 检查 Prometheus 配置中是否包含 Alertmanager
+#=============================
+log_info "检查 Prometheus 是否配置了 Alertmanager..."
+if curl -s "${PROM_URL}/api/v1/status/config" | grep -q "alertmanagers"; then
+  log_success "Prometheus 已配置 Alertmanager 目标"
+else
+  fail_exit "Prometheus 未配置 Alertmanager，请检查 prometheus.yml"
+fi
+
+#=============================
+# Step 4: 创建并加载测试告警规则
+#=============================
+log_info "创建临时测试规则 ${TMP_RULE} ..."
+cat <<EOF > "${TMP_RULE}"
+groups:
+- name: deploy-verify-group
+  rules:
+  - alert: DeployVerifyAlert
+    expr: vector(1)
+    labels:
+      severity: warning
+    annotations:
+      summary: "Deployment verification alert"
+EOF
+
+mkdir -p "${RULE_DIR}"
+cp "${TMP_RULE}" "${RULE_DIR}/test_rule.yml"
+
+log_info "重载 Prometheus 以加载新规则..."
+if curl -s -X POST "${PROM_URL}/-/reload" >/dev/null; then
+  log_success "Prometheus 已重载规则"
+else
+  fail_exit "Prometheus reload 失败，请检查 API 可访问性。"
+fi
+
+#=============================
+# Step 5: 等待并验证 Alertmanager 是否收到告警
+#=============================
+log_info "等待告警触发 (约5秒)..."
+sleep 5
+
+if curl -s "${ALERT_URL}/api/v2/alerts" | grep -q "DeployVerifyAlert"; then
+  log_success "Prometheus → Alertmanager 告警链路验证成功"
+else
+  fail_exit "未在 Alertmanager 中检测到 DeployVerifyAlert，请检查网络或配置。"
+fi
+
+#=============================
+# Step 6: 清理测试规则
+#=============================
+log_info "清理临时测试规则..."
+rm -f "${RULE_DIR}/test_rule.yml" "${TMP_RULE}"
+
+curl -s -X POST "${PROM_URL}/-/reload" >/dev/null \
+  && log_success "Prometheus 已清理验证规则" \
+  || log_error "Prometheus reload 清理失败，请手动确认。"
+
+log_success "部署验证全部通过！Prometheus ↔ Alertmanager 通信正常。"

src/bind/build/Dockerfile

@@ -6,6 +6,11 @@ ENV TZ=Asia/Shanghai
 
 # 设置构建参数
 ARG USE_INTRANET=false
+ARG ARGUS_BUILD_UID=2133
+ARG ARGUS_BUILD_GID=2015
+
+ENV ARGUS_BUILD_UID=${ARGUS_BUILD_UID} \
+    ARGUS_BUILD_GID=${ARGUS_BUILD_GID}
 
 # 配置内网 apt 源 (如果指定了内网选项)
 RUN if [ "$USE_INTRANET" = "true" ]; then \
@@ -21,6 +26,7 @@ RUN apt-get update && \
     apt-get install -y \
     bind9 \
     bind9utils \
+    dnsutils \
     bind9-doc \
     supervisor \
     net-tools \
@@ -29,6 +35,24 @@ RUN apt-get update && \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/*
 
+# 调整 bind 用户与用户组 ID 以匹配宿主机配置
+RUN set -eux; \
+    current_gid="$(getent group bind | awk -F: '{print $3}')"; \
+    if [ -z "$current_gid" ]; then \
+        groupadd -g "${ARGUS_BUILD_GID}" bind; \
+    elif [ "$current_gid" != "${ARGUS_BUILD_GID}" ]; then \
+        groupmod -g "${ARGUS_BUILD_GID}" bind; \
+    fi; \
+    if id bind >/dev/null 2>&1; then \
+        current_uid="$(id -u bind)"; \
+        if [ "$current_uid" != "${ARGUS_BUILD_UID}" ]; then \
+            usermod -u "${ARGUS_BUILD_UID}" bind; \
+        fi; \
+    else \
+        useradd -m -u "${ARGUS_BUILD_UID}" -g "${ARGUS_BUILD_GID}" bind; \
+    fi; \
+    chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" /var/cache/bind /var/lib/bind
+
 # 配置部署时使用的apt源
 RUN if [ "$USE_INTRANET" = "true" ]; then \
 	echo "deb [trusted=yes] https://10.92.132.52/mirrors/ubuntu2204/ jammy main" > /etc/apt/sources.list; \

src/bind/build/argus_dns_sync.sh

@@ -9,6 +9,9 @@ SLEEP_SECONDS=10
 RELOAD_SCRIPT="/usr/local/bin/reload-bind9.sh"   # 这里放你已有脚本的路径
 
 mkdir -p "$(dirname "$LOCKFILE")" "$BACKUP_DIR"
+BACKUP_UID="${ARGUS_BUILD_UID:-2133}"
+BACKUP_GID="${ARGUS_BUILD_GID:-2015}"
+chown -R "$BACKUP_UID:$BACKUP_GID" "$BACKUP_DIR" 2>/dev/null || true
 
 is_ipv4() {
   local ip="$1"
@@ -33,6 +36,7 @@ upsert_record() {
   local changed=0
 
   cp -a "$ZONE_DB" "$BACKUP_DIR/db.argus.com.$ts.bak"
+  chown "$BACKUP_UID:$BACKUP_GID" "$BACKUP_DIR/db.argus.com.$ts.bak" 2>/dev/null || true
 
   local cur_ip
   cur_ip="$(get_current_ip "$name" || true)"
@@ -61,7 +65,10 @@ upsert_record() {
     echo "[SKIP] ${name} unchanged (${new_ip})"
   fi
 
-  return $changed
+  if [[ $changed -eq 1 ]]; then
+    return 0
+  fi
+  return 1
 }
 
 while true; do
@@ -70,7 +77,7 @@ while true; do
     shopt -s nullglob
     NEED_RELOAD=0
 
-    for f in "$WATCH_DIR"/*.argus.com; do
+  for f in "$WATCH_DIR"/*.argus.com; do
       base="$(basename "$f")"
       name="${base%.argus.com}"
       ip="$(grep -Eo '([0-9]{1,3}\.){3}[0-9]{1,3}' "$f" | tail -n1 || true)"
@@ -97,4 +104,3 @@ while true; do
 
   sleep "$SLEEP_SECONDS"
 done
-

src/bind/build/dns-monitor.sh

@@ -17,6 +17,9 @@ log_message() {
 
 log_message "DNS监控脚本启动"
 
+log_message "删除DNS备份文件（如果存在）"
+rm -f $DNS_BACKUP
+
 while true; do
     if [ -f "$DNS_CONF" ]; then
         if [ -f "$DNS_BACKUP" ]; then

src/bind/build/startup.sh

@@ -6,6 +6,8 @@ chmod 777 /private 2>/dev/null || true
 # Create persistent directories for BIND9 configs and DNS sync
 mkdir -p /private/argus/bind
 mkdir -p /private/argus/etc
+chown bind:bind /private/argus 2>/dev/null || true
+chown -R bind:bind /private/argus/bind /private/argus/etc
 
 # Copy configuration files to persistent storage if they don't exist
 if [ ! -f /private/argus/bind/named.conf.local ]; then

src/bind/tests/docker-compose.yml

@@ -3,8 +3,8 @@ services:
     image: argus-bind9:latest
     container_name: argus-bind9-test
     ports:
-      - "53:53/tcp"
+      - "${HOST_DNS_PORT:-1053}:53/tcp"
-      - "53:53/udp"
+      - "${HOST_DNS_PORT:-1053}:53/udp"
     volumes:
       - ./private:/private
     restart: unless-stopped

src/bind/tests/scripts/00_e2e_test.sh

@@ -7,6 +7,9 @@
 set -e
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
+export HOST_DNS_PORT
 
 echo "=========================================="
 echo "BIND9 DNS Server End-to-End Test Suite"

src/bind/tests/scripts/01_start_container.sh

@@ -7,13 +7,17 @@ set -e
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_DIR="$(dirname "$SCRIPT_DIR")"
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
+export HOST_DNS_PORT
 
 cd "$TEST_DIR"
 
 echo "Starting BIND9 test container..."
 
 # Ensure private directory exists with proper permissions
-mkdir -p private
+mkdir -p private/argus/bind
+mkdir -p private/argus/etc
 chmod 777 private
 
 # Start the container
@@ -35,4 +39,4 @@ fi
 
 echo ""
 echo "BIND9 test environment is ready!"
-echo "DNS server listening on localhost:53"
+echo "DNS server listening on localhost:${HOST_DNS_PORT}"

src/bind/tests/scripts/02_dig_test.sh

@@ -5,7 +5,10 @@
 
 set -e
 
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
 echo "Testing DNS resolution with dig..."
+echo "Using DNS server localhost:${HOST_DNS_PORT}"
 
 # Function to test DNS query
 test_dns_query() {
@@ -19,7 +22,7 @@ test_dns_query() {
     echo "Expected IP: $expected_ip"
 
     # Perform dig query
-    result=$(dig @localhost $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
 
     if [ "$result" = "QUERY_FAILED" ]; then
         echo "✗ DNS query failed"

src/bind/tests/scripts/03.5_dns_sync_test.sh

@@ -6,10 +6,13 @@
 
 set -e
 
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_DIR="$(dirname "$SCRIPT_DIR")"
 
 echo "=== DNS Auto-Sync Functionality Test ==="
+echo "Using DNS server localhost:${HOST_DNS_PORT}"
 
 # Check if container is running
 if ! docker compose ps | grep -q "Up"; then
@@ -36,7 +39,7 @@ test_dns_query() {
     # Wait a moment for DNS cache
     sleep 2
 
-    result=$(dig @localhost $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
 
     if [ "$result" = "$expected_ip" ]; then
         echo "✓ $result"
@@ -90,7 +93,7 @@ echo ""
 echo "Step 2: Testing initial DNS configuration..."
 
 # Get current IP for web.argus.com (may have been changed by previous tests)
-current_web_ip=$(dig @localhost web.argus.com A +short 2>/dev/null || echo "UNKNOWN")
+current_web_ip=$(dig @localhost -p "$HOST_DNS_PORT" web.argus.com A +short 2>/dev/null || echo "UNKNOWN")
 echo "Current web.argus.com IP: $current_web_ip"
 
 # Test that DNS is working (regardless of specific IP)
@@ -185,7 +188,7 @@ docker compose exec bind9 bash -c 'echo "this is not an IP address" > /private/a
 wait_for_sync
 
 # Verify invalid record was not added (should fail to resolve)
-result=$(dig @localhost invalid.argus.com A +short 2>/dev/null || echo "NO_RESULT")
+result=$(dig @localhost -p "$HOST_DNS_PORT" invalid.argus.com A +short 2>/dev/null || echo "NO_RESULT")
 if [ "$result" = "NO_RESULT" ] || [ -z "$result" ]; then
     echo "✓ Invalid IP correctly ignored"
 else

src/bind/tests/scripts/03_reload_test.sh

@@ -5,10 +5,13 @@
 
 set -e
 
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_DIR="$(dirname "$SCRIPT_DIR")"
 
 echo "=== DNS Configuration Reload Test ==="
+echo "Using DNS server localhost:${HOST_DNS_PORT}"
 
 # Check if container is running
 if ! docker compose ps | grep -q "Up"; then
@@ -32,7 +35,7 @@ test_dns_query() {
     echo "Testing: $description"
     echo "Query: $hostname.argus.com -> Expected: $expected_ip"
 
-    result=$(dig @localhost $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
 
     if [ "$result" = "$expected_ip" ]; then
         echo "✓ $result"

src/bind/tests/scripts/04_persistence_test.sh

@@ -5,10 +5,13 @@
 
 set -e
 
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_DIR="$(dirname "$SCRIPT_DIR")"
 
 echo "=== Configuration Persistence Test ==="
+echo "Using DNS server localhost:${HOST_DNS_PORT}"
 
 # Check if dig is available
 if ! command -v dig &> /dev/null; then
@@ -25,7 +28,7 @@ test_dns_query() {
     echo "Testing: $description"
     echo "Query: $hostname.argus.com -> Expected: $expected_ip"
 
-    result=$(dig @localhost $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
 
     if [ "$result" = "$expected_ip" ]; then
         echo "✓ $result"

src/bind/tests/scripts/05_cleanup.sh

@@ -7,6 +7,9 @@ set -e
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_DIR="$(dirname "$SCRIPT_DIR")"
+HOST_DNS_PORT="${HOST_DNS_PORT:-1053}"
+
+export HOST_DNS_PORT
 
 # Parse command line arguments
 FULL_CLEANUP=true

src/log/elasticsearch/build/Dockerfile

@@ -3,10 +3,29 @@ FROM docker.elastic.co/elasticsearch/elasticsearch:8.13.4
 # 切换到 root 用户进行系统级安装
 USER root
 
-# 修改elasticsearch用户的UID和GID
+ARG ARGUS_BUILD_UID=2133
-RUN usermod -u 2133 elasticsearch && \
+ARG ARGUS_BUILD_GID=2015
-    groupmod -g 2015 elasticsearch && \
+
-    chown -R elasticsearch:elasticsearch /usr/share/elasticsearch
+ENV ARGUS_BUILD_UID=${ARGUS_BUILD_UID} \
+    ARGUS_BUILD_GID=${ARGUS_BUILD_GID}
+
+# 调整 elasticsearch 用户与用户组 ID 以匹配宿主机配置
+RUN set -eux; \
+    current_gid="$(getent group elasticsearch | awk -F: '{print $3}')"; \
+    if [ -z "$current_gid" ]; then \
+        groupadd -g "${ARGUS_BUILD_GID}" elasticsearch; \
+    elif [ "$current_gid" != "${ARGUS_BUILD_GID}" ]; then \
+        groupmod -g "${ARGUS_BUILD_GID}" elasticsearch; \
+    fi; \
+    if id elasticsearch >/dev/null 2>&1; then \
+        current_uid="$(id -u elasticsearch)"; \
+        if [ "$current_uid" != "${ARGUS_BUILD_UID}" ]; then \
+            usermod -u "${ARGUS_BUILD_UID}" elasticsearch; \
+        fi; \
+    else \
+        useradd -m -u "${ARGUS_BUILD_UID}" -g "${ARGUS_BUILD_GID}" elasticsearch; \
+    fi; \
+    chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" /usr/share/elasticsearch
 
 # 设置构建参数
 ARG USE_INTRANET=false

src/log/fluent-bit/build/start-fluent-bit.sh

@@ -1,47 +1,96 @@
 #!/bin/bash
 set -euo pipefail
 
-echo "[INFO] Starting Fluent Bit setup in Ubuntu container..."
+echo "[INFO] Starting Fluent Bit setup in Ubuntu container (offline-first)..."
 
-# 安装必要的工具
-echo "[INFO] Installing required packages..."
 export DEBIAN_FRONTEND=noninteractive
-apt-get update -qq
-apt-get install -y -qq curl
 
-# 解压bundle到/tmp
+# Stage bundle to /tmp (read-only mount under /private)
-echo "[INFO] Extracting fluent-bit bundle..."
+echo "[INFO] Staging fluent-bit bundle..."
-cp -r /private/etc /tmp
+rm -rf /tmp/flb && mkdir -p /tmp/flb
-cp -r /private/packages /tmp
+cp -r /private/etc /tmp/flb/
-cd /tmp
+mkdir -p /tmp/flb/packages
+cp -r /private/packages/* /tmp/flb/packages/ 2>/dev/null || true
 
-# 安装 Fluent Bit 从 deb 包
+# Helper: check and install a local deb if not already satisfied
-echo "[INFO] Installing Fluent Bit from deb package..."
+ensure_lib() {
-dpkg -i /tmp/packages/fluent-bit_3.1.9_amd64.deb || true
+  local soname="$1"; shift
-apt-get install -f -y -qq  # 解决依赖问题
+  local pattern="$1"; shift
+  if ldconfig -p 2>/dev/null | grep -q "$soname"; then
+    echo "[OK] $soname already present"
+    return 0
+  fi
+  local deb="$(ls /tmp/flb/packages/$pattern 2>/dev/null | head -n1 || true)"
+  if [[ -n "$deb" ]]; then
+    echo "[INFO] Installing local dependency: $(basename "$deb")"
+    dpkg -i "$deb" >/dev/null 2>&1 || true
+  else
+    echo "[WARN] Local deb for $soname not found (pattern=$pattern)"
+  fi
+  if ! ldconfig -p 2>/dev/null | grep -q "$soname"; then
+    echo "[WARN] $soname still missing after local install; attempting apt fallback"
+    apt-get update -qq || true
+    case "$soname" in
+      libpq.so.5) apt-get install -y -qq libpq5 || true ;;
+      libyaml-0.so.2) apt-get install -y -qq libyaml-0-2 || true ;;
+    esac
+  fi
+  ldconfig 2>/dev/null || true
+}
+
+# Offline-first: satisfy runtime deps from local debs, fallback to apt only if necessary
+ensure_lib "libpq.so.5" "libpq5_*_amd64.deb"
+ensure_lib "libyaml-0.so.2" "libyaml-0-2_*_amd64.deb"
+ensure_lib "libsasl2.so.2" "libsasl2-2_*_amd64.deb"
+ensure_lib "libldap-2.5.so.0" "libldap-2.5-0_*_amd64.deb"
+
+# Install fluent-bit main package from local bundle
+FLB_DEB="$(ls /tmp/flb/packages/fluent-bit_*_amd64.deb 2>/dev/null | head -n1 || true)"
+if [[ -z "$FLB_DEB" ]]; then
+  echo "[ERROR] fluent-bit deb not found under /private/packages" >&2
+  exit 1
+fi
+echo "[INFO] Installing Fluent Bit: $(basename "$FLB_DEB")"
+dpkg -i "$FLB_DEB" >/dev/null 2>&1 || true
+
+# If dpkg reported unresolved dependencies, try apt -f only as last resort
+if ! command -v /opt/fluent-bit/bin/fluent-bit >/dev/null 2>&1; then
+  echo "[WARN] fluent-bit binary missing after dpkg; attempting apt --fix-broken"
+  apt-get install -f -y -qq || true
+fi
+
+# Ensure runtime library dependencies are satisfied (libsasl2, libldap are required via libpq/curl)
+MISSING=$(ldd /opt/fluent-bit/bin/fluent-bit 2>/dev/null | awk '/not found/{print $1}' | xargs -r echo || true)
+if [[ -n "$MISSING" ]]; then
+  echo "[WARN] missing shared libs: $MISSING"
+  apt-get update -qq || true
+  apt-get install -y -qq libsasl2-2 libldap-2.5-0 || true
+  apt-get install -f -y -qq || true
+fi
 
-# 验证 Fluent Bit 可以运行
 echo "[INFO] Fluent Bit version:"
-/opt/fluent-bit/bin/fluent-bit --version
+/opt/fluent-bit/bin/fluent-bit --version || { echo "[ERROR] fluent-bit not installed or libraries missing" >&2; exit 1; }
 
-# 创建配置目录
+# Place configuration
 mkdir -p /etc/fluent-bit
-cp -r /tmp/etc/* /etc/fluent-bit/
+cp -r /tmp/flb/etc/* /etc/fluent-bit/
 
-# 创建日志和缓冲区目录
+# Create logs/buffers dirs
 mkdir -p /logs/train /logs/infer /buffers
 chmod 755 /logs/train /logs/infer /buffers
 
-# 等待 Elasticsearch 就绪
+# Wait for Elasticsearch via bash /dev/tcp to avoid curl dependency
-echo "[INFO] Waiting for Elasticsearch to be ready..."
+echo "[INFO] Waiting for Elasticsearch to be ready (tcp ${ES_HOST}:${ES_PORT})..."
-while ! curl -fs http://${ES_HOST}:${ES_PORT}/_cluster/health >/dev/null 2>&1; do
+for i in $(seq 1 120); do
-    echo "  Waiting for ES at ${ES_HOST}:${ES_PORT}..."
+  if exec 3<>/dev/tcp/${ES_HOST}/${ES_PORT}; then
-    sleep 5
+    exec 3<&- 3>&-
+    echo "[INFO] Elasticsearch is ready"
+    break
+  fi
+  [[ $i -eq 120 ]] && { echo "[ERROR] ES not reachable" >&2; exit 1; }
+  sleep 1
 done
-echo "[INFO] Elasticsearch is ready"
 
-# 启动 Fluent Bit
 echo "[INFO] Starting Fluent Bit with configuration from /etc/fluent-bit/"
 echo "[INFO] Command: /opt/fluent-bit/bin/fluent-bit --config=/etc/fluent-bit/fluent-bit.conf"
-exec /opt/fluent-bit/bin/fluent-bit \
+exec /opt/fluent-bit/bin/fluent-bit --config=/etc/fluent-bit/fluent-bit.conf
-    --config=/etc/fluent-bit/fluent-bit.conf

src/log/kibana/build/Dockerfile

@@ -3,10 +3,29 @@ FROM docker.elastic.co/kibana/kibana:8.13.4
 # 切换到 root 用户进行系统级安装
 USER root
 
-# 修改kibana用户的UID和GID
+ARG ARGUS_BUILD_UID=2133
-RUN usermod -u 2133 kibana && \
+ARG ARGUS_BUILD_GID=2015
-    groupmod -g 2015 kibana && \
+
-    chown -R kibana:kibana /usr/share/kibana
+ENV ARGUS_BUILD_UID=${ARGUS_BUILD_UID} \
+    ARGUS_BUILD_GID=${ARGUS_BUILD_GID}
+
+# 调整 kibana 用户与用户组 ID 以匹配宿主机配置
+RUN set -eux; \
+    current_gid="$(getent group kibana | awk -F: '{print $3}')"; \
+    if [ -z "$current_gid" ]; then \
+        groupadd -g "${ARGUS_BUILD_GID}" kibana; \
+    elif [ "$current_gid" != "${ARGUS_BUILD_GID}" ]; then \
+        groupmod -g "${ARGUS_BUILD_GID}" kibana; \
+    fi; \
+    if id kibana >/dev/null 2>&1; then \
+        current_uid="$(id -u kibana)"; \
+        if [ "$current_uid" != "${ARGUS_BUILD_UID}" ]; then \
+            usermod -u "${ARGUS_BUILD_UID}" kibana; \
+        fi; \
+    else \
+        useradd -m -u "${ARGUS_BUILD_UID}" -g "${ARGUS_BUILD_GID}" kibana; \
+    fi; \
+    chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" /usr/share/kibana
 
 # 设置构建参数
 ARG USE_INTRANET=false

src/log/kibana/build/kibana-post-start.sh

@@ -2,7 +2,7 @@
 set -euo pipefail
 
 ES_HOST="${ELASTICSEARCH_HOSTS:-http://es:9200}"
-KB_HOST="http://localhost:5601"
+KB_HOST="${KB_HOST:-http://127.0.0.1:5601}"
 
 echo "[INFO] Starting Kibana post-start configuration..."
 
@@ -83,50 +83,37 @@ fix_replicas_idempotent() {
 }
 
 # 幂等创建数据视图
+create_or_ensure_data_view() {
+    local name="$1"
+    local title="$2"
+
+    local list_response
+    list_response=$(curl -fsS "$KB_HOST/api/data_views" -H 'kbn-xsrf: true' 2>/dev/null || echo "")
+
+    if [ -z "$list_response" ]; then
+        echo "[WARN] Failed to list data views, skipping creation check for $title"
+        return
+    fi
+
+    if echo "$list_response" | grep -Fq "\"title\":\"$title\""; then
+        echo "[INFO] Data view $title already exists, skipping"
+        return
+    fi
+
+    echo "[INFO] Creating data view for $title indices (allowNoIndex)"
+
+    curl -fsS -X POST "$KB_HOST/api/data_views/data_view?allowNoIndex=true" \
+        -H 'kbn-xsrf: true' \
+        -H 'Content-Type: application/json' \
+        -d "{\"data_view\":{\"name\":\"$name\",\"title\":\"$title\",\"timeFieldName\":\"@timestamp\",\"allowNoIndex\":true}}" \
+        >/dev/null && echo "[OK] Created $name data view" || echo "[WARN] Failed to create $name data view"
+}
+
 create_data_views_idempotent() {
     echo "[INFO] Checking and creating data views..."
 
-    # 检查是否存在匹配的索引
+    create_or_ensure_data_view "train" "train-*"
-    local train_indices=$(curl -s "$ES_HOST/_cat/indices/train-*?h=index" 2>/dev/null | wc -l || echo "0")
+    create_or_ensure_data_view "infer" "infer-*"
-    local infer_indices=$(curl -s "$ES_HOST/_cat/indices/infer-*?h=index" 2>/dev/null | wc -l || echo "0")
-
-    # 创建 train 数据视图
-    if [ "$train_indices" -gt 0 ]; then
-        # 检查数据视图是否已存在
-        local train_exists=$(curl -s "$KB_HOST/api/data_views" -H 'kbn-xsrf: true' 2>/dev/null | grep '"title":"train-\*"' | wc -l )
-
-        if [ "$train_exists" -eq 0 ]; then
-            echo "[INFO] Creating data view for train-* indices"
-            curl -fsS -X POST "$KB_HOST/api/data_views/data_view" \
-                -H 'kbn-xsrf: true' \
-                -H 'Content-Type: application/json' \
-                -d '{"data_view":{"name":"train","title":"train-*","timeFieldName":"@timestamp"}}' \
-                >/dev/null && echo "[OK] Created train data view" || echo "[WARN] Failed to create train data view"
-        else
-            echo "[INFO] Train data view already exists, skipping"
-        fi
-    else
-        echo "[INFO] No train-* indices found, skipping train data view creation"
-    fi
-
-    # 创建 infer 数据视图
-    if [ "$infer_indices" -gt 0 ]; then
-        # 检查数据视图是否已存在
-        local infer_exists=$(curl -s "$KB_HOST/api/data_views" -H 'kbn-xsrf: true' 2>/dev/null | grep '"title":"infer-\*"' | wc -l )
-
-        if [ "$infer_exists" -eq 0 ]; then
-            echo "[INFO] Creating data view for infer-* indices"
-            curl -fsS -X POST "$KB_HOST/api/data_views/data_view" \
-                -H 'kbn-xsrf: true' \
-                -H 'Content-Type: application/json' \
-                -d '{"data_view":{"name":"infer","title":"infer-*","timeFieldName":"@timestamp"}}' \
-                >/dev/null && echo "[OK] Created infer data view" || echo "[WARN] Failed to create infer data view"
-        else
-            echo "[INFO] Infer data view already exists, skipping"
-        fi
-    else
-        echo "[INFO] No infer-* indices found, skipping infer data view creation"
-    fi
 }
 
 # 主逻辑

src/log/tests/docker-compose.yml

@@ -17,6 +17,7 @@ services:
       interval: 10s
       timeout: 5s
       retries: 30
+    restart: always
 
   kibana:
     build:
@@ -73,13 +74,11 @@ services:
       interval: 15s
       timeout: 10s
       retries: 30
+    restart: always
 
   bind9:
       image: argus-bind9:latest
-      ports:
-        - "53:53/tcp"
-        - "53:53/udp"
       volumes:
         - ./private/argus:/private/argus/
-      restart: unless-stopped
+      restart: always
 

src/log/tests/scripts/01_bootstrap.sh

@@ -1,6 +1,10 @@
 #!/usr/bin/env bash
 set -euo pipefail
 root="$(cd "$(dirname "${BASH_SOURCE[0]}")/../" && pwd)"
+project_root="$(cd "$root/../../.." && pwd)"
+
+source "$project_root/scripts/common/build_user.sh"
+load_build_user
 
 # 创建新的private目录结构 (基于argus目录结构)
 echo "[INFO] Creating private directory structure for supervisor-based containers..."
@@ -11,9 +15,9 @@ mkdir -p "$root/private/argus/etc/"
 
 # 设置数据目录权限（ES 和 Kibana 容器都使用 UID 1000）
 echo "[INFO] Setting permissions for data directories..."
-sudo chown -R 2133:2015 "$root/private/argus/log/elasticsearch" 2>/dev/null || true
+chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" "$root/private/argus/log/elasticsearch" 2>/dev/null || true
-sudo chown -R 2133:2015 "$root/private/argus/log/kibana" 2>/dev/null || true
+chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" "$root/private/argus/log/kibana" 2>/dev/null || true
-sudo chown -R 2133:2015 "$root/private/argus/etc" 2>/dev/null || true
+chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" "$root/private/argus/etc" 2>/dev/null || true
 
 echo "[INFO] Supervisor-based containers will manage their own scripts and configurations"
 
@@ -28,3 +32,42 @@ fi
 
 echo "[OK] 初始化完成: private/argus/log/{elasticsearch,kibana}"
 echo "[INFO] Fluent-bit files should be in fluent-bit/ directory"
+
+# 准备 Fluent Bit 离线依赖（从 metric all-in-one-full 复制 deb 到 ../fluent-bit/build/packages）
+FLB_BUILD_PACKAGES_DIR="$root/../fluent-bit/build/packages"
+mkdir -p "$FLB_BUILD_PACKAGES_DIR"
+for deb in \
+  "$project_root/src/metric/client-plugins/all-in-one-full/plugins/fluent-bit/bin/libyaml-0-2_"*_amd64.deb \
+  "$project_root/src/metric/client-plugins/all-in-one-full/plugins/fluent-bit/bin/libpq5_"*_amd64.deb ; do
+  if ls $deb >/dev/null 2>&1; then
+    for f in $deb; do
+      base="$(basename "$f")"
+      if [[ ! -f "$FLB_BUILD_PACKAGES_DIR/$base" ]]; then
+        cp "$f" "$FLB_BUILD_PACKAGES_DIR/"
+        echo "  [+] copied $base"
+      fi
+    done
+  fi
+done
+
+# 额外：从 all-in-one-full 的 ubuntu22/curl.tar.gz 解包必要依赖（libsasl2/ldap），便于离线安装
+CURLOPT_TAR="$project_root/src/metric/client-plugins/all-in-one-full/deps/ubuntu22/curl.tar.gz"
+if [[ -f "$CURLOPT_TAR" ]]; then
+  tmpdir=$(mktemp -d)
+  if tar -xzf "$CURLOPT_TAR" -C "$tmpdir" 2>/dev/null; then
+    for p in \
+      libsasl2-2_*_amd64.deb \
+      libsasl2-modules-db_*_amd64.deb \
+      libldap-2.5-0_*_amd64.deb \
+      libidn2-0_*_amd64.deb \
+      libbrotli1_*_amd64.deb \
+      libssl3_*_amd64.deb ; do
+      src=$(ls "$tmpdir"/curl/$p 2>/dev/null | head -n1 || true)
+      if [[ -n "$src" ]]; then
+        base="$(basename "$src")"
+        [[ -f "$FLB_BUILD_PACKAGES_DIR/$base" ]] || cp "$src" "$FLB_BUILD_PACKAGES_DIR/" && echo "  [+] staged $base"
+      fi
+    done
+  fi
+  rm -rf "$tmpdir"
+fi

src/log/tests/scripts/03_send_test_host01.sh

@@ -4,8 +4,22 @@ set -euo pipefail
 # 获取fluent-bit-host01容器名称
 container_name="logging-mvp-fluent-bit-host01-1"
 
-# 检查容器是否存在并运行
+wait_for_container() {
-if ! docker ps | grep -q "$container_name"; then
+    local name="$1"
+    local attempts=30
+    local delay=5
+    local i
+    for ((i = 1; i <= attempts; i++)); do
+        if docker ps --format '{{.Names}}' | grep -Fx "$name" >/dev/null; then
+            return 0
+        fi
+        echo "[INFO] 等待容器 $name 启动中... ($i/$attempts)"
+        sleep "$delay"
+    done
+    return 1
+}
+
+if ! wait_for_container "$container_name"; then
     echo "[ERROR] Fluent Bit容器 $container_name 未运行"
     exit 1
 fi

src/log/tests/scripts/03_send_test_host02.sh

@@ -4,8 +4,22 @@ set -euo pipefail
 # 获取fluent-bit-host02容器名称
 container_name="logging-mvp-fluent-bit-host02-1"
 
-# 检查容器是否存在并运行
+wait_for_container() {
-if ! docker ps | grep -q "$container_name"; then
+    local name="$1"
+    local attempts=30
+    local delay=5
+    local i
+    for ((i = 1; i <= attempts; i++)); do
+        if docker ps --format '{{.Names}}' | grep -Fx "$name" >/dev/null; then
+            return 0
+        fi
+        echo "[INFO] 等待容器 $name 启动中... ($i/$attempts)"
+        sleep "$delay"
+    done
+    return 1
+}
+
+if ! wait_for_container "$container_name"; then
     echo "[ERROR] Fluent Bit容器 $container_name 未运行"
     exit 1
 fi

src/log/tests/scripts/04_query_es.sh

@@ -1,7 +1,42 @@
 #!/usr/bin/env bash
 set -euo pipefail
+
+# ES endpoint and wait strategy
 ES="${ES:-http://localhost:9200}"
+es_wait_attempts="${ES_WAIT_ATTEMPTS:-60}"      # total attempts to wait for ES
+es_wait_interval="${ES_WAIT_INTERVAL:-2}"        # seconds between attempts
+
 echo "[i] 查询 ES 端点：$ES"
+
+wait_for_es() {
+  local attempt=1
+  while (( attempt <= es_wait_attempts )); do
+    # 等待集群达到至少 yellow 状态；请求失败则重试
+    if curl -fsS "$ES/_cluster/health?wait_for_status=yellow&timeout=1s" >/dev/null 2>&1; then
+      echo "[ok] Elasticsearch 已就绪 (attempt=${attempt}/${es_wait_attempts})"
+      return 0
+    fi
+    echo "[..] 等待 Elasticsearch 可用中 (${attempt}/${es_wait_attempts})"
+    sleep "${es_wait_interval}"
+    (( attempt++ ))
+  done
+  echo "[err] Elasticsearch 在 ${es_wait_attempts} 次尝试后仍不可用"
+  return 1
+}
+
+safe_count() {
+  # 对缺失索引返回 0，避免 404 触发失败
+  local pattern="$1"
+  local json
+  json=$(curl -fsS "$ES/${pattern}/_count?ignore_unavailable=true&allow_no_indices=true" 2>/dev/null || echo '{}')
+  echo "$json" | sed -E 's/.*"count":([0-9]+).*/\1/' | awk 'NF{print $0;exit} END{if(NR==0)print 0}'
+}
+
+wait_for_es
+
+# 列出相关索引（可能为空，允许）
 curl -fsS "$ES/_cat/indices?v" | egrep 'train-|infer-|logstash' || true
-printf "train-* 计数："; curl -fsS "$ES/train-*/_count" | sed -E 's/.*"count":([0-9]+).*/\1/'; echo
+
-printf "infer-* 计数："; curl -fsS "$ES/infer-*/_count" | sed -E 's/.*"count":([0-9]+).*/\1/'; echo
+# 打印计数，缺失索引按 0 处理
+printf "train-* 计数："; safe_count "train-*"; echo
+printf "infer-* 计数："; safe_count "infer-*"; echo

src/log/tests/scripts/e2e_test.sh

@@ -19,7 +19,7 @@ get_log_count() {
 # 函数：等待服务就绪
 wait_for_services() {
     echo "[INFO] Waiting for all services to be ready..."
-    local max_attempts=60
+    local max_attempts=${SERVICE_WAIT_ATTEMPTS:-120}
     local attempt=1
 
     while [ $attempt -le $max_attempts ]; do
@@ -115,20 +115,32 @@ show_step "Health" "Check service health"
 echo "[INFO] Checking service health..."
 
 # 检查 Elasticsearch 健康状态
+health_check_ok=1
 es_health=$(curl -s "http://localhost:9200/_cluster/health" | grep -o '"status":"[^"]*"' | cut -d'"' -f4)
 if [ "$es_health" = "green" ] || [ "$es_health" = "yellow" ]; then
     echo "✅ Elasticsearch health: $es_health"
 else
     echo "❌ Elasticsearch health: $es_health"
+    health_check_ok=0
 fi
 
 # 检查 Kibana 状态
 if curl -fs "http://localhost:5601/api/status" >/dev/null 2>&1; then
     kb_status="available"
     echo "✅ Kibana status: $kb_status"
+
+    data_views_json=$(curl -fs "http://localhost:5601/api/data_views" -H 'kbn-xsrf: true' 2>/dev/null || true)
+    if echo "$data_views_json" | grep -F '"title":"train-*"' >/dev/null 2>&1 && \
+       echo "$data_views_json" | grep -F '"title":"infer-*"' >/dev/null 2>&1; then
+        echo "✅ Kibana data views: train-* and infer-* present"
+    else
+        echo "❌ Kibana data views missing: train-* or infer-*"
+        health_check_ok=0
+    fi
 else
     kb_status="unavailable"
     echo "⚠️  Kibana status: $kb_status"
+    health_check_ok=0
 fi
 
 # 检查 Fluent-Bit 指标
@@ -139,6 +151,13 @@ if [ "$fb_host01_uptime" -gt 0 ] && [ "$fb_host02_uptime" -gt 0 ]; then
     echo "✅ Fluent-Bit services: host01 uptime=${fb_host01_uptime}s, host02 uptime=${fb_host02_uptime}s"
 else
     echo "⚠️  Fluent-Bit services: host01 uptime=${fb_host01_uptime}s, host02 uptime=${fb_host02_uptime}s"
+    health_check_ok=0
+fi
+
+if [ "$health_check_ok" -eq 1 ]; then
+    true
+else
+    false
 fi
 
 verify_step "Service health check"

src/master/Dockerfile

@@ -5,6 +5,11 @@ SHELL ["/bin/bash", "-c"]
 ARG PIP_INDEX_URL=
 ARG USE_OFFLINE=0
 ARG USE_INTRANET=false
+ARG ARGUS_BUILD_UID=2133
+ARG ARGUS_BUILD_GID=2015
+
+ENV ARGUS_BUILD_UID=${ARGUS_BUILD_UID} \
+    ARGUS_BUILD_GID=${ARGUS_BUILD_GID}
 
 ENV PIP_NO_CACHE_DIR=1 \
     PYTHONUNBUFFERED=1 \
@@ -36,11 +41,12 @@ RUN if [[ "$USE_INTRANET" == "true" ]]; then \
         if [[ -f /etc/apt/sources.list ]]; then cp /etc/apt/sources.list /etc/apt/sources.list.bak; fi && \
         mkdir -p /etc/apt && \
         echo "deb [trusted=yes] http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        rm -rf /etc/apt/sources.list.d && \
         echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
         echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
     fi && \
     apt-get update && \
-    apt-get install -y supervisor net-tools inetutils-ping vim && \
+    apt-get install -y supervisor net-tools inetutils-ping && \
     apt-get clean && \
     rm -rf /var/lib/apt/lists/*
 
@@ -51,6 +57,18 @@ RUN if [[ "$USE_INTRANET" == "true" ]]; then \
 
 RUN mkdir -p /var/log/supervisor
 
+RUN set -eux; \
+    if getent group argus >/dev/null; then \
+        groupmod -g "${ARGUS_BUILD_GID}" argus; \
+    else \
+        groupadd -g "${ARGUS_BUILD_GID}" argus; \
+    fi; \
+    if id argus >/dev/null 2>&1; then \
+        usermod -u "${ARGUS_BUILD_UID}" -g "${ARGUS_BUILD_GID}" argus; \
+    else \
+        useradd -m -u "${ARGUS_BUILD_UID}" -g "${ARGUS_BUILD_GID}" -s /bin/bash argus; \
+    fi
+
 COPY ./src/master/build/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
 COPY ./src/master/build/start-master.sh /usr/local/bin/start-master.sh
 COPY ./src/master/build/dns-monitor.sh /usr/local/bin/dns-monitor.sh

src/master/README.md

@@ -11,7 +11,7 @@ Argus Master 是基于 Flask + SQLite 的节点管理服务，负责：
 
 ```bash
 cd src/master
-./scripts/build_images.sh            # 生成 argus-master:dev 镜像
+./scripts/build_images.sh            # 生成 argus-master:latest 镜像
 ```
 
 如需离线构建，先在有网环境运行准备脚本：
@@ -25,7 +25,7 @@ cd src/master
 
 ```bash
 cd src/master
-./scripts/build_images.sh --offline --tag argus-master:dev
+./scripts/build_images.sh --offline --tag argus-master:latest
 ```
 
 若内网缺少 `python:3.11-slim`，请提前在外网 `docker save` 后通过离线介质 `docker load`。

src/master/build/start-master.sh

@@ -5,8 +5,17 @@ set -euo pipefail
 DNS_DIR="/private/argus/etc"
 DNS_SCRIPT="${DNS_DIR}/update-dns.sh"
 MASTER_DOMAIN_FILE="${DNS_DIR}/master.argus.com"
+RUNTIME_USER="${ARGUS_RUNTIME_USER:-argus}"
+RUNTIME_UID="${ARGUS_BUILD_UID:-2133}"
+RUNTIME_GID="${ARGUS_BUILD_GID:-2015}"
+MASTER_DATA_DIR="/private/argus/master"
+METRIC_DIR="/private/argus/metric/prometheus"
 
 mkdir -p "$DNS_DIR"
+chown -R "$RUNTIME_UID:$RUNTIME_GID" "$DNS_DIR" 2>/dev/null || true
+mkdir -p "$MASTER_DATA_DIR"
+mkdir -p "$METRIC_DIR"
+chown -R "$RUNTIME_UID:$RUNTIME_GID" "$MASTER_DATA_DIR" "$METRIC_DIR" 2>/dev/null || true
 
 if [[ -x "$DNS_SCRIPT" ]]; then
   echo "[INFO] Running update-dns.sh before master starts"
@@ -21,6 +30,7 @@ MASTER_IP=$(ifconfig | grep -A 1 eth0 | grep inet | awk '{print $2}' || true)
 if [[ -n "${MASTER_IP}" ]]; then
   echo "current IP: ${MASTER_IP}"
   echo "${MASTER_IP}" > "$MASTER_DOMAIN_FILE"
+  chown "$RUNTIME_UID:$RUNTIME_GID" "$MASTER_DOMAIN_FILE" 2>/dev/null || true
 else
   echo "[WARN] Failed to detect master IP via ifconfig"
 fi
@@ -35,4 +45,15 @@ else
   EXTRA_ARRAY=()
 fi
 
-exec gunicorn --bind "$BIND_ADDR" --workers "$WORKERS" "${EXTRA_ARRAY[@]}" "app:create_app()"
+command=(gunicorn --bind "$BIND_ADDR" --workers "$WORKERS")
+if [[ ${#EXTRA_ARRAY[@]} -gt 0 ]]; then
+  command+=("${EXTRA_ARRAY[@]}")
+fi
+command+=("app:create_app()")
+
+if command -v runuser >/dev/null 2>&1; then
+  exec runuser -u "$RUNTIME_USER" -- "${command[@]}"
+else
+  printf -v _cmd_str '%q ' "${command[@]}"
+  exec su -s /bin/bash -m "$RUNTIME_USER" -c "exec ${_cmd_str}"
+fi

src/master/scripts/build_images.sh

@@ -3,22 +3,28 @@ set -euo pipefail
 
 usage() {
   cat >&2 <<'USAGE'
-Usage: $0 [--intranet] [--offline] [--tag <image_tag>]
+Usage: $0 [--intranet] [--offline] [--tag <image_tag>] [--no-cache]
 
 Options:
   --intranet           使用指定的 PyPI 镜像源（默认清华镜像）。
   --offline            完全离线构建，依赖 offline_wheels/ 目录中的离线依赖包。
-  --tag <image_tag>    自定义镜像标签，默认 argus-master:dev。
+  --tag <image_tag>    自定义镜像标签，默认 argus-master:latest。
+  --no-cache           不使用 Docker 构建缓存。
 USAGE
 }
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 PROJECT_ROOT="$(cd "$SCRIPT_DIR/../../.." && pwd)"
 MODULE_ROOT="$PROJECT_ROOT/src/master"
-IMAGE_TAG="${IMAGE_TAG:-argus-master:dev}"
+IMAGE_TAG="${IMAGE_TAG:-argus-master:latest}"
 DOCKERFILE="src/master/Dockerfile"
 BUILD_ARGS=()
 OFFLINE_MODE=0
+NO_CACHE=0
+
+source "$PROJECT_ROOT/scripts/common/build_user.sh"
+load_build_user
+BUILD_ARGS+=("--build-arg" "ARGUS_BUILD_UID=${ARGUS_BUILD_UID}" "--build-arg" "ARGUS_BUILD_GID=${ARGUS_BUILD_GID}")
 
 cd "$PROJECT_ROOT"
 
@@ -41,6 +47,11 @@ while [[ "$#" -gt 0 ]]; do
       IMAGE_TAG="$2"
       shift 2
       ;;
+    --no-cache)
+      NO_CACHE=1
+      BUILD_ARGS+=("--no-cache")
+      shift
+      ;;
     -h|--help)
       usage
       exit 0

src/master/scripts/save_images.sh

@@ -8,7 +8,7 @@ usage() {
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 DEFAULT_OUTPUT="$PROJECT_ROOT/images/argus-master-dev.tar"
-IMAGE_TAG="${IMAGE_TAG:-argus-master:dev}"
+IMAGE_TAG="${IMAGE_TAG:-argus-master:latest}"
 OUTPUT_PATH="$DEFAULT_OUTPUT"
 
 while [[ "$#" -gt 0 ]]; do

src/master/tests/docker-compose.yml

@@ -1,6 +1,6 @@
 services:
   master:
-    image: ${MASTER_IMAGE_TAG:-argus-master:dev}
+    image: ${MASTER_IMAGE_TAG:-argus-master:latest}
     container_name: argus-master-e2e
     environment:
       - OFFLINE_THRESHOLD_SECONDS=6

src/master/tests/scripts/00_e2e_test.sh

@@ -17,7 +17,7 @@ SCRIPTS=(
 
 for script in "${SCRIPTS[@]}"; do
   echo "[TEST] Running $script"
-  MASTER_IMAGE_TAG="${MASTER_IMAGE_TAG:-argus-master:dev}" "$SCRIPT_DIR/$script"
+  MASTER_IMAGE_TAG="${MASTER_IMAGE_TAG:-argus-master:latest}" "$SCRIPT_DIR/$script"
   echo "[TEST] $script completed"
   echo
 done

src/master/tests/scripts/01_up_master.sh

@@ -44,7 +44,7 @@ fi
 
 pushd "$TEST_ROOT" >/dev/null
 compose down --remove-orphans || true
-MASTER_IMAGE_TAG="${MASTER_IMAGE_TAG:-argus-master:dev}" compose up -d
+MASTER_IMAGE_TAG="${MASTER_IMAGE_TAG:-argus-master:latest}" compose up -d
 popd >/dev/null
 
 echo "[INFO] Master container is up on http://localhost:31300"

src/metric/.gitignore

@@ -0,0 +1,7 @@
+/prometheus/data/
+/client-plugins/dcgm-exporter-installer/
+/client-plugins/demo-all-in-one/artifact/
+/client-plugins/demo-all-in-one/publish/
+/client-plugins/demo-all-in-one/checklist
+/client-plugins/demo-all-in-one/VERSION
+/client-plugins/all-in-one-full/artifact/

src/metric/client-plugins/all-in-one-demo/README.md

@@ -0,0 +1,65 @@
+# 客户侧组件安装包构建、发布流程
+
+## 第一步：配置版本和组件
+
+首先搞定配置文件：
+
+1. 把 `.checklist.example` 重命名成 `checklist`
+2. 把 `.VERSION.example` 重命名成 `VERSION`
+
+### checklist 文件格式
+```
+# 组件名称 目录路径 版本号 [依赖组件] [安装顺序]
+dcgm-exporter-installer /path/to/dcgm-exporter-installer 1.1.0
+node-exporter-installer /path/to/node-exporter-installer 1.1.0
+```
+
+### VERSION 文件
+设置需要发布的版本号，比如 `1.29.0`
+
+> 建议用 `version-manager.sh` 来管理版本
+
+## 第二步：构建安装包
+
+直接跑脚本：
+```bash
+./package_artifact.sh
+```
+
+构建完的东西会放在 `artifact/` 目录下，按版本分文件夹。
+
+如果版本已经存在了，想要覆盖重新构建：
+```bash
+./package_artifact.sh --force
+```
+
+构建完可以手工测试安装包。
+
+## 第三步：发布安装包
+
+用这个脚本发布：
+```bash
+./publish_artifact.sh
+```
+
+发布后的内容在 `publish/` 目录里，包含：
+- 压缩版本的安装包
+- 一键安装的bash脚本
+
+## 第四步：部署到FTP服务器
+
+把发布的内容上传到FTP服务器，客户端就可以通过一键命令安装：
+
+```bash
+curl -fsSL 'ftp://{$USER}:{$PASSWD}@{$your-ftp-server}/setup.sh' -o setup.sh
+
+# root用户直接执行，非root用户需要使用sudo
+chmod +x setup.sh
+bash setup.sh --server {$your-ftp-server} --user {$USER} --password {$PASSWD}
+
+示例：
+curl -fsS 'ftp://ftpuser:ZGClab1234!@177.177.70.200/setup.sh' -o setup.sh
+chmod +x setup.sh
+bash setup.sh --server {$域名} --user ftpuser --password 'ZGClab1234!'
+
+```

src/metric/client-plugins/all-in-one-demo/config/.VERSION.example

@@ -0,0 +1 @@
+1.29.0

src/metric/client-plugins/all-in-one-demo/config/.checklist.example

@@ -0,0 +1,3 @@
+# 组件名称 目录路径 版本号 [依赖组件] [安装顺序]
+dcgm-exporter-installer /Users/sundapeng/Project/nlp/aiops/client-plugins/dcgm-exporter-installer 1.1.0
+node-exporter-installer /Users/sundapeng/Project/nlp/aiops/client-plugins/node-exporter-installer 1.1.0

src/metric/client-plugins/all-in-one-demo/config/.config.env.example

@@ -0,0 +1,8 @@
+# Argus Metric 配置文件示例
+# 复制此文件为 config.env 并根据需要修改配置
+
+# 连接master服务
+MASTER_ENDPOINT=master.argus.com:3000
+
+# 上报状态间隔描述（秒）
+REPORT_INTERVAL_SECONDS=60

src/metric/client-plugins/all-in-one-demo/config/config.env

@@ -0,0 +1,3 @@
+# Elasticsearch
+ES_HOST=es.log.argus.com
+ES_PORT=9200

src/metric/client-plugins/all-in-one-demo/config/dns.conf.example

@@ -0,0 +1 @@
+177.177.17.106

src/metric/client-plugins/all-in-one-demo/plugins/node-exporter/check_health.sh

@@ -0,0 +1,55 @@
+#!/bin/bash
+
+# Node Exporter 健康检查脚本
+# 输出 JSON 格式结果
+
+set -e
+
+# 检查 Node Exporter 健康状态
+check_health() {
+    local url="http://localhost:9100"
+    local metrics_url="$url/metrics"
+    local name="node-exporter"
+    local status="unhealth"
+    local reason=""
+    
+    # 检查 curl 是否可用
+    if ! command -v curl &> /dev/null; then
+        reason="curl 命令不可用，无法进行健康检查"
+        echo "{\"name\": \"$name\", \"status\": \"$status\", \"reason\": \"$reason\"}"
+        exit 1
+    fi
+    
+    # 测试根路径连接
+    local http_code=$(curl -s -o /dev/null -w "%{http_code}" "$url" 2>/dev/null || echo "000")
+    
+    if [[ "$http_code" == "200" ]]; then
+        # 测试 metrics 端点
+        local metrics_code=$(curl -s -o /dev/null -w "%{http_code}" "$metrics_url" 2>/dev/null || echo "000")
+        
+        if [[ "$metrics_code" == "200" ]]; then
+            status="health"
+            reason="success"
+            echo "{\"name\": \"$name\", \"status\": \"$status\", \"reason\": \"$reason\"}"
+            exit 0
+        else
+            reason="Metrics 端点异常 (HTTP $metrics_code)"
+            echo "{\"name\": \"$name\", \"status\": \"$status\", \"reason\": \"$reason\"}"
+            exit 1
+        fi
+    else
+        reason="HTTP 服务异常 (HTTP $http_code)，请检查 Node Exporter 是否正在运行在端口 9100"
+        echo "{\"name\": \"$name\", \"status\": \"$status\", \"reason\": \"$reason\"}"
+        exit 1
+    fi
+}
+
+# 主函数
+main() {
+    check_health
+}
+
+# 脚本入口
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    main "$@"
+fi

src/metric/client-plugins/all-in-one-demo/plugins/node-exporter/install.sh

@@ -0,0 +1,343 @@
+#!/bin/bash
+
+set -e
+
+# 颜色定义
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# 日志函数
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1"
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1"
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+# 更新安装记录
+update_install_record() {
+    local pid="$1"
+    # 使用传入的安装目录参数，如果没有则使用默认值
+    local install_base_dir="${2:-/opt/argus-metric/current}"
+    local install_record="$install_base_dir/.install_record"
+    
+    # 如果安装记录文件不存在，说明是首次安装，由主安装脚本统一创建
+    if [[ ! -f "$install_record" ]]; then
+        log_info "安装记录文件不存在，将由主安装脚本创建"
+        return 0
+    fi
+    
+    # 如果文件存在，说明是重启场景，只更新 PID 字段
+    if command -v jq &> /dev/null; then
+        # 读取当前 PID
+        local current_pid=$(jq -r '.components."node-exporter".pid // ""' "$install_record" 2>/dev/null)
+        
+        if [[ -z "$current_pid" ]]; then
+            log_warning "无法读取当前 PID，跳过更新"
+            return 1
+        fi
+        
+        # 使用 jq 只更新 pid 字段，保持字符串类型，保留其他字段
+        jq --arg new_pid "$pid" '.components."node-exporter".pid = $new_pid' "$install_record" > "$install_record.tmp" && mv "$install_record.tmp" "$install_record"
+        log_info "PID 已更新: $current_pid -> $pid"
+    else
+        log_warning "jq 命令不可用，无法更新安装记录文件"
+    fi
+}
+
+# 显示帮助信息
+show_help() {
+    echo "Node Exporter 安装脚本"
+    echo
+    echo "用法: $0 [选项]"
+    echo
+    echo "选项:"
+    echo "  --help     显示此帮助信息"
+    echo
+    echo "示例:"
+    echo "  $0          # 安装 Node Exporter"
+    echo
+}
+
+# 解析命令行参数
+INSTALL_DIR=""
+for arg in "$@"; do
+    case $arg in
+        --help|-h)
+            show_help
+            exit 0
+            ;;
+        *)
+            # 如果参数不是以--开头，则认为是安装目录
+            if [[ ! "$arg" =~ ^-- ]]; then
+                INSTALL_DIR="$arg"
+            else
+                log_error "未知参数: $arg"
+                show_help
+                exit 1
+            fi
+            ;;
+    esac
+done
+
+# 检查是否为 root 用户
+check_root() {
+    if [[ $EUID -ne 0 ]]; then
+        log_error "此脚本需要 root 权限运行"
+        log_info "请使用: sudo $0"
+        exit 1
+    fi
+}
+
+# 检查系统要求
+check_system() {
+    log_info "检查系统要求..."
+    
+    # 检查操作系统
+    if [[ ! -f /etc/os-release ]]; then
+        log_error "无法检测操作系统版本"
+        exit 1
+    fi
+    
+    source /etc/os-release
+    log_info "检测到操作系统: $NAME $VERSION"
+    
+    # 检查是否为 Linux 系统
+    if [[ "$ID" != "ubuntu" && "$ID" != "debian" && "$ID" != "centos" && "$ID" != "rhel" && "$ID" != "fedora" ]]; then
+        log_warning "此脚本主要针对常见 Linux 发行版，其他系统可能需要调整"
+    fi
+    
+    # 检查系统架构
+    local arch=$(uname -m)
+    log_info "系统架构: $arch"
+    
+    if [[ "$arch" != "x86_64" && "$arch" != "amd64" ]]; then
+        log_warning "当前架构为 $arch，node_exporter 主要支持 x86_64/amd64"
+    fi
+}
+
+stop_existing_service() {
+    log_info "检查并停止可能运行的 Node Exporter 服务..."
+
+    # 当前脚本 PID，防止误杀
+    SELF_PID=$$
+
+    # 1. 停止 systemd 服务（如果存在）
+    if systemctl list-units --full -all | grep -q "node_exporter.service"; then
+        log_info "检测到 systemd 服务 node_exporter，正在停止..."
+        systemctl stop node_exporter || true
+        systemctl disable node_exporter || true
+    fi
+
+    # 2. 清理可能存在的 PID 文件
+    for pid_file in /var/run/node-exporter.pid /var/run/node_exporter.pid /tmp/node_exporter.pid; do
+        if [[ -f "$pid_file" ]]; then
+            local pid=$(cat "$pid_file")
+            if kill -0 "$pid" 2>/dev/null; then
+                log_info "发现 Node Exporter (PID: $pid)，正在停止..."
+                kill "$pid"
+                sleep 2
+                kill -0 "$pid" 2>/dev/null && kill -9 "$pid"
+            fi
+            rm -f "$pid_file"
+        fi
+    done
+
+    # 3. 用 pgrep 查找进程，排除当前脚本
+    local pids=$(pgrep -f "node_exporter|node-exporter|/usr/local/bin/node-exporter" | grep -vw "$SELF_PID" || true)
+    if [[ -n "$pids" ]]; then
+        log_info "发现 Node Exporter 进程 (PID: $pids)，正在停止..."
+        for pid in $pids; do
+            if kill -0 "$pid" 2>/dev/null; then
+                kill "$pid" 2>/dev/null || true
+                sleep 1
+                kill -0 "$pid" 2>/dev/null && kill -9 "$pid" 2>/dev/null || true
+            fi
+        done
+    fi
+
+    # 4. 兜底：检查是否有进程占用 9100 端口
+    local listen_pids=$(lsof -ti:9100 2>/dev/null || true)
+    if [[ -n "$listen_pids" ]]; then
+        log_warning "发现占用 9100 端口的进程 (PID: $listen_pids)，强制终止..."
+        for pid in $listen_pids; do
+            kill -9 "$pid" 2>/dev/null || true
+        done
+        sleep 1
+    fi
+
+    # 5. 最终验证
+    if netstat -tuln 2>/dev/null | grep -q ":9100 "; then
+        log_error "端口 9100 仍被占用，请手动检查"
+        return 1
+    else
+        log_success "旧的 Node Exporter 已完全停止"
+    fi
+}
+
+
+# 安装 Node Exporter 二进制文件
+install_node_exporter() {
+    log_info "安装 Node Exporter..."
+    
+    local binary_file="bin/node_exporter"
+    local install_dir="/usr/local/bin"
+    
+    if [[ ! -f "$binary_file" ]]; then
+        log_error "找不到 Node Exporter 二进制文件: $binary_file"
+        exit 1
+    fi
+    
+    # 停止可能运行的服务
+    stop_existing_service
+    
+    # 复制二进制文件并重命名为统一格式
+    cp "$binary_file" "$install_dir/node-exporter"
+    chmod +x "$install_dir/node-exporter"
+    
+    log_success "Node Exporter 二进制文件安装完成"
+}
+
+# 创建用户和组
+create_user() {
+    log_info "创建 node_exporter 用户..."
+    
+    # 检查用户是否已存在
+    if id "node_exporter" &>/dev/null; then
+        log_info "用户 node_exporter 已存在"
+    else
+        useradd --no-create-home --shell /bin/false node_exporter
+        log_success "用户 node_exporter 创建完成"
+    fi
+}
+
+# 安装配置文件
+install_config() {
+    log_info "安装配置文件..."
+    
+    local config_dir="/etc/node_exporter"
+    
+    # 创建配置目录
+    mkdir -p "$config_dir"
+    
+    # 创建文本文件收集器目录
+    mkdir -p "/var/lib/node_exporter/textfile_collector"
+    chown node_exporter:node_exporter "/var/lib/node_exporter/textfile_collector"
+}
+
+# 启动 Node Exporter 服务
+start_node_exporter() {
+    log_info "启动 Node Exporter 服务..."
+    
+    local binary_path="/usr/local/bin/node-exporter"
+    local log_file="/var/log/node-exporter.log"
+    local pid_file="/var/run/node-exporter.pid"
+    
+    # 检查服务是否已经在运行
+    if [[ -f "$pid_file" ]]; then
+        local pid=$(cat "$pid_file")
+        if kill -0 "$pid" 2>/dev/null; then
+            log_info "Node Exporter 服务已在运行 (PID: $pid)"
+            return 0
+        else
+            log_warning "发现过期的 PID 文件，正在清理..."
+            rm -f "$pid_file"
+        fi
+    fi
+    
+    # 检查端口是否被占用
+    if netstat -tuln 2>/dev/null | grep -q ":9100 "; then
+        log_warning "端口 9100 已被占用，请检查是否有其他服务在运行"
+        return 1
+    fi
+    
+    # 启动服务
+    log_info "正在启动 Node Exporter..."
+    nohup "$binary_path" --web.listen-address=:9100 > "$log_file" 2>&1 &
+    local pid=$!
+    
+    # 保存 PID
+    echo "$pid" > "$pid_file"
+    
+    # 等待服务启动
+    sleep 2
+    
+    # 检查服务是否成功启动
+    if kill -0 "$pid" 2>/dev/null; then
+        log_success "Node Exporter 服务启动成功 (PID: $pid)"
+        log_info "日志文件: $log_file"
+        log_info "PID 文件: $pid_file"
+        
+        # 更新安装记录
+        update_install_record "$pid" "$INSTALL_DIR"
+    else
+        log_error "Node Exporter 服务启动失败"
+        rm -f "$pid_file"
+        return 1
+    fi
+}
+
+
+
+# 显示安装信息
+show_install_info() {
+    log_success "Node Exporter 安装完成！"
+    echo
+    echo "安装信息:"
+    echo "  二进制文件: /usr/local/bin/node-exporter"
+    echo "  运行用户: node_exporter"
+    echo "  配置目录: /etc/node_exporter/"
+    echo "  默认端口: 9100"
+    echo
+    echo "使用方法:"
+    echo "  手动启动: /usr/local/bin/node-exporter --web.listen-address=:9100"
+    echo "  后台启动: nohup /usr/local/bin/node-exporter --web.listen-address=:9100 &"
+    echo
+    echo "测试连接:"
+    echo "  curl http://localhost:9100/metrics"
+    echo "  curl http://localhost:9100"
+    echo
+    echo "Prometheus 配置示例:"
+    echo "  - job_name: 'node_exporter'"
+    echo "    static_configs:"
+    echo "      - targets: ['localhost:9100']"
+    echo
+}
+
+# 主函数
+main() {
+    echo "=========================================="
+    echo "    Node Exporter 安装脚本 v1.0"
+    echo "=========================================="
+    echo
+    
+    check_root
+    check_system
+    
+    log_info "开始安装 Node Exporter..."
+    
+    install_node_exporter
+    create_user
+    install_config
+    start_node_exporter
+    
+    show_install_info
+}
+
+# 脚本入口
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    main "$@"
+fi
+

src/metric/client-plugins/all-in-one-demo/plugins/node-exporter/package.sh

@@ -0,0 +1,87 @@
+#!/bin/bash
+
+set -e
+
+# 颜色定义
+GREEN='\033[0;32m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1"
+}
+
+# 获取当前目录
+CURRENT_DIR=$(pwd)
+PACKAGE_NAME="node-exporter-$(date +%Y%m%d-%H%M%S)"
+PACKAGE_FILE="${PACKAGE_NAME}.tar.gz"
+
+log_info "开始打包 Node Exporter 安装包..."
+
+# 检查必要文件
+log_info "检查必要文件..."
+
+required_files=(
+    "install.sh"
+    "uninstall.sh"
+    "bin/node_exporter"
+    "check_health.sh"
+)
+
+missing_files=()
+for file in "${required_files[@]}"; do
+    if [[ ! -f "$file" ]]; then
+        missing_files+=("$file")
+    fi
+done
+
+if [[ ${#missing_files[@]} -gt 0 ]]; then
+    echo "缺少以下文件:"
+    for file in "${missing_files[@]}"; do
+        echo "  - $file"
+    done
+    exit 1
+fi
+
+log_success "所有必要文件检查完成"
+
+# 创建临时目录
+TEMP_DIR=$(mktemp -d)
+log_info "创建临时目录: $TEMP_DIR"
+
+# 复制文件到临时目录
+cp -r . "$TEMP_DIR/$PACKAGE_NAME"
+
+# 进入临时目录
+cd "$TEMP_DIR"
+
+# 创建压缩包
+log_info "创建压缩包: $PACKAGE_FILE"
+tar -czf "$PACKAGE_FILE" "$PACKAGE_NAME"
+
+# 移动压缩包到原目录
+mv "$PACKAGE_FILE" "$CURRENT_DIR/"
+
+# 清理临时目录
+rm -rf "$TEMP_DIR"
+
+# 返回原目录
+cd "$CURRENT_DIR"
+
+# 显示结果
+log_success "打包完成！"
+echo
+echo "安装包文件: $PACKAGE_FILE"
+echo "文件大小: $(du -h "$PACKAGE_FILE" | cut -f1)"
+echo
+echo "使用方法:"
+echo "1. 将 $PACKAGE_FILE 传输到目标服务器"
+echo "2. 解压: tar -xzf $PACKAGE_FILE"
+echo "3. 进入目录: cd $PACKAGE_NAME"
+echo "4. 运行安装: sudo ./install.sh"
+echo
+echo "注意: 请确保所有必要文件都存在"

src/metric/client-plugins/all-in-one-demo/plugins/node-exporter/uninstall.sh

@@ -0,0 +1,239 @@
+#!/bin/bash
+
+# Node Exporter 卸载脚本
+# 版本: 1.0
+# 作者: AIOps Team
+# 日期: $(date +%Y-%m-%d)
+
+set -e
+
+# 颜色定义
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# 日志函数
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1"
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1"
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+# 检查是否为 root 用户
+check_root() {
+    if [[ $EUID -ne 0 ]]; then
+        log_error "此脚本需要 root 权限运行"
+        log_info "请使用: sudo $0"
+        exit 1
+    fi
+}
+
+# 停止运行中的进程
+stop_processes() {
+    log_info "停止 Node Exporter 进程..."
+    
+    local pid_file="/var/run/node-exporter.pid"
+    local stopped=false
+    
+    # 首先尝试通过 PID 文件停止服务
+    if [[ -f "$pid_file" ]]; then
+        local pid=$(cat "$pid_file")
+        if kill -0 "$pid" 2>/dev/null; then
+            log_info "通过 PID 文件停止服务 (PID: $pid)..."
+            kill "$pid"
+            sleep 3
+            
+            # 检查进程是否已停止
+            if kill -0 "$pid" 2>/dev/null; then
+                log_warning "进程未响应，强制终止..."
+                kill -9 "$pid" 2>/dev/null || true
+            fi
+            log_success "Node Exporter 进程已停止"
+            stopped=true
+        else
+            log_warning "PID 文件存在但进程已不存在，清理 PID 文件"
+            rm -f "$pid_file"
+        fi
+    fi
+    
+    # 查找并杀死所有 node_exporter 和 node-exporter 进程
+    local pids=$(pgrep -f "node_exporter\|node-exporter" 2>/dev/null || true)
+    if [[ -n "$pids" ]]; then
+        log_info "发现 node_exporter 或 node-exporter 进程，正在停止..."
+        for pid in $pids; do
+            log_info "停止进程 PID: $pid"
+            kill "$pid" 2>/dev/null || true
+        done
+        sleep 2
+        
+        # 检查是否还有进程在运行，如果有则强制终止
+        local remaining_pids=$(pgrep -f "node_exporter\|node-exporter" 2>/dev/null || true)
+        if [[ -n "$remaining_pids" ]]; then
+            log_warning "进程未响应，强制终止..."
+            for pid in $remaining_pids; do
+                log_info "强制终止进程 PID: $pid"
+                kill -9 "$pid" 2>/dev/null || true
+            done
+            sleep 1
+        fi
+        
+        # 最终检查
+        if pgrep -f "node_exporter\|node-exporter" > /dev/null; then
+            log_error "无法停止所有 node_exporter 进程"
+        else
+            log_success "所有 Node Exporter 进程已停止"
+            stopped=true
+        fi
+    else
+        log_info "Node Exporter 进程未运行"
+    fi
+    
+    # 清理 PID 文件
+    rm -f "$pid_file"
+    
+    if [[ "$stopped" == "false" ]]; then
+        log_warning "未发现需要停止的 Node Exporter 进程"
+    fi
+}
+
+# 删除二进制文件
+remove_binary() {
+    log_info "删除 Node Exporter 二进制文件..."
+    
+    local binary_files=(
+        "/usr/local/bin/node-exporter"
+        "/usr/local/bin/node_exporter"
+    )
+    
+    local deleted=false
+    for binary_file in "${binary_files[@]}"; do
+        if [[ -f "$binary_file" ]]; then
+            rm -f "$binary_file"
+            log_success "二进制文件已删除: $binary_file"
+            deleted=true
+        fi
+    done
+    
+    if [[ "$deleted" == "false" ]]; then
+        log_info "二进制文件不存在"
+    fi
+}
+
+# 删除配置文件
+remove_config() {
+    log_info "删除配置文件..."
+    
+    local config_dir="/etc/node_exporter"
+    
+    if [[ -d "$config_dir" ]]; then
+        rm -rf "$config_dir"
+        log_success "配置目录已删除"
+    else
+        log_info "配置目录不存在"
+    fi
+}
+
+# 删除数据目录
+remove_data_dir() {
+    log_info "删除数据目录..."
+    
+    local data_dir="/var/lib/node_exporter"
+    
+    if [[ -d "$data_dir" ]]; then
+        rm -rf "$data_dir"
+        log_success "数据目录已删除"
+    else
+        log_info "数据目录不存在"
+    fi
+}
+
+# 检查用户状态（可选）
+check_user_status() {
+    log_info "检查 node_exporter 用户状态..."
+    
+    if id "node_exporter" &>/dev/null; then
+        log_info "检测到 node_exporter 用户存在"
+        log_warning "node_exporter 是系统用户，可能被其他服务使用"
+        log_info "为了系统稳定性，将保留 node_exporter 用户"
+        log_info "如需手动删除，请运行: sudo userdel node_exporter"
+    else
+        log_info "node_exporter 用户不存在"
+    fi
+}
+
+# 清理日志文件
+cleanup_logs() {
+    log_info "清理日志文件..."
+    
+    # 清理 journal 日志
+    journalctl --vacuum-time=1s --quiet || true
+    
+    # 删除安装脚本创建的日志文件
+    rm -f /var/log/node-exporter.log
+    
+    log_success "日志文件已清理"
+}
+
+# 显示卸载信息
+show_uninstall_info() {
+    log_success "Node Exporter 卸载完成！"
+    echo
+    echo "已删除的内容:"
+    echo "  - 二进制文件: /usr/local/bin/node-exporter"
+    echo "  - 配置目录: /etc/node_exporter"
+    echo "  - 数据目录: /var/lib/node_exporter"
+    echo "  - 相关日志文件"
+    echo
+    echo "注意:"
+    echo "  - node_exporter 用户已保留（系统用户，可能被其他服务使用）"
+    echo "  - 如需完全清理，请手动检查并删除相关文件"
+    echo
+}
+
+# 主函数
+main() {
+    echo "=========================================="
+    echo "    Node Exporter 卸载脚本 v1.0"
+    echo "=========================================="
+    echo
+    
+    check_root
+    
+    log_warning "此操作将完全卸载 Node Exporter"
+    read -p "确认继续？(y/N): " confirm
+    
+    if [[ "$confirm" != "y" && "$confirm" != "Y" ]]; then
+        log_info "取消卸载操作"
+        exit 0
+    fi
+    
+    log_info "开始卸载 Node Exporter..."
+    
+    stop_processes
+    remove_binary
+    remove_config
+    remove_data_dir
+    cleanup_logs
+    
+    # 检查用户状态
+    check_user_status
+    
+    show_uninstall_info
+}
+
+# 脚本入口
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    main "$@"
+fi

src/metric/client-plugins/all-in-one-demo/scripts/check_health.sh

@@ -0,0 +1,286 @@
+#!/bin/bash
+
+# 整体健康检查脚本，调用各个组件的健康检查并将结果写入 .health_log 文件
+
+set -e
+
+# PID 文件检测，防止重复执行
+PIDFILE="/var/run/check_health.pid"
+if [ -f "$PIDFILE" ] && kill -0 $(cat "$PIDFILE") 2>/dev/null; then
+    echo "健康检查脚本已在运行中，跳过本次执行" >&2
+    exit 0
+fi
+echo $$ > "$PIDFILE"
+trap "rm -f $PIDFILE" EXIT
+
+# 获取脚本所在目录
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+HEALTH_LOG_FILE="$SCRIPT_DIR/.health_log"
+INSTALL_RECORD_FILE="$SCRIPT_DIR/.install_record"
+
+# 颜色定义
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# 日志函数 - 输出到 stderr 避免影响 JSON 结果
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1" >&2
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1" >&2
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1" >&2
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1" >&2
+}
+
+# 检查单个组件健康状态
+check_component() {
+    local component_name="$1"
+    local check_script_path="$2"
+
+    log_info "检查 $component_name 健康状态..."
+
+    if [[ ! -f "$check_script_path" ]]; then
+        log_error "健康检查脚本不存在: $check_script_path"
+        echo "{\"name\": \"$component_name\", \"status\": \"unhealth\", \"reason\": \"健康检查脚本不存在: $check_script_path\"}"
+        return 1
+    fi
+
+    if [[ ! -x "$check_script_path" ]]; then
+        log_error "健康检查脚本无执行权限: $check_script_path"
+        echo "{\"name\": \"$component_name\", \"status\": \"unhealth\", \"reason\": \"健康检查脚本无执行权限: $check_script_path\"}"
+        return 1
+    fi
+
+    # 执行健康检查脚本，只捕获 stdout，stderr 输出到终端
+    local result
+    if result=$("$check_script_path" 2>/dev/null); then
+        log_success "$component_name 健康检查通过"
+        echo "$result"
+        return 0
+    else
+        log_warning "$component_name 健康检查失败"
+        echo "$result"
+        return 1
+    fi
+}
+
+# 生成时间戳
+get_timestamp() {
+    date '+%Y-%m-%d %H:%M:%S'
+}
+
+# 生成UTC时间戳
+get_utc_timestamp() {
+    date -u '+%Y-%m-%dT%H:%M:%SZ'
+}
+
+# 获取主机名
+get_hostname() {
+    echo "${HOSTNAME:-$(hostname)}"
+}
+
+# 创建健康状态目录
+create_health_dir() {
+    local hostname=$(get_hostname)
+    local health_dir="/private/argus/agent/$hostname/health"
+    
+    if [[ ! -d "$health_dir" ]]; then
+        log_info "创建健康状态目录: $health_dir"
+        mkdir -p "$health_dir"
+    fi
+    
+    echo "$health_dir"
+}
+
+# 写入单个模块的健康状态JSON文件
+write_component_health_json() {
+    local component_name="$1"
+    local status="$2"
+    local error_msg="$3"
+    local health_dir="$4"
+    
+    # 生成模块名前缀-xxx.json格式的文件名
+    local module_prefix="metric"
+    local filename="${module_prefix}-${component_name}.json"
+    local filepath="$health_dir/$filename"
+    
+    # 生成UTC时间戳
+    local timestamp=$(get_utc_timestamp)
+    
+    # 构建JSON内容
+    local json_content=$(cat << EOF
+{
+    "status": "$status",
+    "error": "$error_msg",
+    "timestamp": "$timestamp"
+}
+EOF
+)
+    
+    # 写入文件
+    echo "$json_content" > "$filepath"
+    log_info "已写入模块健康状态文件: $filepath"
+}
+
+# 从安装记录文件中读取组件安装目录
+read_install_record() {
+    local install_record_file="$1"
+
+    if [[ ! -f "$install_record_file" ]]; then
+        log_error "安装记录文件不存在: $install_record_file"
+        return 1
+    fi
+
+    # 检查是否有 jq 命令来解析 JSON
+    if command -v jq &> /dev/null; then
+        # 使用 jq 解析 JSON
+        local components_json
+        if components_json=$(jq -r '.components | to_entries[] | "\(.key):\(.value.install_dir)"' "$install_record_file" 2>/dev/null); then
+            echo "$components_json"
+            return 0
+        else
+            log_error "无法解析安装记录文件 JSON 格式: $install_record_file"
+            return 1
+        fi
+    else
+        # 如果没有 jq，尝试简单的文本解析
+        log_warning "jq 命令不可用，尝试简单文本解析"
+
+        # 查找所有 install_dir 行
+        local components=()
+        while IFS= read -r line; do
+            if [[ "$line" =~ \"install_dir\":[[:space:]]*\"([^\"]+)\" ]]; then
+                local install_dir="${BASH_REMATCH[1]}"
+                # 从路径中提取组件名称
+                local component_name=$(basename "$install_dir")
+                components+=("$component_name:$install_dir")
+            fi
+        done < "$install_record_file"
+
+        if [[ ${#components[@]} -gt 0 ]]; then
+            printf '%s\n' "${components[@]}"
+            return 0
+        else
+            log_error "无法从安装记录文件中提取组件信息"
+            return 1
+        fi
+    fi
+}
+
+# 主函数
+main() {
+    echo "==========================================" >&2
+    echo "    整体健康检查脚本" >&2
+    echo "==========================================" >&2
+    echo >&2
+
+    # 记录健康检查开始时间
+    local start_time=$(get_timestamp)
+    log_info "健康检查开始时间: $start_time"
+
+    # 创建健康状态目录
+    local health_dir
+    health_dir=$(create_health_dir)
+
+    # 从安装记录文件中读取组件信息
+    log_info "从安装记录文件读取组件信息: $INSTALL_RECORD_FILE"
+    local components_info
+    if ! components_info=$(read_install_record "$INSTALL_RECORD_FILE"); then
+        log_error "无法读取安装记录文件，健康检查终止"
+        exit 1
+    fi
+
+    # 存储所有检查结果
+    local all_results=()
+    local overall_status="health"
+
+    # 逐个检查组件
+    while IFS= read -r component_info; do
+        if [[ -n "$component_info" ]]; then
+            IFS=':' read -r component_name install_dir <<< "$component_info"
+            local check_script_path="$install_dir/check_health.sh"
+
+            local result
+            local component_status="healthy"
+            local error_msg=""
+            
+            if result=$(check_component "$component_name" "$check_script_path"); then
+                all_results+=("$result")
+            else
+                all_results+=("$result")
+                overall_status="unhealth"
+                component_status="unhealthy"
+                # 从结果中提取错误信息
+                if command -v jq &> /dev/null; then
+                    error_msg=$(echo "$result" | jq -r '.reason // ""' 2>/dev/null || echo "")
+                else
+                    # 简单的文本解析提取错误信息
+                    if [[ "$result" =~ \"reason\":[[:space:]]*\"([^\"]+)\" ]]; then
+                        error_msg="${BASH_REMATCH[1]}"
+                    fi
+                fi
+            fi
+            
+            # 写入单个模块的健康状态JSON文件
+            write_component_health_json "$component_name" "$component_status" "$error_msg" "$health_dir"
+        fi
+    done <<< "$components_info"
+
+    # 记录健康检查结束时间
+    local end_time=$(get_timestamp)
+    log_info "健康检查结束时间: $end_time"
+
+    # 构建完整的健康检查结果 JSON
+    local health_check_result=$(cat << EOF
+{
+  "start_time": "$start_time",
+  "end_time": "$end_time",
+  "overall_status": "$overall_status",
+  "components": [
+$(printf '%s,\n' "${all_results[@]}" | sed '$s/,$//')
+  ]
+}
+EOF
+)
+
+    # 写入健康日志文件
+    log_info "将健康检查结果写入日志文件: $HEALTH_LOG_FILE"
+    echo "$health_check_result" >> "$HEALTH_LOG_FILE"
+
+    # 输出 JSON 结果到 stdout
+    echo "$health_check_result"
+
+    # 显示总结到 stderr
+    echo >&2
+    echo "==========================================" >&2
+    echo "    健康检查总结" >&2
+    echo "==========================================" >&2
+    echo "开始时间: $start_time" >&2
+    echo "结束时间: $end_time" >&2
+    echo "整体状态: $overall_status" >&2
+    echo "日志文件: $HEALTH_LOG_FILE" >&2
+    echo >&2
+
+    if [[ "$overall_status" == "health" ]]; then
+        log_success "所有组件健康检查通过！"
+        exit 0
+    else
+        log_error "部分组件健康检查失败，请查看上述详细信息"
+        exit 1
+    fi
+}
+
+# 脚本入口
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    main "$@"
+fi

src/metric/client-plugins/all-in-one-demo/scripts/check_version.sh

@@ -0,0 +1,240 @@
+#!/bin/bash
+
+# 版本校验脚本
+# 比较本地 LATEST_VERSION 与 FTP 的 VERSION 版本，如果不一致则更新对应版本
+
+set -e
+
+# 颜色定义
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# 日志函数 - 输出到 stderr 避免影响函数返回值
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1" >&2
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1" >&2
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1" >&2
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1" >&2
+}
+
+# 获取脚本所在目录
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+
+# 动态获取当前版本目录
+get_current_version_dir() {
+    # 查找 /opt/argus-metric/versions/ 下的最新版本目录
+    local versions_dir="/opt/argus-metric/versions"
+    if [[ -d "$versions_dir" ]]; then
+        # 按版本号排序，获取最新的版本目录
+        local latest_version_dir=$(ls -1 "$versions_dir" 2>/dev/null | sort -V | tail -1)
+        if [[ -n "$latest_version_dir" ]]; then
+            echo "$versions_dir/$latest_version_dir"
+        else
+            echo "/opt/argus-metric"
+        fi
+    else
+        echo "/opt/argus-metric"
+    fi
+}
+
+# 获取当前版本目录
+CURRENT_VERSION_DIR=$(get_current_version_dir)
+# LATEST_VERSION 文件在根目录
+LOCAL_VERSION_FILE="/opt/argus-metric/LATEST_VERSION"
+REMOTE_VERSION_URL=""
+LOG_FILE="$CURRENT_VERSION_DIR/.version_check.log"
+
+# 从环境变量或配置文件获取 FTP 服务器信息
+get_ftp_config() {
+    # 优先从环境变量获取配置
+    log_info "获取 FTP 配置信息..."
+    
+    # 如果环境变量中没有设置，则尝试从配置文件读取
+    if [[ -z "$FTP_SERVER" || -z "$FTP_USER" || -z "$FTP_PASSWORD" ]]; then
+        local config_file="$SCRIPT_DIR/../config/config.env"
+        if [[ -f "$config_file" ]]; then
+            log_info "从配置文件读取 FTP 配置: $config_file"
+            source "$config_file"
+        fi
+    else
+        log_info "使用环境变量中的 FTP 配置"
+    fi
+    
+    # 设置默认值（如果环境变量和配置文件都没有设置）
+    FTP_SERVER="${FTP_SERVER:-localhost}"
+    FTP_USER="${FTP_USER:-ftpuser}"
+    FTP_PASSWORD="${FTP_PASSWORD:-ZGClab1234!}"
+    
+    # 构建远程版本文件 URL
+    REMOTE_VERSION_URL="ftp://${FTP_USER}:${FTP_PASSWORD}@${FTP_SERVER}/LATEST_VERSION"
+    
+    log_info "FTP 配置来源: ${FTP_CONFIG_SOURCE:-环境变量/配置文件}"
+}
+
+# 获取远程版本号
+get_remote_version() {
+    log_info "从 FTP 服务器获取远程版本号..."
+    log_info "远程地址: $REMOTE_VERSION_URL"
+    
+    # 先测试 FTP 连接
+    log_info "测试 FTP 连接..."
+    if curl -u "${FTP_USER}:${FTP_PASSWORD}" -sfI "ftp://${FTP_SERVER}/" >/dev/null 2>&1; then
+        log_success "FTP 服务器连接成功"
+    else
+        log_error "无法连接到 FTP 服务器: $FTP_SERVER"
+        return 1
+    fi
+    
+    # 测试 LATEST_VERSION 文件是否存在
+    log_info "检查远程 LATEST_VERSION 文件是否存在..."
+    if curl -u "${FTP_USER}:${FTP_PASSWORD}" -sfI "ftp://${FTP_SERVER}/LATEST_VERSION" >/dev/null 2>&1; then
+        log_success "远程 LATEST_VERSION 文件存在"
+    else
+        log_error "远程 LATEST_VERSION 文件不存在或无法访问"
+        return 1
+    fi
+    
+    # 获取远程版本号
+    local remote_version
+    if remote_version=$(curl -u "${FTP_USER}:${FTP_PASSWORD}" -sfL "ftp://${FTP_SERVER}/LATEST_VERSION" 2>/dev/null | tr -d '[:space:]'); then
+        if [[ -n "$remote_version" ]]; then
+            log_success "获取到远程版本号: $remote_version"
+            echo "$remote_version"
+        else
+            log_error "远程版本号为空"
+            return 1
+        fi
+    else
+        log_error "获取远程版本号失败"
+        return 1
+    fi
+}
+
+# 获取本地版本号
+get_local_version() {
+    if [[ -f "$LOCAL_VERSION_FILE" ]]; then
+        local local_version=$(cat "$LOCAL_VERSION_FILE" 2>/dev/null | tr -d '[:space:]')
+        if [[ -n "$local_version" ]]; then
+            log_info "本地版本号: $local_version"
+            echo "$local_version"
+        else
+            log_warning "本地版本文件为空"
+            echo ""
+        fi
+    else
+        log_warning "本地版本文件不存在: $LOCAL_VERSION_FILE"
+        echo ""
+    fi
+}
+
+# 更新到新版本
+update_to_version() {
+    local new_version="$1"
+    local temp_dir="/tmp/argus-update-$$"
+    local setup_script="$temp_dir/setup.sh"
+    
+    log_info "开始更新到版本: $new_version"
+    
+    # 创建临时目录
+    mkdir -p "$temp_dir"
+    
+    # 下载最新的 setup.sh
+    log_info "从 FTP 服务器下载最新的安装脚本..."
+    local setup_url="ftp://${FTP_USER}:${FTP_PASSWORD}@${FTP_SERVER}/setup.sh"
+    
+    if curl -fsS "$setup_url" -o "$setup_script"; then
+        log_success "安装脚本下载完成"
+    else
+        log_error "下载安装脚本失败: $setup_url"
+        rm -rf "$temp_dir"
+        return 1
+    fi
+    
+    # 添加执行权限
+    chmod +x "$setup_script"
+    
+    # 执行安装脚本
+    log_info "执行安装脚本进行版本更新..."
+    if "$setup_script" --server "$FTP_SERVER" --user "$FTP_USER" --password "$FTP_PASSWORD" --version "$new_version"; then
+        log_success "版本更新完成: $new_version"
+        rm -rf "$temp_dir"
+        return 0
+    else
+        log_error "版本更新失败: $new_version"
+        rm -rf "$temp_dir"
+        return 1
+    fi
+}
+
+# 记录检查日志
+log_check() {
+    local message="$1"
+    local timestamp=$(date '+%Y-%m-%d %H:%M:%S')
+    echo "[$timestamp] $message" >> "$LOG_FILE"
+}
+
+# 主函数
+main() {
+    log_info "开始版本校验检查..."
+    log_check "版本校验检查开始"
+    
+    # 确保系统目录存在
+    mkdir -p "/opt/argus-metric"
+    mkdir -p "$CURRENT_VERSION_DIR"
+    
+    log_info "当前版本目录: $CURRENT_VERSION_DIR"
+    
+    # 获取 FTP 配置
+    get_ftp_config
+    
+    # 获取本地版本号
+    local local_version
+    local_version=$(get_local_version)
+    
+    # 获取远程版本号
+    local remote_version
+    if ! remote_version=$(get_remote_version); then
+        log_error "无法获取远程版本号，跳过本次检查"
+        log_check "版本校验失败：无法获取远程版本号"
+        exit 1
+    fi
+    
+    # 比较版本号
+    if [[ "$local_version" == "$remote_version" ]]; then
+        log_info "版本一致，无需更新 (本地: $local_version, 远程: $remote_version)"
+        log_check "版本校验完成：版本一致 ($local_version)"
+    else
+        log_info "检测到版本不一致 (本地: $local_version, 远程: $remote_version)"
+        log_check "检测到版本不一致：本地($local_version) -> 远程($remote_version)"
+        
+        # 更新到新版本
+        if update_to_version "$remote_version"; then
+            log_success "版本更新成功: $local_version -> $remote_version"
+            log_check "版本更新成功：$local_version -> $remote_version"
+        else
+            log_error "版本更新失败"
+            log_check "版本更新失败：$local_version -> $remote_version"
+            exit 1
+        fi
+    fi
+    
+    log_success "版本校验检查完成"
+    log_check "版本校验检查完成"
+}
+
+# 脚本入口
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    main "$@"
+fi

src/metric/client-plugins/all-in-one-demo/scripts/install_artifact.sh

@@ -0,0 +1,995 @@
+#!/bin/bash
+
+set -e
+
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m'
+
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1"
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1"
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+# 配置变量
+INSTALL_DIR="${1:-$(pwd)}"  # 使用第一个参数作为安装目录，如果没有参数则使用当前目录
+TEMP_DIR="/tmp/metrics-install-$$"
+VERSION_FILE="version.json"
+
+
+# 加载配置文件
+load_config() {
+    local script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+    local config_file="$script_dir/config.env"
+    
+    if [[ -f "$config_file" ]]; then
+        log_info "加载配置文件: $config_file"
+        # 导出配置文件中的环境变量
+        set -a  # 自动导出所有变量
+        source "$config_file"
+        set +a  # 关闭自动导出
+        log_success "配置文件加载完成"
+    else
+        log_warning "配置文件不存在: $config_file，使用默认配置"
+    fi
+}
+
+# 复制配置文件到安装目录
+copy_config_files() {
+    log_info "复制配置文件到安装目录..."
+    
+    local script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+    local source_config="$script_dir/../config/config.env"
+    local target_config="$INSTALL_DIR/config.env"
+    
+    if [[ -f "$source_config" ]]; then
+        # 检查源文件和目标文件是否是同一个文件
+        if [[ "$source_config" == "$target_config" ]]; then
+            log_info "配置文件已在目标位置，跳过复制"
+            log_success "配置文件已存在: $target_config"
+        else
+            if cp "$source_config" "$target_config"; then
+                log_success "配置文件复制完成: $target_config"
+            else
+                log_error "配置文件复制失败"
+                return 1
+            fi
+        fi
+    else
+        log_warning "源配置文件不存在: $source_config"
+    fi
+    
+    # 复制版本校验脚本
+    log_info "复制版本校验脚本到安装目录..."
+    local target_check_version="$INSTALL_DIR/check_version.sh"
+    
+    # 检查目标文件是否已存在（从 artifact 包中解压出来的）
+    if [[ -f "$target_check_version" ]]; then
+        log_info "版本校验脚本已存在，设置执行权限..."
+        chmod +x "$target_check_version"
+        log_success "版本校验脚本权限设置完成: $target_check_version"
+    else
+        log_warning "版本校验脚本不存在: $target_check_version"
+        log_info "请确保 check_version.sh 已包含在 artifact 包中"
+    fi
+}
+
+check_root() {
+    if [[ $EUID -ne 0 ]]; then
+        log_error "此脚本需要 root 权限运行"
+        log_info "请使用: sudo $0 [安装目录]"
+        log_info "如果不指定安装目录，将使用当前目录: $(pwd)"
+        exit 1
+    fi
+}
+
+# 检查系统要求
+check_system() {
+    log_info "检查系统要求..."
+    
+    # 检查操作系统
+    if [[ ! -f /etc/os-release ]]; then
+        log_error "无法检测操作系统版本"
+        exit 1
+    fi
+    
+    source /etc/os-release
+    log_info "检测到操作系统: $NAME $VERSION"
+    
+    # 检查系统架构
+    arch=$(uname -m)
+    log_info "系统架构: $arch"
+    
+    # 检查磁盘空间
+    available_space=$(df / | awk 'NR==2 {print $4}')
+    if [[ $available_space -lt 10485760 ]]; then  # 10GB in KB
+        log_warning "可用磁盘空间不足 10GB，当前可用: $(($available_space / 1024 / 1024))GB"
+    fi
+    
+    # 检查内存
+    total_mem=$(free -m | awk 'NR==2{print $2}')
+    if [[ $total_mem -lt 4096 ]]; then  # 4GB
+        log_warning "系统内存不足 4GB，当前: ${total_mem}MB"
+    fi
+}
+
+# 查找版本文件
+find_version_file() {
+    log_info "查找版本信息文件..."
+    
+    # 在当前目录查找
+    if [[ -f "$VERSION_FILE" ]]; then
+        VERSION_FILE_PATH="$(pwd)/$VERSION_FILE"
+        log_success "找到版本文件: $VERSION_FILE"
+        return 0
+    fi
+    
+    # 在 artifact 目录查找
+    for version_dir in artifact/*/; do
+        if [[ -f "${version_dir}${VERSION_FILE}" ]]; then
+            VERSION_FILE_PATH="$(cd "$(dirname "${version_dir}${VERSION_FILE}")" && pwd)/$(basename "${version_dir}${VERSION_FILE}")"
+            log_success "找到版本文件: $VERSION_FILE_PATH"
+            return 0
+        fi
+    done
+    
+    log_error "未找到版本信息文件 $VERSION_FILE"
+    exit 1
+}
+
+# 解析版本信息
+parse_version_info() {
+    log_info "解析版本信息..."
+    
+    if [[ ! -f "$VERSION_FILE_PATH" ]]; then
+        log_error "版本文件不存在: $VERSION_FILE_PATH"
+        exit 1
+    fi
+    
+    # 使用 jq 解析 JSON（如果可用）
+    if command -v jq &> /dev/null; then
+        # 验证JSON文件格式
+        if ! jq empty "$VERSION_FILE_PATH" 2>/dev/null; then
+            log_error "JSON文件格式错误，请检查 $VERSION_FILE_PATH"
+            exit 1
+        fi
+        
+        VERSION=$(jq -r '.version' "$VERSION_FILE_PATH")
+        BUILD_TIME=$(jq -r '.build_time' "$VERSION_FILE_PATH")
+        
+        # 解析 artifact_list
+        if jq -e '.artifact_list' "$VERSION_FILE_PATH" > /dev/null 2>&1; then
+            jq -r '.artifact_list | to_entries[] | "\(.key):\(.value)"' "$VERSION_FILE_PATH" > "$TEMP_DIR/components.txt"
+        else
+            log_error "version.json 中缺少 artifact_list 字段"
+            exit 1
+        fi
+        
+        # 解析 checksums
+        if jq -e '.checksums' "$VERSION_FILE_PATH" > /dev/null 2>&1; then
+            jq -r '.checksums | to_entries[] | "\(.key):\(.value)"' "$VERSION_FILE_PATH" > "$TEMP_DIR/checksums.txt"
+        else
+            log_error "version.json 中缺少 checksums 字段"
+            exit 1
+        fi
+        
+        # 解析 install_order（现在包含完整的文件名）
+        if jq -e '.install_order' "$VERSION_FILE_PATH" > /dev/null 2>&1; then
+            jq -r '.install_order[]' "$VERSION_FILE_PATH" > "$TEMP_DIR/install_order.txt"
+        else
+            log_error "version.json 中缺少 install_order 字段"
+            exit 1
+        fi
+        
+    else
+        log_warning "jq 未安装，使用简单的 JSON 解析"
+        # 简单的 JSON 解析
+        VERSION=$(grep '"version"' "$VERSION_FILE_PATH" | sed 's/.*"version": *"\([^"]*\)".*/\1/')
+        BUILD_TIME=$(grep '"build_time"' "$VERSION_FILE_PATH" | sed 's/.*"build_time": *"\([^"]*\)".*/\1/')
+        
+        # 解析 artifact_list（跳过字段名本身）
+        grep -A 100 '"artifact_list"' "$VERSION_FILE_PATH" | grep -v '"artifact_list"' | grep -E '^\s*"[^"]+":\s*"[^"]+"' | while read line; do
+            component=$(echo "$line" | sed 's/.*"\([^"]*\)":\s*"[^"]*".*/\1/')
+            version=$(echo "$line" | sed 's/.*"[^"]*":\s*"\([^"]*\)".*/\1/')
+            echo "$component:$version" >> "$TEMP_DIR/components.txt"
+        done
+        
+        # 解析 checksums（跳过字段名本身）
+        grep -A 100 '"checksums"' "$VERSION_FILE_PATH" | grep -v '"checksums"' | grep -E '^\s*"[^"]+":\s*"[^"]+"' | while read line; do
+            component=$(echo "$line" | sed 's/.*"\([^"]*\)":\s*"[^"]*".*/\1/')
+            checksum=$(echo "$line" | sed 's/.*"[^"]*":\s*"\([^"]*\)".*/\1/')
+            echo "$component:$checksum" >> "$TEMP_DIR/checksums.txt"
+        done
+        
+        # 解析 install_order（跳过字段名本身，只取数组元素）
+        grep -A 100 '"install_order"' "$VERSION_FILE_PATH" | grep -v '"install_order"' | grep -E '^\s*"[^"]+"' | while read line; do
+            component=$(echo "$line" | sed 's/.*"\([^"]*\)".*/\1/')
+            echo "$component" >> "$TEMP_DIR/install_order.txt"
+        done
+        
+        # 验证解析结果
+        if [[ ! -f "$TEMP_DIR/components.txt" || ! -s "$TEMP_DIR/components.txt" ]]; then
+            log_error "无法解析 artifact_list，请检查 version.json 格式"
+            exit 1
+        fi
+        
+        if [[ ! -f "$TEMP_DIR/checksums.txt" || ! -s "$TEMP_DIR/checksums.txt" ]]; then
+            log_error "无法解析 checksums，请检查 version.json 格式"
+            exit 1
+        fi
+        
+        if [[ ! -f "$TEMP_DIR/install_order.txt" || ! -s "$TEMP_DIR/install_order.txt" ]]; then
+            log_error "无法解析 install_order，请检查 version.json 格式"
+            exit 1
+        fi
+    fi
+    
+    log_success "版本信息解析完成"
+    log_info "  版本: $VERSION"
+    log_info "  构建时间: $BUILD_TIME"
+    
+    component_count=0
+    if [[ -f "$TEMP_DIR/components.txt" ]]; then
+        component_count=$(wc -l < "$TEMP_DIR/components.txt")
+        log_info "  组件数量: $component_count"
+        log_info "  组件列表:"
+        while IFS= read -r line; do
+            component=$(echo "$line" | cut -d':' -f1)
+            version=$(echo "$line" | cut -d':' -f2)
+            log_info "    - $component v$version"
+        done < "$TEMP_DIR/components.txt"
+    else
+        log_error "components.txt 文件不存在"
+        exit 1
+    fi
+}
+
+# 验证文件完整性
+verify_checksums() {
+    log_info "验证文件完整性..."
+    
+    artifact_dir=$(dirname "$VERSION_FILE_PATH")
+    log_info "Artifact 目录: $artifact_dir"
+    failed_verification=0
+    
+    if [[ -f "$TEMP_DIR/checksums.txt" ]]; then
+        while IFS= read -r line; do
+            component=$(echo "$line" | cut -d':' -f1)
+            expected_checksum=$(echo "$line" | cut -d':' -f2-)
+            
+            # 查找匹配的 tar 文件
+            actual_file=""
+            for file in "$artifact_dir/${component}-"*.tar.gz; do
+                if [[ -f "$file" ]]; then
+                    actual_file="$file"
+                    break
+                fi
+            done
+            
+            if [[ -z "$actual_file" ]]; then
+                log_error "找不到组件文件: $component"
+                failed_verification=1
+                continue
+            fi
+            
+            # 计算实际校验和
+            actual_checksum="sha256:$(sha256sum "$actual_file" | cut -d' ' -f1)"
+            
+            if [[ "$actual_checksum" == "$expected_checksum" ]]; then
+                log_success "  $component: 校验通过"
+            else
+                log_error "  $component: 校验失败"
+                log_error "    期望: $expected_checksum"
+                log_error "    实际: $actual_checksum"
+                failed_verification=1
+            fi
+        done < "$TEMP_DIR/checksums.txt"
+    fi
+    
+    if [[ $failed_verification -eq 1 ]]; then
+        log_error "文件完整性验证失败"
+        exit 1
+    fi
+    
+    log_success "所有文件校验通过"
+}
+
+# 创建安装目录
+create_install_dirs() {
+    log_info "创建安装目录..."
+    
+    mkdir -p "$INSTALL_DIR"
+    mkdir -p "$TEMP_DIR"
+    
+    log_success "安装目录创建完成: $INSTALL_DIR"
+}
+
+# 获取系统版本
+get_system_version() {
+    if [[ ! -f /etc/os-release ]]; then
+        log_error "无法检测操作系统版本"
+        return 1
+    fi
+    
+    source /etc/os-release
+    
+    # 提取主版本号
+    case "$VERSION_ID" in
+        "20.04")
+            echo "ubuntu20"
+            ;;
+        "22.04")
+            echo "ubuntu22"
+            ;;
+        *)
+            log_warning "未识别的Ubuntu版本: $VERSION_ID，尝试使用ubuntu22"
+            echo "ubuntu22"
+            ;;
+    esac
+}
+
+# 安装系统依赖包
+install_system_deps() {
+    log_info "检查系统依赖包..."
+
+    local script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+    local deps_dir="$script_dir/deps"
+
+    # 检查deps目录是否存在
+    if [[ ! -d "$deps_dir" ]]; then
+        log_info "deps 目录不存在，跳过系统依赖包安装"
+        return 0
+    fi
+
+    # 获取系统版本对应的依赖目录
+    local system_version=$(get_system_version)
+    local version_deps_dir="$deps_dir/$system_version"
+    
+    log_info "检测到系统版本: $system_version"
+    
+    # 检查版本特定的依赖目录是否存在
+    if [[ ! -d "$version_deps_dir" ]]; then
+        log_warning "未找到 $system_version 版本的依赖目录: $version_deps_dir"
+        # 回退到旧的逻辑，检查根deps目录
+        local deps_count=$(find "$deps_dir" -name "*.tar.gz" | wc -l)
+        if [[ $deps_count -eq 0 ]]; then
+            log_info "deps 目录中没有 tar.gz 文件，跳过系统依赖包安装"
+            return 0
+        fi
+        version_deps_dir="$deps_dir"
+    else
+        # 检查版本目录中是否有tar.gz文件
+        local deps_count=$(find "$version_deps_dir" -name "*.tar.gz" | wc -l)
+        if [[ $deps_count -eq 0 ]]; then
+            log_info "$system_version 版本目录中没有 tar.gz 文件，跳过系统依赖包安装"
+            return 0
+        fi
+    fi
+
+    log_info "找到 $system_version 版本的依赖包，开始安装..."
+
+    # 创建临时目录用于解压依赖包
+    local deps_temp_dir="${TEMP_DIR:-/tmp}/deps"
+    mkdir -p "$deps_temp_dir"
+
+    # 定义要检查的核心依赖
+    local CORE_DEPS=(jq cron curl)
+    local FAILED_DEPS=()
+
+    # 处理每个tar.gz文件
+    find "$version_deps_dir" -name "*.tar.gz" | while read tar_file; do
+        local tar_basename=$(basename "$tar_file")
+        local extract_name="${tar_basename%.tar.gz}"
+
+        log_info "处理依赖包: $tar_basename"
+
+        # 解压到临时目录
+        local extract_dir="$deps_temp_dir/$extract_name"
+        mkdir -p "$extract_dir"
+
+        if tar -xzf "$tar_file" -C "$extract_dir" 2>/dev/null; then
+            log_success "  $tar_basename 解压完成"
+        else
+            log_error "  $tar_basename 解压失败"
+            continue
+        fi
+
+        # 进入解压目录，查找deb包
+        cd "$extract_dir" || continue
+        local deb_files=(*.deb)
+        if [[ ${#deb_files[@]} -gt 0 ]]; then
+            log_info "  找到 ${#deb_files[@]} 个 deb 包，开始安装..."
+
+            for deb in "${deb_files[@]}"; do
+                local pkg_name
+                pkg_name=$(dpkg-deb -f "$deb" Package 2>/dev/null)
+
+                # 如果已安装，则跳过
+                if dpkg -s "$pkg_name" &>/dev/null; then
+                    log_success "  $pkg_name 已安装，跳过"
+                    continue
+                fi
+
+                # 尝试安装
+                log_info "  安装 $pkg_name..."
+                if DEBIAN_FRONTEND=noninteractive dpkg -i "$deb" &>/dev/null; then
+                    log_success "  $pkg_name 安装成功"
+                else
+                    log_warning "  $pkg_name 安装失败，尝试修复依赖..."
+                    if DEBIAN_FRONTEND=noninteractive apt-get install -f -y &>/dev/null; then
+                        if dpkg -s "$pkg_name" &>/dev/null; then
+                            log_success "  $pkg_name 修复安装成功"
+                        else
+                            log_error "  $pkg_name 仍未安装成功"
+                            FAILED_DEPS+=("$pkg_name")
+                        fi
+                    else
+                        log_error "  $pkg_name 自动修复失败"
+                        FAILED_DEPS+=("$pkg_name")
+                    fi
+                fi
+            done
+        else
+            log_info "  $tar_basename 中没有找到deb包，跳过"
+        fi
+
+        # 返回到依赖临时目录
+        cd "$deps_temp_dir" || continue
+    done
+
+    # 检查并启动 cron 服务
+    start_cron_service
+
+    # 总结安装结果
+    if [[ ${#FAILED_DEPS[@]} -gt 0 ]]; then
+        log_error "以下系统依赖未能成功安装，安装终止，请手动安装后重试："
+        for f in "${FAILED_DEPS[@]}"; do
+            echo "  - $f"
+        done
+        exit 1
+    else
+        log_success "系统依赖包安装完成，全部就绪"
+    fi
+}
+
+# 启动 cron 服务
+start_cron_service() {
+    log_info "检查并启动 cron 服务..."
+    
+    # 检查 cron 是否已经在运行
+    if pgrep -x "cron" > /dev/null; then
+        log_success "cron 服务已在运行"
+        return 0
+    fi
+    
+    # 检查 /usr/sbin/cron 是否存在
+    if [[ ! -f "/usr/sbin/cron" ]]; then
+        log_warning "cron 可执行文件不存在，跳过启动"
+        return 1
+    fi
+    
+    # 启动 cron 服务
+    log_info "启动 cron 服务..."
+    if /usr/sbin/cron start 2>/dev/null || /usr/sbin/cron 2>/dev/null; then
+        log_success "cron 服务启动成功"
+        
+        sleep 2
+        
+        if pgrep -x "cron" > /dev/null; then
+            log_success "cron 服务运行正常"
+        else
+            log_warning "cron 服务可能未正常启动"
+        fi
+    else
+        log_error "cron 服务启动失败"
+        return 1
+    fi
+}
+
+# 安装组件
+install_components() {
+    log_info "开始安装组件..."
+    
+    artifact_dir=$(dirname "$VERSION_FILE_PATH")
+    log_info "Artifact 目录: $artifact_dir"
+    install_count=0
+    total_count=0
+    
+    if [[ -f "$TEMP_DIR/install_order.txt" ]]; then
+        total_count=$(wc -l < "$TEMP_DIR/install_order.txt")
+    fi
+    
+    if [[ -f "$TEMP_DIR/install_order.txt" ]]; then
+        while IFS= read -r filename; do
+            install_count=$((install_count + 1))
+            
+            # 从文件名中提取组件名（去掉时间戳后缀）
+            component=$(echo "$filename" | sed 's/-[0-9]\{8\}-[0-9]\{6\}\.tar\.gz$//')
+            
+            log_info "[$install_count/$total_count] 安装 $component..."
+            log_info "  文件名: $filename"
+            
+            # 直接使用完整的文件名
+            tar_file="$artifact_dir/$filename"
+            
+            if [[ ! -f "$tar_file" ]]; then
+                log_error "找不到组件文件: $filename"
+                log_info "  期望路径: $tar_file"
+                log_info "  当前目录: $(pwd)"
+                log_info "  目录内容:"
+                ls -la "$artifact_dir" | while read line; do
+                    log_info "    $line"
+                done
+                exit 1
+            fi
+            
+            log_info "  找到文件: $tar_file"
+            
+            # 解压到临时目录
+            component_temp_dir="$TEMP_DIR/$component"
+            mkdir -p "$component_temp_dir"
+            
+            if tar -xzf "$tar_file" -C "$component_temp_dir" 2>/dev/null; then
+                log_success "  $component 解压完成"
+            else
+                log_error "  $component 解压失败"
+                exit 1
+            fi
+            
+            # 查找解压后的目录
+            extracted_dir=""
+            for dir in "$component_temp_dir"/*; do
+                if [[ -d "$dir" ]]; then
+                    extracted_dir="$dir"
+                    break
+                fi
+            done
+            
+            if [[ -z "$extracted_dir" ]]; then
+                log_error "  $component 解压后未找到目录"
+                exit 1
+            fi
+            
+            # 执行安装脚本
+            if [[ -f "$extracted_dir/install.sh" ]]; then
+                log_info "  执行 $component 安装脚本..."
+                if (cd "$extracted_dir" && ./install.sh "$INSTALL_DIR"); then
+                    log_success "  $component 安装完成"
+                else
+                    log_error "  $component 安装失败"
+                    exit 1
+                fi
+            else
+                log_error "  $component 缺少 install.sh 文件"
+                exit 1
+            fi
+            
+            # 将解压后的目录移动到安装目录，保留组件目录
+            component_install_dir="$INSTALL_DIR/$component"
+            # 简化安装逻辑：直接删除旧目录，不进行备份
+            if [[ -d "$component_install_dir" ]]; then
+                log_info "  组件目录已存在，删除旧版本: $component_install_dir"
+                rm -rf "$component_install_dir"
+                # log_info "  组件目录已存在，备份后更新: $component_install_dir"
+                # mv "$component_install_dir" "${component_install_dir}.backup.$(date +%Y%m%d_%H%M%S)"
+            fi
+            mv "$extracted_dir" "$component_install_dir"
+            log_success "  组件目录已保存: $component_install_dir"
+            
+            # 清理临时文件
+            rm -rf "$component_temp_dir"
+        done < "$TEMP_DIR/install_order.txt"
+    fi
+    
+    log_success "所有组件安装完成"
+}
+
+# 创建安装记录
+create_install_record() {
+    log_info "创建安装记录..."
+    
+    # 等待一段时间确保所有进程都已启动
+    log_info "等待进程启动..."
+    sleep 3
+    
+    local install_time=$(date -u +"%Y-%m-%dT%H:%M:%SZ")
+    local install_record_file="$INSTALL_DIR/.install_record"
+    
+    # 创建 JSON 格式的安装记录
+    cat > "$install_record_file" << EOF
+{
+  "version": "$VERSION",
+  "build_time": "$BUILD_TIME",
+  "install_time": "$install_time",
+  "install_dir": "$INSTALL_DIR",
+  "install_pid": $$,
+  "components": {
+EOF
+
+    # 添加组件信息
+    local first_component=true
+    if [[ -f "$TEMP_DIR/components.txt" ]]; then
+        while IFS= read -r line; do
+            component=$(echo "$line" | cut -d':' -f1)
+            version=$(echo "$line" | cut -d':' -f2)
+            
+            # 获取组件的进程信息
+            local component_pid=""
+            
+            # 根据组件名查找进程，使用多种方法确保能找到PID
+            case "$component" in
+                "node-exporter")
+                    # 尝试多种方式查找node_exporter进程
+                    component_pid=$(pgrep -f "node_exporter" | head -1)
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(pgrep -f "node-exporter" | head -1)
+                    fi
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(ps aux | grep -v grep | grep "node_exporter" | awk '{print $2}' | head -1)
+                    fi
+                    ;;
+                "dcgm-exporter")
+                    # 查找dcgm-exporter进程
+                    component_pid=$(pgrep -f "dcgm-exporter" | head -1)
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(pgrep -f "dcgm_exporter" | head -1)
+                    fi
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(ps aux | grep -v grep | grep "dcgm-exporter" | awk '{print $2}' | head -1)
+                    fi
+                    ;;
+                "fluent-bit")
+                    # 查找fluent-bit进程
+                    component_pid=$(pgrep -f "fluent-bit" | head -1)
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(pgrep -f "fluent_bit" | head -1)
+                    fi
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(ps aux | grep -v grep | grep "fluent-bit" | awk '{print $2}' | head -1)
+                    fi
+                    ;;
+                "argus-agent")
+                    # 查找argus-agent进程
+                    component_pid=$(pgrep -f "argus-agent" | head -1)
+                    if [[ -z "$component_pid" ]]; then
+                        component_pid=$(ps aux | grep -v grep | grep "argus-agent" | awk '{print $2}' | head -1)
+                    fi
+                    ;;
+            esac
+            
+            # 记录找到的PID信息
+            if [[ -n "$component_pid" ]]; then
+                log_info "  找到 $component 进程 PID: $component_pid"
+            else
+                log_warning "  未找到 $component 进程"
+            fi
+            
+            # 添加逗号分隔符
+            if [[ "$first_component" == "true" ]]; then
+                first_component=false
+            else
+                echo "," >> "$install_record_file"
+            fi
+            
+            # 添加组件信息
+            cat >> "$install_record_file" << EOF
+    "$component": {
+      "version": "$version",
+      "pid": "$component_pid",
+      "install_dir": "$INSTALL_DIR/$component"
+    }
+EOF
+        done < "$TEMP_DIR/components.txt"
+    fi
+    
+    # 结束 JSON
+    cat >> "$install_record_file" << EOF
+  }
+}
+EOF
+
+    log_success "安装记录已创建: $install_record_file"
+}
+
+# 检查cron任务是否已存在
+check_cron_task_exists() {
+    local task_pattern="$1"
+    local temp_cron="$2"
+    
+    if grep -q "$task_pattern" "$temp_cron"; then
+        return 0  # 任务已存在
+    else
+        return 1  # 任务不存在
+    fi
+}
+
+# 设置健康检查定时任务
+setup_health_check_cron() {
+    log_info "设置健康检查定时任务..."
+    
+    # 直接使用当前安装目录，不依赖current软链接
+    # INSTALL_DIR 是 /opt/argus-metric/versions/1.34.0
+    local check_health_script="$INSTALL_DIR/check_health.sh"
+    
+    # 检查健康检查脚本是否存在
+    if [[ ! -f "$check_health_script" ]]; then
+        log_error "健康检查脚本不存在: $check_health_script"
+        return 1
+    fi
+    
+    # 确保脚本有执行权限
+    chmod +x "$check_health_script"
+    
+    # 创建临时crontab文件
+    local temp_cron="/tmp/crontab_$$"
+    
+    # 获取当前用户的crontab（如果存在）
+    crontab -l 2>/dev/null > "$temp_cron" || touch "$temp_cron"
+    
+    # 检查并删除旧的健康检查任务
+    if check_cron_task_exists "check_health.sh" "$temp_cron"; then
+        log_info "发现旧的健康检查定时任务，正在更新..."
+        # 删除所有包含check_health.sh的行
+        grep -v "check_health.sh" "$temp_cron" > "$temp_cron.new"
+        mv "$temp_cron.new" "$temp_cron"
+        log_info "旧的健康检查定时任务已删除"
+    fi
+    
+    # 添加新的定时任务（每5分钟执行一次）
+    echo "# Argus-Metrics 健康检查定时任务" >> "$temp_cron"
+    echo "*/5 * * * * $check_health_script >> $INSTALL_DIR/.health_cron.log 2>&1" >> "$temp_cron"
+    
+    # 安装新的crontab
+    if crontab "$temp_cron"; then
+        log_success "健康检查定时任务设置成功"
+        log_info "  执行频率: 每5分钟"
+        log_info "  日志文件: $INSTALL_DIR/.health_cron.log"
+        log_info "  查看定时任务: crontab -l"
+        log_info "  删除定时任务: crontab -e"
+    else
+        log_error "健康检查定时任务设置失败"
+        rm -f "$temp_cron"
+        return 1
+    fi
+    
+    # 清理临时文件
+    rm -f "$temp_cron"
+    
+    log_info "健康检查通过crontab自动执行"
+}
+
+# 设置 DNS 同步定时任务
+setup_dns_sync_cron() {
+    log_info "设置 DNS 同步定时任务..."
+    
+    # 使用当前版本目录中的 DNS 同步脚本
+    local sync_dns_script="$INSTALL_DIR/sync_dns.sh"
+    
+    # 检查 DNS 同步脚本是否存在
+    if [[ ! -f "$sync_dns_script" ]]; then
+        log_warning "DNS 同步脚本不存在: $sync_dns_script"
+        log_warning "跳过 DNS 同步定时任务设置"
+        return 0
+    fi
+    
+    # 确保脚本有执行权限
+    chmod +x "$sync_dns_script"
+    
+    # 创建临时crontab文件
+    local temp_cron="/tmp/crontab_$$"
+    
+    # 获取当前用户的crontab（如果存在）
+    crontab -l 2>/dev/null > "$temp_cron" || touch "$temp_cron"
+    
+    # 检查并删除旧的 DNS 同步任务
+    if check_cron_task_exists "sync_dns.sh" "$temp_cron"; then
+        log_info "发现旧的 DNS 同步定时任务，正在更新..."
+        # 删除所有包含sync_dns.sh的行
+        grep -v "sync_dns.sh" "$temp_cron" > "$temp_cron.new"
+        mv "$temp_cron.new" "$temp_cron"
+        log_info "旧的 DNS 同步定时任务已删除"
+    fi
+    
+    # 添加新的定时任务（每1分钟执行一次）
+    # 直接使用版本目录中的 DNS 同步脚本
+    echo "# Argus-Metrics DNS 同步定时任务" >> "$temp_cron"
+    echo "* * * * * $sync_dns_script >> $INSTALL_DIR/.dns_sync.log 2>&1" >> "$temp_cron"
+    
+    # 安装新的crontab
+    if crontab "$temp_cron"; then
+        log_success "DNS 同步定时任务设置成功"
+        log_info "  执行频率: 每1分钟"
+        log_info "  日志文件: $INSTALL_DIR/.dns_sync.log"
+        log_info "  查看定时任务: crontab -l"
+        log_info "  删除定时任务: crontab -e"
+    else
+        log_error "DNS 同步定时任务设置失败"
+        rm -f "$temp_cron"
+        return 1
+    fi
+    
+    # 清理临时文件
+    rm -f "$temp_cron"
+    
+    log_info "DNS 同步通过crontab自动执行"
+}
+
+# 设置版本校验定时任务
+setup_version_check_cron() {
+    log_info "设置版本校验定时任务..."
+    
+    # 使用当前版本目录中的版本校验脚本
+    local check_version_script="$INSTALL_DIR/check_version.sh"
+    
+    # 检查脚本是否存在
+    if [[ ! -f "$check_version_script" ]]; then
+        log_warning "版本校验脚本不存在: $check_version_script"
+        log_info "跳过版本校验定时任务设置"
+        return 0
+    fi
+    
+    # 确保脚本可执行
+    chmod +x "$check_version_script"
+    
+    # 创建临时crontab文件
+    local temp_cron="/tmp/crontab_$$"
+    crontab -l > "$temp_cron" 2>/dev/null || touch "$temp_cron"
+    
+    # 检查是否已存在版本校验定时任务
+    if check_cron_task_exists "check_version.sh" "$temp_cron"; then
+        log_info "发现旧的版本校验定时任务，正在更新..."
+        # 删除所有包含check_version.sh的行
+        grep -v "check_version.sh" "$temp_cron" > "$temp_cron.new"
+        mv "$temp_cron.new" "$temp_cron"
+        log_info "旧的版本校验定时任务已删除"
+    fi
+    
+    # 添加新的定时任务（每30分钟执行一次）
+    echo "# Argus-Metrics 版本校验定时任务" >> "$temp_cron"
+    echo "*/1 * * * * $check_version_script >> $INSTALL_DIR/.version_check.log 2>&1" >> "$temp_cron"
+    
+    # 安装新的crontab
+    if crontab "$temp_cron"; then
+        log_success "版本校验定时任务设置成功"
+        log_info "  执行频率: 每1分钟"
+        log_info "  日志文件: $INSTALL_DIR/.version_check.log"
+        log_info "  查看定时任务: crontab -l"
+        log_info "  删除定时任务: crontab -e"
+    else
+        log_error "版本校验定时任务设置失败"
+        rm -f "$temp_cron"
+        return 1
+    fi
+    
+    # 清理临时文件
+    rm -f "$temp_cron"
+    
+    log_info "版本校验通过crontab自动执行"
+}
+
+# 设置自动重启定时任务
+setup_restart_cron() {
+    log_info "设置自动重启定时任务..."
+    
+    # 使用当前版本目录中的重启脚本
+    local restart_script="$INSTALL_DIR/restart_unhealthy.sh"
+    
+    # 检查脚本是否存在
+    if [[ ! -f "$restart_script" ]]; then
+        log_warning "重启脚本不存在: $restart_script"
+        log_info "跳过自动重启定时任务设置"
+        return 0
+    fi
+    
+    # 确保脚本可执行
+    chmod +x "$restart_script"
+    
+    # 创建临时crontab文件
+    local temp_cron="/tmp/crontab_$$"
+    crontab -l > "$temp_cron" 2>/dev/null || touch "$temp_cron"
+    
+    # 检查是否已存在自动重启定时任务
+    if check_cron_task_exists "restart_unhealthy.sh" "$temp_cron"; then
+        log_info "发现旧的自动重启定时任务，正在更新..."
+        # 删除所有包含restart_unhealthy.sh的行
+        grep -v "restart_unhealthy.sh" "$temp_cron" > "$temp_cron.new"
+        mv "$temp_cron.new" "$temp_cron"
+        log_info "旧的自动重启定时任务已删除"
+    fi
+    
+    # 添加新的定时任务（每2分钟执行一次）
+    echo "# Argus-Metrics 自动重启定时任务" >> "$temp_cron"
+    echo "*/2 * * * * $restart_script >> $INSTALL_DIR/.restart.log 2>&1" >> "$temp_cron"
+    
+    # 安装新的crontab
+    if crontab "$temp_cron"; then
+        log_success "自动重启定时任务设置成功"
+        log_info "  执行频率: 每2分钟"
+        log_info "  日志文件: $INSTALL_DIR/.restart.log"
+        log_info "  查看定时任务: crontab -l"
+        log_info "  删除定时任务: crontab -e"
+    else
+        log_error "自动重启定时任务设置失败"
+        rm -f "$temp_cron"
+        return 1
+    fi
+    
+    # 清理临时文件
+    rm -f "$temp_cron"
+    
+    log_info "自动重启检查通过crontab自动执行"
+}
+
+# 显示安装信息
+show_install_info() {
+    log_success "Argus-Metrics All-in-One 安装完成！"
+}
+
+cleanup() {
+    if [[ -d "$TEMP_DIR" ]]; then
+        rm -rf "$TEMP_DIR"
+    fi
+}
+
+trap cleanup EXIT
+
+# 主函数
+main() {
+    echo "=========================================="
+    echo "    Argus-Metrics All-in-One 安装脚本 v1.0"
+    echo "=========================================="
+    echo
+    
+    # 加载配置文件
+    load_config
+    
+    log_info "安装目录: $INSTALL_DIR"
+    echo
+    
+    check_root
+    check_system
+    find_version_file
+    create_install_dirs
+    install_system_deps     
+    parse_version_info    
+    verify_checksums
+    install_components
+    copy_config_files
+    create_install_record
+    setup_health_check_cron
+    setup_dns_sync_cron
+    setup_version_check_cron
+    setup_restart_cron
+    
+    # 注释掉立即执行健康检查，避免与cron任务重复执行
+    # log_info "立即执行一次健康检查..."
+    # local check_health_script="$INSTALL_DIR/check_health.sh"
+    # if [[ -f "$check_health_script" ]]; then
+    #     if "$check_health_script" >> "$INSTALL_DIR/.health_check.log" 2>&1; then
+    #         log_success "健康检查执行完成"
+    #     else
+    #         log_warning "健康检查执行失败，请检查日志: $INSTALL_DIR/.health_check.log"
+    #     fi
+    # else
+    #     log_warning "健康检查脚本不存在: $check_health_script"
+    # fi
+    
+    show_install_info
+}
+
+if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
+    main "$@"
+fi

src/metric/client-plugins/all-in-one-demo/scripts/package_artifact.sh

@@ -0,0 +1,474 @@
+#!/bin/bash
+
+set -e
+
+# 颜色定义
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+# 日志函数
+log_info() {
+    echo -e "${BLUE}[INFO]${NC} $1"
+}
+
+log_success() {
+    echo -e "${GREEN}[SUCCESS]${NC} $1"
+}
+
+log_warning() {
+    echo -e "${YELLOW}[WARNING]${NC} $1"
+}
+
+log_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+# 显示帮助信息
+show_help() {
+    echo "AIOps All-in-One 打包脚本"
+    echo
+    echo "用法: $0 [选项]"
+    echo
+    echo "选项:"
+    echo "  --force     强制重新打包，即使版本已存在"
+    echo "  --help     显示此帮助信息"
+    echo
+    echo "示例:"
+    echo "  $0              # 正常打包，跳过已存在的版本"
+    echo "  $0 --force      # 强制重新打包"
+    echo
+}
+
+# 解析命令行参数
+FORCE_PACKAGE=false
+if [[ "$1" == "--force" ]]; then
+    FORCE_PACKAGE=true
+    log_info "强制重新打包模式"
+elif [[ "$1" == "--help" || "$1" == "-h" ]]; then
+    show_help
+    exit 0
+fi
+
+# 获取当前目录和版本
+CURRENT_DIR=$(pwd)
+VERSION=$(cat config/VERSION 2>/dev/null || echo "1.0.0")
+ARTIFACT_DIR="artifact/$VERSION"
+
+log_info "开始打包 AIOps All-in-One 安装包 v$VERSION"
+
+# 检查必要文件
+log_info "检查必要文件..."
+if [[ ! -f "config/VERSION" ]]; then
+    log_error "VERSION 文件不存在"
+    exit 1
+fi
+
+if [[ ! -f "config/checklist" ]]; then
+    log_error "checklist 文件不存在"
+    exit 1
+fi
+
+# 检查是否已存在该版本
+if [[ -d "$ARTIFACT_DIR" && "$FORCE_PACKAGE" == "false" ]]; then
+    log_info "检查版本 $VERSION 是否已存在..."
+    
+    # 检查 version.json 是否存在
+    if [[ -f "$ARTIFACT_DIR/version.json" ]]; then
+        log_info "找到已存在的版本信息文件"
+        
+        # 检查是否所有组件文件都存在
+        missing_files=0
+        existing_components=0
+        
+        # 解析已存在的 version.json 来检查文件
+        if command -v jq &> /dev/null; then
+            # 使用 jq 解析
+            while IFS= read -r component; do
+                existing_components=$((existing_components + 1))
+                # 查找对应的 tar 文件
+                found_file=false
+                for file in "$ARTIFACT_DIR/${component}-"*.tar.gz; do
+                    if [[ -f "$file" ]]; then
+                        found_file=true
+                        break
+                    fi
+                done
+                if [[ "$found_file" == "false" ]]; then
+                    missing_files=$((missing_files + 1))
+                    log_warning "  缺少文件: $component"
+                fi
+            done < <(jq -r '.artifact_list | keys[]' "$ARTIFACT_DIR/version.json" 2>/dev/null)
+        else
+            # 简单的文件检查
+            for file in "$ARTIFACT_DIR"/*.tar.gz; do
+                if [[ -f "$file" ]]; then
+                    existing_components=$((existing_components + 1))
+                fi
+            done
+        fi
+        
+        # 如果所有文件都存在，则跳过打包
+        if [[ $missing_files -eq 0 && $existing_components -gt 0 ]]; then
+            log_success "版本 $VERSION 已完整打包，跳过重复打包"
+            echo
+            echo "现有文件:"
+            ls -la "$ARTIFACT_DIR"
+            echo
+            echo "如需强制重新打包，请删除目录: rm -rf $ARTIFACT_DIR"
+            echo "或使用: ./package.sh --force"
+            exit 0
+        else
+            log_warning "版本 $VERSION 存在但不完整，将重新打包"
+            log_info "  现有组件: $existing_components"
+            log_info "  缺少文件: $missing_files"
+        fi
+    else
+        log_warning "版本目录存在但缺少 version.json，将重新打包"
+    fi
+fi
+
+# 创建 artifact 目录
+mkdir -p "$ARTIFACT_DIR"
+log_info "创建输出目录: $ARTIFACT_DIR"
+
+# 创建临时文件存储数据
+TEMP_DIR=$(mktemp -d)
+COMPONENTS_FILE="$TEMP_DIR/components.txt"
+VERSIONS_FILE="$TEMP_DIR/versions.txt"
+DEPENDENCIES_FILE="$TEMP_DIR/dependencies.txt"
+INSTALL_ORDER_FILE="$TEMP_DIR/install_order.txt"
+CHECKSUMS_FILE="$TEMP_DIR/checksums.txt"
+ARTIFACT_LIST_FILE="$TEMP_DIR/artifact_list.txt"
+
+# 解析 checklist 文件
+log_info "解析组件清单..."
+line_num=0
+component_count=0
+
+while IFS= read -r line; do
+    [[ -z "$line" || "$line" =~ ^[[:space:]]*# ]] && continue
+    
+    line_num=$((line_num + 1))
+    
+    # 解析行: 组件名 目录路径 版本 [依赖组件] [安装顺序]
+    read -r component component_path version dep_component order <<< "$line"
+    
+    if [[ -z "$component" || -z "$component_path" || -z "$version" ]]; then
+        log_warning "跳过无效行 $line_num: $line"
+        continue
+    fi
+    
+    # 存储组件信息
+    echo "$component" >> "$COMPONENTS_FILE"
+    echo "$component:$version" >> "$VERSIONS_FILE"
+    echo "$component:$component_path" >> "$TEMP_DIR/component_paths.txt"
+    
+    if [[ -n "$dep_component" && "$dep_component" != "$component" ]]; then
+        echo "$component:$dep_component" >> "$DEPENDENCIES_FILE"
+    fi
+    
+    if [[ -n "$order" && "$order" =~ ^[0-9]+$ ]]; then
+        echo "$order:$component" >> "$INSTALL_ORDER_FILE"
+    else
+        # 如果没有指定顺序，按解析顺序分配
+        echo "$line_num:$component" >> "$INSTALL_ORDER_FILE"
+    fi
+    
+    component_count=$((component_count + 1))
+    log_info "  - $component v$version"
+done < config/checklist
+
+if [[ $component_count -eq 0 ]]; then
+    log_error "没有找到有效的组件"
+    rm -rf "$TEMP_DIR"
+    exit 1
+fi
+
+log_success "找到 $component_count 个组件"
+
+# 检查组件目录是否存在
+log_info "检查组件目录..."
+missing_components=()
+
+while IFS= read -r component; do
+    # 获取组件路径
+    component_path=$(grep "^$component:" "$TEMP_DIR/component_paths.txt" | cut -d':' -f2-)
+    if [[ -z "$component_path" ]]; then
+        log_error "未找到组件 $component 的路径配置"
+        log_info "请检查 component_paths.txt 文件或添加路径配置"
+        exit 1
+    fi
+    
+    if [[ ! -d "$component_path" ]]; then
+        missing_components+=("$component:$component_path")
+    fi
+done < "$COMPONENTS_FILE"
+
+if [[ ${#missing_components[@]} -gt 0 ]]; then
+    log_error "以下组件目录不存在:"
+    for component_path in "${missing_components[@]}"; do
+        echo "  - $component_path"
+    done
+    rm -rf "$TEMP_DIR"
+    exit 1
+fi
+
+# 打包各个组件
+log_info "开始打包组件..."
+
+while IFS= read -r component; do
+    # 获取组件版本和路径
+    version=$(grep "^$component:" "$VERSIONS_FILE" | cut -d':' -f2)
+    component_path=$(grep "^$component:" "$TEMP_DIR/component_paths.txt" | cut -d':' -f2-)
+    if [[ -z "$component_path" ]]; then
+        log_error "未找到组件 $component 的路径配置"
+        log_info "请检查 component_paths.txt 文件或添加路径配置"
+        exit 1
+    fi
+    
+    log_info "打包 $component v$version..."
+    log_info "  组件路径: $component_path"
+    
+    # 进入组件目录
+    cd "$component_path"
+    
+    # 检查组件是否有 package.sh
+    if [[ ! -f "package.sh" ]]; then
+        log_error "$component 缺少 package.sh 文件"
+        cd "$CURRENT_DIR"
+        rm -rf "$TEMP_DIR"
+        exit 1
+    fi
+    
+    # 执行组件的打包脚本
+    if ./package.sh; then
+        # 查找生成的 tar 包
+        tar_file=$(find . -name "*.tar.gz" -type f | head -1)
+        if [[ -n "$tar_file" ]]; then
+            # 移动到 artifact 目录
+            mv "$tar_file" "$CURRENT_DIR/$ARTIFACT_DIR/"
+            tar_filename=$(basename "$tar_file")
+            
+            # 计算校验和
+            checksum=$(sha256sum "$CURRENT_DIR/$ARTIFACT_DIR/$tar_filename" | cut -d' ' -f1)
+            echo "$component:sha256:$checksum" >> "$CHECKSUMS_FILE"
+            echo "$component:$version" >> "$ARTIFACT_LIST_FILE"
+            
+            # 将完整的文件名存储到安装顺序文件中
+            echo "$tar_filename" >> "$TEMP_DIR/install_order_files.txt"
+            
+            log_success "  $component 打包完成: $tar_filename"
+        else
+            log_error "$component 打包失败，未找到生成的 tar 包"
+            cd "$CURRENT_DIR"
+            rm -rf "$TEMP_DIR"
+            exit 1
+        fi
+    else
+        log_error "$component 打包失败"
+        cd "$CURRENT_DIR"
+        rm -rf "$TEMP_DIR"
+        exit 1
+    fi
+    
+    # 返回主目录
+    cd "$CURRENT_DIR"
+done < "$COMPONENTS_FILE"
+
+# 生成 version.json
+log_info "生成版本信息文件..."
+version_json="$ARTIFACT_DIR/version.json"
+
+# 构建依赖关系 JSON
+deps_json=""
+if [[ -f "$DEPENDENCIES_FILE" ]]; then
+    first=true
+    while IFS= read -r line; do
+        component=$(echo "$line" | cut -d':' -f1)
+        dep=$(echo "$line" | cut -d':' -f2)
+        if [[ "$first" == "true" ]]; then
+            deps_json="\"$component\":[\"$dep\"]"
+            first=false
+        else
+            deps_json="$deps_json,\"$component\":[\"$dep\"]"
+        fi
+    done < "$DEPENDENCIES_FILE"
+fi
+
+# 构建安装顺序数组
+order_array=""
+if [[ -f "$TEMP_DIR/install_order_files.txt" ]]; then
+    first=true
+    while IFS= read -r filename; do
+        if [[ "$first" == "true" ]]; then
+            order_array="\"$filename\""
+            first=false
+        else
+            order_array="$order_array,\"$filename\""
+        fi
+    done < "$TEMP_DIR/install_order_files.txt"
+fi
+
+# 构建 artifact_list JSON
+artifact_json=""
+if [[ -f "$ARTIFACT_LIST_FILE" ]]; then
+    first=true
+    while IFS= read -r line; do
+        component=$(echo "$line" | cut -d':' -f1)
+        version=$(echo "$line" | cut -d':' -f2)
+        if [[ "$first" == "true" ]]; then
+            artifact_json="\"$component\":\"$version\""
+            first=false
+        else
+            artifact_json="$artifact_json,\"$component\":\"$version\""
+        fi
+    done < "$ARTIFACT_LIST_FILE"
+fi
+
+# 构建 checksums JSON
+checksums_json=""
+if [[ -f "$CHECKSUMS_FILE" ]]; then
+    first=true
+    while IFS= read -r line; do
+        component=$(echo "$line" | cut -d':' -f1)
+        checksum=$(echo "$line" | cut -d':' -f2-)
+        if [[ "$first" == "true" ]]; then
+            checksums_json="\"$component\":\"$checksum\""
+            first=false
+        else
+            checksums_json="$checksums_json,\"$component\":\"$checksum\""
+        fi
+    done < "$CHECKSUMS_FILE"
+fi
+
+# 生成完整的 version.json
+cat > "$version_json" << EOF
+{
+  "version": "$VERSION",
+  "build_time": "$(date -u +%Y-%m-%dT%H:%M:%SZ)",
+  "artifact_list": {
+    $artifact_json
+  },
+  "checksums": {
+    $checksums_json
+  },
+  "dependencies": {
+    $deps_json
+  },
+  "install_order": [
+    $order_array
+  ]
+}
+EOF
+
+log_success "版本信息文件生成完成: $version_json"
+
+# 复制`安装`脚本到 artifact 目录
+log_info "复制安装脚本..."
+if [[ -f "scripts/install_artifact.sh" ]]; then
+    cp "scripts/install_artifact.sh" "$ARTIFACT_DIR/install.sh"
+    chmod +x "$ARTIFACT_DIR/install.sh"
+    log_success "安装脚本复制完成: $ARTIFACT_DIR/install.sh"
+else
+    log_warning "scripts/install_artifact.sh 文件不存在"
+fi
+
+# 复制`卸载`脚本到 artifact 目录
+log_info "复制卸载脚本..."
+if [[ -f "scripts/uninstall_artifact.sh" ]]; then
+    cp "scripts/uninstall_artifact.sh" "$ARTIFACT_DIR/uninstall.sh"
+    chmod +x "$ARTIFACT_DIR/uninstall.sh"
+    log_success "卸载脚本复制完成: $ARTIFACT_DIR/uninstall.sh"
+else
+    log_warning "scripts/uninstall_artifact.sh 文件不存在"
+fi
+
+# 复制`健康检查`脚本到 artifact 目录
+log_info "复制健康检查脚本..."
+if [[ -f "scripts/check_health.sh" ]]; then
+    cp "scripts/check_health.sh" "$ARTIFACT_DIR/check_health.sh"
+    chmod +x "$ARTIFACT_DIR/check_health.sh"
+    log_success "健康检查脚本复制完成: $ARTIFACT_DIR/check_health.sh"
+else
+    log_warning "scripts/check_health.sh 文件不存在"
+fi
+
+# 复制`DNS 同步`脚本到 artifact 目录
+log_info "复制 DNS 同步脚本..."
+if [[ -f "scripts/sync_dns.sh" ]]; then
+    cp "scripts/sync_dns.sh" "$ARTIFACT_DIR/sync_dns.sh"
+    chmod +x "$ARTIFACT_DIR/sync_dns.sh"
+    log_success "DNS 同步脚本复制完成: $ARTIFACT_DIR/sync_dns.sh"
+else
+    log_warning "scripts/sync_dns.sh 文件不存在"
+fi
+
+# 复制`版本校验`脚本到 artifact 目录
+log_info "复制版本校验脚本..."
+if [[ -f "scripts/check_version.sh" ]]; then
+    cp "scripts/check_version.sh" "$ARTIFACT_DIR/check_version.sh"
+    chmod +x "$ARTIFACT_DIR/check_version.sh"
+    log_success "版本校验脚本复制完成: $ARTIFACT_DIR/check_version.sh"
+else
+    log_warning "scripts/check_version.sh 文件不存在"
+fi
+
+# 复制`自动重启`脚本到 artifact 目录
+log_info "复制自动重启脚本..."
+if [[ -f "scripts/restart_unhealthy.sh" ]]; then
+    cp "scripts/restart_unhealthy.sh" "$ARTIFACT_DIR/restart_unhealthy.sh"
+    chmod +x "$ARTIFACT_DIR/restart_unhealthy.sh"
+    log_success "自动重启脚本复制完成: $ARTIFACT_DIR/restart_unhealthy.sh"
+else
+    log_warning "scripts/restart_unhealthy.sh 文件不存在"
+fi
+
+# 复制配置文件到 artifact 目录
+log_info "复制配置文件..."
+if [[ -f "config/config.env" ]]; then
+    cp "config/config.env" "$ARTIFACT_DIR/"
+    log_success "配置文件复制完成: $ARTIFACT_DIR/config.env"
+else
+    log_warning "config 目录不存在，跳过配置文件复制"
+fi
+
+# DNS 配置文件不需要复制到版本目录，直接从 FTP 服务器根目录获取
+
+# 复制 deps 目录到 artifact 目录
+log_info "复制系统依赖包..."
+if [[ -d "deps" ]]; then
+    cp -r "deps" "$ARTIFACT_DIR/"
+    log_success "系统依赖包复制完成: $ARTIFACT_DIR/deps"
+    
+    # 显示deps目录内容
+    log_info "  依赖包列表:"
+    find "$ARTIFACT_DIR/deps" -name "*.tar.gz" -exec basename {} \; | while read dep_file; do
+        log_info "    - $dep_file"
+    done
+else
+    log_warning "deps 目录不存在，跳过依赖包复制"
+fi
+
+# 显示打包结果
+log_success "打包完成！"
+echo
+echo "版本: $VERSION"
+echo "输出目录: $ARTIFACT_DIR"
+echo "包含组件:"
+if [[ -f "$ARTIFACT_LIST_FILE" ]]; then
+    while IFS= read -r line; do
+        component=$(echo "$line" | cut -d':' -f1)
+        version=$(echo "$line" | cut -d':' -f2)
+        echo "  - $component v$version"
+    done < "$ARTIFACT_LIST_FILE"
+fi
+echo
+echo "文件列表:"
+ls -la "$ARTIFACT_DIR"
+echo
+
+# 清理临时文件
+rm -rf "$TEMP_DIR"