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

@@ -0,0 +1,409 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+show_help() {
+  cat <<'EOF'
+ARGUS Unified Build System - Image Build Tool
+
+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 --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
+    --intranet)
+      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
+      ;;
+    *)
+      echo "Unknown option: $1" >&2
+      show_help
+      exit 1
+      ;;
+  esac
+done
+
+root="$(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="$master_root/offline_wheels"
+
+if [[ "$build_master_offline" == true ]]; then
+  if [[ ! -f "$master_offline_tar" ]]; then
+    echo "❌ offline wheels tar not found: $master_offline_tar" >&2
+    echo "   请提前准备好 offline_wheels.tar.gz 后再执行 --master-offline" >&2
+    exit 1
+  fi
+  echo "📦 Preparing offline wheels for master (extracting $master_offline_tar)"
+  rm -rf "$master_offline_dir"
+  mkdir -p "$master_offline_dir"
+  tar -xzf "$master_offline_tar" -C "$master_root"
+  has_wheel=$(find "$master_offline_dir" -maxdepth 1 -type f -name '*.whl' -print -quit)
+  if [[ -z "$has_wheel" ]]; then
+    echo "❌ offline_wheels extraction failed或无 wheel: $master_offline_dir" >&2
+    exit 1
+  fi
+fi
+
+echo "======================================="
+echo "ARGUS Unified Build System"
+echo "======================================="
+
+if [[ "$use_intranet" == true ]]; then
+  echo "🌐 Mode: Intranet (Using internal mirror: 10.68.64.1)"
+else
+  echo "🌐 Mode: Public (Using default package sources)"
+fi
+
+echo "👤 Build user UID:GID -> ${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}"
+
+echo "📁 Build context: $root"
+echo ""
+
+build_image() {
+  local image_name=$1
+  local dockerfile_path=$2
+  local tag=$3
+  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 "$tag" "$context"; then
+    echo "✅ $image_name image built successfully"
+    return 0
+  else
+    echo "❌ Failed to build $image_name image"
+    return 1
+  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_core" == true ]]; then
+  if build_image "Elasticsearch" "src/log/elasticsearch/build/Dockerfile" "argus-elasticsearch:latest"; then
+    images_built+=("argus-elasticsearch:latest")
+  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 [[ "$build_master" == true ]]; then
+  echo ""
+  echo "🔄 Building Master image..."
+  pushd "$master_root" >/dev/null
+  master_args=("--tag" "argus-master:latest")
+  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 "======================================="
+
+if [[ ${#images_built[@]} -gt 0 ]]; then
+  echo "✅ Successfully built images:"
+  for image in "${images_built[@]}"; do
+    echo "   • $image"
+  done
+fi
+
+if [[ "$build_failed" == true ]]; then
+  echo ""
+  echo "❌ Some images failed to build. Please check the errors above."
+  exit 1
+fi
+
+if [[ "$use_intranet" == true ]]; then
+  echo ""
+  echo "🌐 Built with intranet mirror configuration"
+fi
+
+if [[ "$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"
+echo "   cd src/master/tests && MASTER_IMAGE_TAG=argus-master:offline ./scripts/00_e2e_test.sh"
+echo ""

build/save_images.sh

@@ -0,0 +1,229 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 帮助信息
+show_help() {
+    cat << EOF
+ARGUS Unified Build System - Image Export Tool
+
+Usage: $0 [OPTIONS]
+
+Options:
+  --compress    Compress exported images with gzip
+  -h, --help    Show this help message
+
+Examples:
+  $0                # Export all images without compression
+  $0 --compress     # Export all images with gzip compression
+
+EOF
+}
+
+# 解析命令行参数
+use_compression=false
+
+while [[ $# -gt 0 ]]; do
+    case $1 in
+        --compress)
+            use_compression=true
+            shift
+            ;;
+        -h|--help)
+            show_help
+            exit 0
+            ;;
+        *)
+            echo "Unknown option: $1"
+            show_help
+            exit 1
+            ;;
+    esac
+done
+
+# 获取项目根目录
+root="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+cd "$root"
+
+# 创建镜像输出目录
+images_dir="$root/images"
+mkdir -p "$images_dir"
+
+echo "======================================="
+echo "ARGUS Unified Build System - Image Export"
+echo "======================================="
+echo ""
+
+if [[ "$use_compression" == true ]]; then
+    echo "🗜️  Mode: With gzip compression"
+else
+    echo "📦 Mode: No compression"
+fi
+
+echo "📁 Output directory: $images_dir"
+echo ""
+
+# 定义镜像列表
+declare -A images=(
+    ["argus-elasticsearch:latest"]="argus-elasticsearch-latest.tar"
+    ["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
+check_image() {
+    local image_name="$1"
+    if docker images --format "{{.Repository}}:{{.Tag}}" | grep -q "^$image_name$"; then
+        echo "✅ Image found: $image_name"
+        return 0
+    else
+        echo "❌ Image not found: $image_name"
+        return 1
+    fi
+}
+
+# 函数：显示镜像信息
+show_image_info() {
+    local image_name="$1"
+    echo "📋 Image info for $image_name:"
+    docker images "$image_name" --format "   Size: {{.Size}}, Created: {{.CreatedSince}}, ID: {{.ID}}"
+}
+
+# 函数：保存镜像
+save_image() {
+    local image_name="$1"
+    local output_file="$2"
+    local output_path="$images_dir/$output_file"
+
+    echo "🔄 Saving $image_name to $output_file..."
+
+    # 删除旧的镜像文件（如果存在）
+    if [[ -f "$output_path" ]]; then
+        echo "   Removing existing file: $output_file"
+        rm "$output_path"
+    fi
+
+    if [[ "$use_compression" == true && -f "$output_path.gz" ]]; then
+        echo "   Removing existing compressed file: $output_file.gz"
+        rm "$output_path.gz"
+    fi
+
+    # 保存镜像
+    docker save "$image_name" -o "$output_path"
+
+    if [[ "$use_compression" == true ]]; then
+        echo "   Compressing with gzip..."
+        gzip "$output_path"
+        output_path="$output_path.gz"
+        output_file="$output_file.gz"
+    fi
+
+    # 检查文件大小
+    local file_size=$(du -h "$output_path" | cut -f1)
+    echo "✅ Saved successfully: $output_file ($file_size)"
+}
+
+echo "🔍 Checking for ARGUS images..."
+echo ""
+
+# 检查所有镜像
+available_images=()
+missing_images=()
+
+for image_name in "${!images[@]}"; do
+    if check_image "$image_name"; then
+        show_image_info "$image_name"
+        available_images+=("$image_name")
+    else
+        missing_images+=("$image_name")
+    fi
+    echo ""
+done
+
+# 如果没有镜像存在，提示构建
+if [[ ${#available_images[@]} -eq 0 ]]; then
+    echo "❌ No ARGUS images found to export."
+    echo ""
+    echo "🔧 Please build the images first with:"
+    echo "   ./build/build_images.sh"
+    exit 1
+fi
+
+# 显示缺失的镜像
+if [[ ${#missing_images[@]} -gt 0 ]]; then
+    echo "⚠️  Missing images (will be skipped):"
+    for image_name in "${missing_images[@]}"; do
+        echo "   • $image_name"
+    done
+    echo ""
+fi
+
+echo "💾 Starting image export process..."
+echo ""
+
+# 保存所有可用的镜像
+exported_files=()
+for image_name in "${available_images[@]}"; do
+    output_file="${images[$image_name]}"
+    save_image "$image_name" "$output_file"
+
+    if [[ "$use_compression" == true ]]; then
+        exported_files+=("$output_file.gz")
+    else
+        exported_files+=("$output_file")
+    fi
+    echo ""
+done
+
+echo "======================================="
+echo "📦 Export Summary"
+echo "======================================="
+
+# 显示导出的文件
+echo "📁 Exported files in $images_dir:"
+total_size=0
+for file in "${exported_files[@]}"; do
+    full_path="$images_dir/$file"
+    if [[ -f "$full_path" ]]; then
+        size=$(du -h "$full_path" | cut -f1)
+        size_bytes=$(du -b "$full_path" | cut -f1)
+        total_size=$((total_size + size_bytes))
+        echo "   ✅ $file ($size)"
+    fi
+done
+
+# 显示总大小
+if [[ $total_size -gt 0 ]]; then
+    total_size_human=$(numfmt --to=iec --suffix=B $total_size)
+    echo ""
+    echo "📊 Total size: $total_size_human"
+fi
+
+echo ""
+echo "🚀 Usage instructions:"
+echo "   To load these images on another system:"
+
+if [[ "$use_compression" == true ]]; then
+    for file in "${exported_files[@]}"; do
+        if [[ -f "$images_dir/$file" ]]; then
+            base_name="${file%.gz}"
+            echo "     gunzip $file && docker load -i $base_name"
+        fi
+    done
+else
+    for file in "${exported_files[@]}"; do
+        if [[ -f "$images_dir/$file" ]]; then
+            echo "     docker load -i $file"
+        fi
+    done
+fi
+
+echo ""
+echo "✅ Image export completed successfully!"
+echo ""

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/.gitignore

@@ -0,0 +1,2 @@
+
+__pycache__/

src/agent/.gitignore

@@ -0,0 +1,5 @@
+build/
+*.egg-info/
+__pycache__/
+
+.env

src/agent/README.md

@@ -0,0 +1,78 @@
+# Argus Agent 模块
+
+Argus Agent 是一个轻量级 Python 进程，负责向 Argus Master 注册节点、汇报健康数据，并维护本地持久化信息。模块现以 PyInstaller 打包为独立可执行文件，便于在普通容器或虚机中直接运行。
+
+## 构建可执行文件
+
+```bash
+cd src/agent
+./scripts/build_binary.sh  # 生成 dist/argus-agent
+```
+
+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
+
+Agent 不再依赖配置文件；所有参数均由环境变量与主机名推导：
+
+| 变量 | 必填 | 默认值 | 说明 |
+| --- | --- | --- | --- |
+| `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/<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。
+
+## 日志与持久化
+
+- Agent 会在成功注册、状态上报、异常重试等关键节点输出结构化日志，便于聚合分析。
+- `node.json` 保存 Master 返回的最新节点对象，用于重启后继续使用既有节点 ID。
+
+## 端到端测试
+
+仓库内提供 Docker Compose 测试栈（master + ubuntu 容器）：
+
+```bash
+cd src/agent/tests
+./scripts/00_e2e_test.sh
+```
+
+测试脚本会：
+
+1. 构建 master 镜像与 agent 可执行文件。
+2. 以 `ubuntu:24.04` 启动 agent 容器，并通过环境变量注入 `MASTER_ENDPOINT`、`REPORT_INTERVAL_SECONDS`。
+3. 验证注册、健康上报、nodes.json 生成、统计接口，以及“容器重启 + IP 变化”重注册流程。
+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/client.py

@@ -0,0 +1,60 @@
+from __future__ import annotations
+
+import json
+from typing import Any, Dict, Optional
+
+import requests
+
+from .log import get_logger
+
+LOGGER = get_logger("argus.agent.client")
+
+
+class MasterAPIError(Exception):
+    def __init__(self, message: str, status_code: int, payload: Optional[Dict[str, Any]] = None) -> None:
+        super().__init__(message)
+        self.status_code = status_code
+        self.payload = payload or {}
+
+
+class AgentClient:
+    def __init__(self, base_url: str, *, timeout: int = 10) -> None:
+        self._base_url = base_url.rstrip("/")
+        self._timeout = timeout
+        self._session = requests.Session()
+
+    def register_node(self, body: Dict[str, Any]) -> Dict[str, Any]:
+        """调用 master 注册接口，返回节点对象。"""
+        url = f"{self._base_url}/api/v1/master/nodes"
+        response = self._session.post(url, json=body, timeout=self._timeout)
+        return self._parse_response(response, "Failed to register node")
+
+    def update_status(self, node_id: str, body: Dict[str, Any]) -> Dict[str, Any]:
+        """上报健康信息，由 master 更新 last_report。"""
+        url = f"{self._base_url}/api/v1/master/nodes/{node_id}/status"
+        response = self._session.put(url, json=body, timeout=self._timeout)
+        return self._parse_response(response, "Failed to update node status")
+
+    def _parse_response(self, response: requests.Response, error_prefix: str) -> Dict[str, Any]:
+        content_type = response.headers.get("Content-Type", "")
+        payload: Dict[str, Any] | None = None
+        if "application/json" in content_type:
+            try:
+                payload = response.json()
+            except json.JSONDecodeError:
+                LOGGER.warning("Response contained invalid JSON", extra={"status": response.status_code})
+
+        if response.status_code >= 400:
+            message = payload.get("error") if isinstance(payload, dict) else response.text
+            raise MasterAPIError(
+                f"{error_prefix}: {message}",
+                status_code=response.status_code,
+                payload=payload if isinstance(payload, dict) else None,
+            )
+
+        if payload is None:
+            try:
+                payload = response.json()
+            except json.JSONDecodeError as exc:
+                raise MasterAPIError("Master returned non-JSON payload", response.status_code) from exc
+        return payload

src/agent/app/collector.py

@@ -0,0 +1,110 @@
+from __future__ import annotations
+
+import os
+import re
+import socket
+import subprocess
+from pathlib import Path
+from typing import Any, Dict
+
+from .config import AgentConfig
+from .log import get_logger
+
+LOGGER = get_logger("argus.agent.collector")
+
+_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
+    meta = {
+        "hostname": hostname,
+        "ip": _detect_ip_address(),
+        "env": config.environment,
+        "user": config.user,
+        "instance": config.instance,
+        "cpu_number": _detect_cpu_count(),
+        "memory_in_bytes": _detect_memory_bytes(),
+        "gpu_number": _detect_gpu_count(),
+    }
+    return meta
+
+
+def _parse_hostname(hostname: str) -> tuple[str, str, str]:
+    """按照约定的 env-user-instance 前缀拆解主机名。"""
+    match = _HOSTNAME_PATTERN.match(hostname)
+    if not match:
+        LOGGER.warning("Hostname does not match expected pattern", extra={"hostname": hostname})
+        return "", "", ""
+    return match.group(1), match.group(2), match.group(3)
+
+
+def _detect_cpu_count() -> int:
+    count = os.cpu_count()
+    return count if count is not None else 0
+
+
+def _detect_memory_bytes() -> int:
+    """优先读取 cgroup 限额，失败时退回 /proc/meminfo。"""
+    cgroup_path = Path("/sys/fs/cgroup/memory.max")
+    try:
+        raw = cgroup_path.read_text(encoding="utf-8").strip()
+        if raw and raw != "max":
+            return int(raw)
+    except FileNotFoundError:
+        LOGGER.debug("cgroup memory.max not found, falling back to /proc/meminfo")
+    except ValueError:
+        LOGGER.warning("Failed to parse memory.max, falling back", extra={"value": raw})
+
+    try:
+        with open("/proc/meminfo", "r", encoding="utf-8") as handle:
+            for line in handle:
+                if line.startswith("MemTotal:"):
+                    parts = line.split()
+                    if len(parts) >= 2:
+                        return int(parts[1]) * 1024
+    except FileNotFoundError:
+        LOGGER.error("/proc/meminfo not found; defaulting memory to 0")
+    return 0
+
+
+def _detect_gpu_count() -> int:
+    """采集 GPU 数量，如无法探测则默认为 0。"""
+    try:
+        proc = subprocess.run(
+            ["nvidia-smi", "-L"],
+            check=False,
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            timeout=5,
+        )
+    except FileNotFoundError:
+        LOGGER.debug("nvidia-smi not available; assuming 0 GPUs")
+        return 0
+    except subprocess.SubprocessError as exc:
+        LOGGER.warning("nvidia-smi invocation failed", extra={"error": str(exc)})
+        return 0
+
+    if proc.returncode != 0:
+        LOGGER.debug("nvidia-smi returned non-zero", extra={"stderr": proc.stderr.strip()})
+        return 0
+
+    count = sum(1 for line in proc.stdout.splitlines() if line.strip())
+    return count
+
+
+def _detect_ip_address() -> str:
+    """尝试通过 UDP socket 获得容器出口 IP，失败则回退解析主机名。"""
+    try:
+        with socket.socket(socket.AF_INET, socket.SOCK_DGRAM) as sock:
+            sock.connect(("8.8.8.8", 80))
+            return sock.getsockname()[0]
+    except OSError:
+        LOGGER.debug("UDP socket trick failed; falling back to hostname lookup")
+    try:
+        return socket.gethostbyname(socket.gethostname())
+    except OSError:
+        LOGGER.warning("Unable to resolve hostname to IP; defaulting to 127.0.0.1")
+        return "127.0.0.1"

src/agent/app/config.py

@@ -0,0 +1,141 @@
+from __future__ import annotations
+
+import os
+import socket
+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
+    report_interval_seconds: int
+    health_dir: str
+    request_timeout_seconds: int = 10
+
+
+def _normalise_master_endpoint(value: str) -> str:
+    value = value.strip()
+    if not value:
+        raise ValueError("MASTER_ENDPOINT environment variable is required")
+    if not value.startswith("http://") and not value.startswith("https://"):
+        value = f"http://{value}"
+    return value.rstrip("/")
+
+
+def _read_report_interval(raw_value: str | None) -> int:
+    if raw_value is None or raw_value.strip() == "":
+        return DEFAULT_REPORT_INTERVAL_SECONDS
+    try:
+        interval = int(raw_value)
+    except ValueError as exc:
+        raise ValueError("REPORT_INTERVAL_SECONDS must be an integer") from exc
+    if interval <= 0:
+        raise ValueError("REPORT_INTERVAL_SECONDS must be positive")
+    return interval
+
+
+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"
+    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:
+        raise ValueError("MASTER_ENDPOINT environment variable is not set")
+    master_endpoint = _normalise_master_endpoint(master_endpoint_env)
+
+    report_interval_seconds = _read_report_interval(os.environ.get("REPORT_INTERVAL_SECONDS"))
+
+    Path(node_file).parent.mkdir(parents=True, exist_ok=True)
+    Path(health_dir).mkdir(parents=True, exist_ok=True)
+
+    return AgentConfig(
+        hostname=hostname,
+        environment=environment,
+        user=user,
+        instance=instance,
+        node_file=node_file,
+        version=VERSION,
+        master_endpoint=master_endpoint,
+        report_interval_seconds=report_interval_seconds,
+        health_dir=health_dir,
+    )

src/agent/app/health_reader.py

@@ -0,0 +1,32 @@
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from typing import Any, Dict
+
+from .log import get_logger
+
+LOGGER = get_logger("argus.agent.health")
+
+
+def read_health_directory(path: str) -> Dict[str, Any]:
+    """读取目录中所有 <prefix>-*.json 文件并返回 JSON 映射。"""
+    result: Dict[str, Any] = {}
+    directory = Path(path)
+    if not directory.exists():
+        LOGGER.debug("Health directory does not exist", extra={"path": str(directory)})
+        return result
+
+    for health_file in sorted(directory.glob("*.json")):
+        if "-" not in health_file.stem:
+            LOGGER.debug("Skipping non-prefixed health file", extra={"file": health_file.name})
+            continue
+        try:
+            with health_file.open("r", encoding="utf-8") as handle:
+                content = json.load(handle)
+            result[health_file.stem] = content
+        except json.JSONDecodeError as exc:
+            LOGGER.warning("Failed to parse health file", extra={"file": health_file.name, "error": str(exc)})
+        except OSError as exc:
+            LOGGER.warning("Failed to read health file", extra={"file": health_file.name, "error": str(exc)})
+    return result

src/agent/app/log.py

@@ -0,0 +1,18 @@
+from __future__ import annotations
+
+import logging
+import os
+
+
+_LOG_FORMAT = "%(asctime)s %(levelname)s %(name)s - %(message)s"
+
+
+def setup_logging() -> None:
+    level_name = os.environ.get("AGENT_LOG_LEVEL", "INFO").upper()
+    level = getattr(logging, level_name, logging.INFO)
+    logging.basicConfig(level=level, format=_LOG_FORMAT)
+
+
+def get_logger(name: str) -> logging.Logger:
+    setup_logging()
+    return logging.getLogger(name)

src/agent/app/main.py

@@ -0,0 +1,163 @@
+from __future__ import annotations
+
+import signal
+import time
+from datetime import datetime, timezone
+from typing import Optional
+
+from .client import AgentClient, MasterAPIError
+from .collector import collect_metadata
+from .config import AgentConfig, load_config
+from .health_reader import read_health_directory
+from .log import get_logger, setup_logging
+from .state import clear_node_state, load_node_state, save_node_state
+
+LOGGER = get_logger("argus.agent")
+
+
+def _current_timestamp() -> str:
+    return datetime.now(timezone.utc).replace(microsecond=0).isoformat().replace("+00:00", "Z")
+
+
+class StopSignal:
+    def __init__(self) -> None:
+        self._stop = False
+
+    def set(self, *_args) -> None:  # type: ignore[override]
+        self._stop = True
+
+    def is_set(self) -> bool:
+        return self._stop
+
+
+def main(argv: Optional[list[str]] = None) -> int:  # noqa: ARG001 - 保留签名以兼容入口调用
+    setup_logging()
+
+    stop_signal = StopSignal()
+    signal.signal(signal.SIGTERM, stop_signal.set)
+    signal.signal(signal.SIGINT, stop_signal.set)
+
+    try:
+        config = load_config()
+    except Exception as exc:
+        LOGGER.error("Failed to load configuration", extra={"error": str(exc)})
+        return 1
+
+    LOGGER.info(
+        "Agent starting",
+        extra={
+            "hostname": config.hostname,
+            "master_endpoint": config.master_endpoint,
+            "node_file": config.node_file,
+        },
+    )
+
+    client = AgentClient(config.master_endpoint, timeout=config.request_timeout_seconds)
+
+    node_state = load_node_state(config.node_file) or {}
+    node_id = node_state.get("id")
+
+    # 与 master 建立注册关系（支持重注册），失败则重试
+    register_response = _register_with_retry(client, config, node_id, stop_signal)
+    if register_response is None:
+        LOGGER.info("Registration aborted due to shutdown signal")
+        return 0
+
+    node_id = register_response.get("id")
+    if not node_id:
+        LOGGER.error("Master did not return node id; aborting")
+        return 1
+    save_node_state(config.node_file, register_response)
+
+    LOGGER.info("Entering status report loop", extra={"node_id": node_id})
+    _status_loop(client, config, node_id, stop_signal)
+    return 0
+
+
+def _register_with_retry(
+    client: AgentClient,
+    config: AgentConfig,
+    node_id: Optional[str],
+    stop_signal: StopSignal,
+):
+    backoff = 5
+    while not stop_signal.is_set():
+        payload = {
+            "name": config.hostname,
+            "type": "agent",
+            "meta_data": collect_metadata(config),
+            "version": config.version,
+        }
+        if node_id:
+            payload["id"] = node_id
+
+        try:
+            response = client.register_node(payload)
+            LOGGER.info("Registration successful", extra={"node_id": response.get("id")})
+            save_node_state(config.node_file, response)
+            return response
+        except MasterAPIError as exc:
+            if exc.status_code == 404 and node_id:
+                LOGGER.warning(
+                    "Master does not recognise node id; clearing local node state",
+                    extra={"node_id": node_id},
+                )
+                clear_node_state(config.node_file)
+                node_id = None
+            elif exc.status_code == 500 and node_id:
+                # id 与 name 不匹配通常意味着配置异常，记录但继续重试
+                LOGGER.error(
+                    "Master rejected node due to id/name mismatch; will retry",
+                    extra={"node_id": node_id},
+                )
+            else:
+                LOGGER.error("Registration failed", extra={"status_code": exc.status_code, "error": str(exc)})
+            time.sleep(min(backoff, 60))
+            backoff = min(backoff * 2, 60)
+        except Exception as exc:  # pragma: no cover - defensive
+            LOGGER.exception("Unexpected error during registration", extra={"error": str(exc)})
+            time.sleep(min(backoff, 60))
+            backoff = min(backoff * 2, 60)
+    return None
+
+
+def _status_loop(
+    client: AgentClient,
+    config: AgentConfig,
+    node_id: str,
+    stop_signal: StopSignal,
+) -> None:
+    interval = config.report_interval_seconds
+    while not stop_signal.is_set():
+        timestamp = _current_timestamp()
+        health_payload = read_health_directory(config.health_dir)
+        body = {
+            "timestamp": timestamp,
+            "health": health_payload,
+        }
+        try:
+            response = client.update_status(node_id, body)
+            LOGGER.info(
+                "Status report succeeded",
+                extra={"node_id": node_id, "health_keys": list(health_payload.keys())},
+            )
+            save_node_state(config.node_file, response)
+        except MasterAPIError as exc:
+            # 保持循环继续执行，等待下一次重试
+            LOGGER.error(
+                "Failed to report status",
+                extra={"status_code": exc.status_code, "error": str(exc)},
+            )
+        except Exception as exc:  # pragma: no cover - defensive
+            LOGGER.exception("Unexpected error during status report", extra={"error": str(exc)})
+
+        for _ in range(interval):
+            if stop_signal.is_set():
+                break
+            time.sleep(1)
+
+    LOGGER.info("Stop signal received; exiting status loop")
+
+
+if __name__ == "__main__":
+    sys.exit(main())

src/agent/app/state.py

@@ -0,0 +1,44 @@
+from __future__ import annotations
+
+import json
+import os
+import tempfile
+from pathlib import Path
+from typing import Any, Dict, Optional
+
+from .log import get_logger
+
+LOGGER = get_logger("argus.agent.state")
+
+
+def load_node_state(path: str) -> Optional[Dict[str, Any]]:
+    """读取本地 node.json，容器重启后沿用之前的 ID。"""
+    try:
+        with open(path, "r", encoding="utf-8") as handle:
+            return json.load(handle)
+    except FileNotFoundError:
+        return None
+    except json.JSONDecodeError as exc:
+        LOGGER.warning("node.json is invalid JSON; ignoring", extra={"error": str(exc)})
+        return None
+
+
+def save_node_state(path: str, data: Dict[str, Any]) -> None:
+    """原子化写入 node.json，避免并发读取坏数据。"""
+    directory = Path(path).parent
+    directory.mkdir(parents=True, exist_ok=True)
+    with tempfile.NamedTemporaryFile("w", dir=directory, delete=False, encoding="utf-8") as tmp:
+        json.dump(data, tmp, separators=(",", ":"))
+        tmp.flush()
+        os.fsync(tmp.fileno())
+        temp_path = tmp.name
+    os.replace(temp_path, path)
+
+
+def clear_node_state(path: str) -> None:
+    try:
+        os.remove(path)
+    except FileNotFoundError:
+        return
+    except OSError as exc:
+        LOGGER.warning("Failed to remove node state file", extra={"error": str(exc), "path": path})

src/agent/app/version.py

@@ -0,0 +1,69 @@
+from __future__ import annotations
+
+import os
+import sys
+from pathlib import Path
+from typing import Optional
+
+import importlib.metadata
+
+try:
+    import tomllib
+except ModuleNotFoundError:  # pragma: no cover
+    import tomli as tomllib  # type: ignore[no-redef]
+
+
+def _candidate_paths() -> list[Path]:
+    paths = []
+    bundle_dir: Optional[str] = getattr(sys, "_MEIPASS", None)
+    if bundle_dir:
+        paths.append(Path(bundle_dir) / "pyproject.toml")
+    paths.append(Path(__file__).resolve().parent.parent / "pyproject.toml")
+    paths.append(Path(__file__).resolve().parent / "pyproject.toml")
+    paths.append(Path.cwd() / "pyproject.toml")
+    return paths
+
+
+def _read_from_pyproject() -> Optional[str]:
+    for path in _candidate_paths():
+        if not path.exists():
+            continue
+        try:
+            with path.open("rb") as handle:
+                data = tomllib.load(handle)
+        except (OSError, tomllib.TOMLDecodeError):
+            continue
+        project = data.get("project")
+        if isinstance(project, dict):
+            version = project.get("version")
+            if isinstance(version, str):
+                return version
+        tool = data.get("tool")
+        if isinstance(tool, dict):
+            argus_cfg = tool.get("argus")
+            if isinstance(argus_cfg, dict):
+                version = argus_cfg.get("version")
+                if isinstance(version, str):
+                    return version
+    return None
+
+
+def _detect_version() -> str:
+    try:
+        return importlib.metadata.version("argus-agent")
+    except importlib.metadata.PackageNotFoundError:
+        pass
+    override = os.environ.get("AGENT_VERSION_OVERRIDE")
+    if override:
+        return override
+    fallback = _read_from_pyproject()
+    if fallback:
+        return fallback
+    return "0.0.0"
+
+
+VERSION: str = _detect_version()
+
+
+def get_version() -> str:
+    return VERSION

src/agent/entry.py

@@ -0,0 +1,10 @@
+#!/usr/bin/env python3
+from __future__ import annotations
+
+import sys
+
+from app.main import main as agent_main
+
+
+if __name__ == "__main__":
+    sys.exit(agent_main())

src/agent/pyproject.toml

@@ -0,0 +1,19 @@
+[project]
+name = "argus-agent"
+version = "1.1.0"
+description = "Argus agent binary"
+readme = "README.md"
+requires-python = ">=3.11"
+dependencies = [
+    "requests==2.31.0"
+]
+
+[build-system]
+requires = ["setuptools>=69", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.argus]
+entry = "app.main:main"
+
+[tool.setuptools]
+packages = ["app"]

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

@@ -0,0 +1,276 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+MODULE_ROOT="$(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="$BUILD_ROOT/venv"
+
+AGENT_BUILD_IMAGE="${AGENT_BUILD_IMAGE:-python:3.11-slim-bullseye}"
+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
+
+run_host_build() {
+  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 [ "${AGENT_BUILD_IGNORE_PROXY:-1}" = "1" ]; then
+  unset http_proxy https_proxy HTTP_PROXY HTTPS_PROXY PIP_INDEX_URL PIP_EXTRA_INDEX_URL PIP_TRUSTED_HOST
+fi
+pip install --upgrade pip
+pip install .
+pip install pyinstaller==6.6.0
+pyinstaller \
+  --clean \
+  --onefile \
+  --name argus-agent \
+  --distpath dist \
+  --workpath build/pyinstaller/work \
+  --specpath build/pyinstaller/spec \
+  --add-data /workspace/pyproject.toml:. \
+  entry.py
+chmod +x dist/argus-agent
+
+TARGET_UID="${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/.gitignore

@@ -0,0 +1,2 @@
+private/
+tmp/

src/agent/tests/docker-compose.yml

@@ -0,0 +1,99 @@
+services:
+  bind:
+    image: ${BIND_IMAGE_TAG:-argus-bind9:latest}
+    container_name: argus-bind-agent-e2e
+    volumes:
+      - ./private:/private
+    networks:
+      default:
+        ipv4_address: 172.28.0.2
+    environment:
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
+    restart: always
+
+  master:
+    image: argus-master:latest
+    container_name: argus-master-agent-e2e
+    depends_on:
+      - bind
+    environment:
+      - OFFLINE_THRESHOLD_SECONDS=6
+      - ONLINE_THRESHOLD_SECONDS=2
+      - SCHEDULER_INTERVAL_SECONDS=1
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
+    ports:
+      - "32300:3000"
+    volumes:
+      - ./private/argus/master:/private/argus/master
+      - ./private/argus/metric/prometheus:/private/argus/metric/prometheus
+      - ./private/argus/etc:/private/argus/etc
+    networks:
+      default:
+        ipv4_address: 172.28.0.10
+    restart: always
+
+  agent:
+    image: ubuntu:22.04
+    container_name: argus-agent-e2e
+    hostname: dev-e2euser-e2einst-pod-0
+    depends_on:
+      - master
+      - bind
+    environment:
+      - MASTER_ENDPOINT=http://master.argus.com:3000
+      - REPORT_INTERVAL_SECONDS=2
+      - "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
+      - "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
+    volumes:
+      - ./private/argus/agent/dev-e2euser-e2einst-pod-0:/private/argus/agent/dev-e2euser-e2einst-pod-0
+      - ./private/argus/agent/dev-e2euser-e2einst-pod-0/health:/private/argus/agent/dev-e2euser-e2einst-pod-0/health
+      - ./private/argus/etc:/private/argus/etc
+      - ../dist/argus-agent:/usr/local/bin/argus-agent:ro
+      - ./scripts/agent_entrypoint.sh:/usr/local/bin/agent-entrypoint.sh:ro
+      - ../scripts/agent_deployment_verify.sh:/usr/local/bin/agent_deployment_verify.sh:ro
+    entrypoint:
+      - /usr/local/bin/agent-entrypoint.sh
+    networks:
+      default:
+        ipv4_address: 172.28.0.20
+    restart: always
+
+  agent_env:
+    image: ubuntu:22.04
+    container_name: argus-agent-env-e2e
+    hostname: host_abc
+    depends_on:
+      - master
+      - bind
+    environment:
+      - MASTER_ENDPOINT=http://master.argus.com:3000
+      - REPORT_INTERVAL_SECONDS=2
+      - AGENT_ENV=prod
+      - AGENT_USER=ml
+      - AGENT_INSTANCE=node-3
+      - AGENT_HOSTNAME=host_abc
+      - "ARGUS_BUILD_UID=${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:
+    driver: bridge
+    ipam:
+      driver: default
+      config:
+        - subnet: 172.28.0.0/16

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

@@ -0,0 +1,23 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+SCRIPTS=(
+  "01_bootstrap.sh"
+  "02_up.sh"
+  "03_wait_and_assert_registration.sh"
+  "04_write_health_files.sh"
+  "05_verify_agent.sh"
+  "06_assert_status_on_master.sh"
+  "07_restart_agent_and_reregister.sh"
+  "08_down.sh"
+)
+
+for script in "${SCRIPTS[@]}"; do
+  echo "[TEST] Running $script"
+  "$SCRIPT_DIR/$script"
+  echo "[TEST] $script completed"
+  echo
+done
+
+echo "[TEST] Agent module E2E tests completed"

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

@@ -0,0 +1,63 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+AGENT_ROOT="$(cd "$TEST_ROOT/.." && pwd)"
+MASTER_ROOT="$(cd "$AGENT_ROOT/../master" && pwd)"
+REPO_ROOT="$(cd "$AGENT_ROOT/../.." && pwd)"
+PRIVATE_ROOT="$TEST_ROOT/private"
+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/$AGENT_HOSTNAME/health"
+MASTER_PRIVATE_DIR="$PRIVATE_ROOT/argus/master"
+METRIC_PRIVATE_DIR="$PRIVATE_ROOT/argus/metric/prometheus"
+DNS_DIR="$PRIVATE_ROOT/argus/etc"
+BIND_IMAGE_TAG="${BIND_IMAGE_TAG:-argus-bind9:latest}"
+BIND_ROOT="$(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"
+mkdir -p "$METRIC_PRIVATE_DIR"
+mkdir -p "$TMP_ROOT"
+mkdir -p "$DNS_DIR"
+
+touch "$AGENT_HEALTH_DIR/.keep"
+
+# 中文提示：准备 bind 模块提供的 update-dns.sh，模拟生产下发
+if [[ -f "$BIND_ROOT/build/update-dns.sh" ]]; then
+  cp "$BIND_ROOT/build/update-dns.sh" "$DNS_DIR/update-dns.sh"
+  chmod +x "$DNS_DIR/update-dns.sh"
+else
+  echo "[WARN] bind update script missing at $BIND_ROOT/build/update-dns.sh"
+fi
+
+ensure_image "argus-master:latest"
+ensure_image "$BIND_IMAGE_TAG"
+
+AGENT_BINARY="$AGENT_ROOT/dist/argus-agent"
+
+pushd "$AGENT_ROOT" >/dev/null
+./scripts/build_binary.sh
+popd >/dev/null
+
+if [[ ! -x "$AGENT_BINARY" ]]; then
+  echo "[ERROR] Agent binary not found at $AGENT_BINARY" >&2
+  exit 1
+fi
+
+echo "$AGENT_BINARY" > "$TMP_ROOT/agent_binary_path"
+echo "$BIND_IMAGE_TAG" > "$TMP_ROOT/bind_image_tag"
+
+echo "[INFO] Agent E2E bootstrap complete"

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

@@ -0,0 +1,53 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+REPO_ROOT="$(cd "$TEST_ROOT/../../.." && pwd)"
+
+TMP_ROOT="$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
+  exit 1
+fi
+
+AGENT_BINARY="$(cat "$TMP_ROOT/agent_binary_path")"
+if [[ ! -x "$AGENT_BINARY" ]]; then
+  echo "[ERROR] Agent binary not executable: $AGENT_BINARY" >&2
+  exit 1
+fi
+
+BIND_IMAGE_TAG_VALUE="argus-bind9:latest"
+if [[ -f "$TMP_ROOT/bind_image_tag" ]]; then
+  BIND_IMAGE_TAG_VALUE="$(cat "$TMP_ROOT/bind_image_tag")"
+fi
+
+compose() {
+  if docker compose version >/dev/null 2>&1; then
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+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
+
+pushd "$TEST_ROOT" >/dev/null
+compose down --remove-orphans || true
+BIND_IMAGE_TAG="$BIND_IMAGE_TAG_VALUE" compose up -d
+popd >/dev/null
+
+echo "[INFO] Master+Agent stack started"

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

@@ -0,0 +1,106 @@
+#!/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"
+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 "$TMP_ROOT"
+
+primary_node_id=""
+env_node_id=""
+for _ in {1..30}; do
+  sleep 2
+  response=$(curl -sS "$API_BASE/nodes" || true)
+  if [[ -z "$response" ]]; then
+    continue
+  fi
+  list_file="$TMP_ROOT/nodes_list.json"
+  echo "$response" > "$list_file"
+  readarray -t node_ids < <(python3 - "$list_file" "$AGENT_HOSTNAME" "$ENV_AGENT_HOSTNAME" <<'PY'
+import json, sys
+
+with open(sys.argv[1]) as handle:
+    nodes = json.load(handle)
+
+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
+  )
+
+  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 "$primary_node_id" ]]; then
+  echo "[ERROR] Primary agent did not register within timeout" >&2
+  exit 1
+fi
+
+if [[ -z "$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 $NODE_FILE" >&2
+  exit 1
+fi
+
+python3 - "$NODE_FILE" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(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/$primary_node_id" -o "$detail_file"
+python3 - "$detail_file" "$TMP_ROOT/initial_ip" <<'PY'
+import json, sys, pathlib
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+ip = node["meta_data"].get("ip")
+if not ip:
+    raise SystemExit("meta_data.ip missing")
+pathlib.Path(sys.argv[2]).write_text(ip)
+PY
+
+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

@@ -0,0 +1,22 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+HEALTH_DIR="$TEST_ROOT/private/argus/agent/dev-e2euser-e2einst-pod-0/health"
+
+cat > "$HEALTH_DIR/log-fluentbit.json" <<JSON
+{
+  "status": "healthy",
+  "timestamp": "2023-10-05T12:05:00Z"
+}
+JSON
+
+cat > "$HEALTH_DIR/metric-node-exporter.json" <<JSON
+{
+  "status": "healthy",
+  "timestamp": "2023-10-05T12:05:00Z"
+}
+JSON
+
+echo "[INFO] Health files written"

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

@@ -0,0 +1,78 @@
+#!/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="$(cat "$TMP_ROOT/node_id_host_abc")"
+ENV_HOSTNAME="host_abc"
+NODES_JSON="$TEST_ROOT/private/argus/metric/prometheus/nodes.json"
+
+success=false
+detail_file="$TMP_ROOT/agent_status_detail.json"
+for _ in {1..20}; do
+  sleep 2
+  if ! curl -sS "$API_BASE/nodes/$NODE_ID" -o "$detail_file"; then
+    continue
+  fi
+  if python3 - "$detail_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+if node["status"] != "online":
+    raise SystemExit(1)
+health = node.get("health", {})
+if "log-fluentbit" not in health or "metric-node-exporter" not in health:
+    raise SystemExit(1)
+PY
+  then
+    success=true
+    break
+  fi
+done
+
+if [[ "$success" != true ]]; then
+  echo "[ERROR] Node did not report health data in time" >&2
+  exit 1
+fi
+
+if [[ ! -f "$NODES_JSON" ]]; then
+  echo "[ERROR] nodes.json missing at $NODES_JSON" >&2
+  exit 1
+fi
+
+python3 - "$NODES_JSON" "$NODE_ID" "$ENV_NODE_ID" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    nodes = json.load(handle)
+
+expected_primary = sys.argv[2]
+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/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

@@ -0,0 +1,36 @@
+#!/usr/bin/env bash
+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
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+docker container rm -f argus-agent-e2e argus-agent-env-e2e >/dev/null 2>&1 || true
+
+pushd "$TEST_ROOT" >/dev/null
+compose down --remove-orphans
+popd >/dev/null
+
+if [[ -d "$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

@@ -0,0 +1,79 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+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
+    break
+  fi
+  log "等待 update-dns.sh 准备就绪..."
+  sleep 1
+done
+
+if [[ -x "$DNS_SCRIPT" ]]; then
+  log "执行 update-dns.sh 更新容器 DNS"
+  while true; do
+    if "$DNS_SCRIPT"; then
+      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
+
+# 中文提示：记录当前 dns.conf 内容，便于排查
+if [[ -f "$DNS_CONF" ]]; then
+  log "dns.conf 内容: $(tr '\n' ' ' < "$DNS_CONF")"
+else
+  log "dns.conf 暂未生成"
+fi
+
+# 中文提示：尝试解析 master 域名，失败不阻塞但会打日志
+for _ in {1..30}; do
+  if getent hosts "$TARGET_DOMAIN" >/dev/null 2>&1; then
+    MASTER_IP=$(getent hosts "$TARGET_DOMAIN" | awk '{print $1}' | head -n 1)
+    log "master.argus.com 解析成功: $MASTER_IP"
+    break
+  fi
+  sleep 1
+done
+
+log "启动 argus-agent"
+exec su -s /bin/bash -c /usr/local/bin/argus-agent "$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/.gitignore

@@ -0,0 +1,2 @@
+
+images/

src/bind/build/Dockerfile

@@ -0,0 +1,90 @@
+FROM ubuntu:22.04
+
+# Set timezone and avoid interactive prompts
+ENV DEBIAN_FRONTEND=noninteractive
+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 [ "$USE_INTRANET" = "true" ]; then \
+        echo "Configuring intranet apt sources..." && \
+        cp /etc/apt/sources.list /etc/apt/sources.list.bak && \
+        echo "deb [trusted=yes] http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
+        echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
+    fi
+
+# Update package list and install required packages
+RUN apt-get update && \
+    apt-get install -y \
+    bind9 \
+    bind9utils \
+    dnsutils \
+    bind9-doc \
+    supervisor \
+    net-tools \
+    inetutils-ping \
+    vim \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# 调整 bind 用户与用户组 ID 以匹配宿主机配置
+RUN set -eux; \
+    current_gid="$(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; \
+    fi
+
+# Create supervisor configuration directory
+RUN mkdir -p /etc/supervisor/conf.d
+
+# Copy supervisor configuration
+COPY src/bind/build/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+
+# Copy BIND9 configuration files
+COPY src/bind/build/named.conf.local /etc/bind/named.conf.local
+COPY src/bind/build/db.argus.com /etc/bind/db.argus.com
+
+# Copy startup and reload scripts
+COPY src/bind/build/startup.sh /usr/local/bin/startup.sh
+COPY src/bind/build/reload-bind9.sh /usr/local/bin/reload-bind9.sh
+COPY src/bind/build/argus_dns_sync.sh /usr/local/bin/argus_dns_sync.sh
+COPY src/bind/build/update-dns.sh /usr/local/bin/update-dns.sh
+
+# Make scripts executable
+RUN chmod +x /usr/local/bin/startup.sh /usr/local/bin/reload-bind9.sh  /usr/local/bin/argus_dns_sync.sh /usr/local/bin/update-dns.sh
+
+# Set proper ownership for BIND9 files
+RUN chown bind:bind /etc/bind/named.conf.local /etc/bind/db.argus.com
+
+# Expose DNS port
+EXPOSE 53/tcp 53/udp
+
+# Use root user as requested
+USER root
+
+# Start with startup script
+CMD ["/usr/local/bin/startup.sh"]

src/bind/build/argus_dns_sync.sh

@@ -0,0 +1,106 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+WATCH_DIR="/private/argus/etc"
+ZONE_DB="/private/argus/bind/db.argus.com"
+LOCKFILE="/var/lock/argus_dns_sync.lock"
+BACKUP_DIR="/private/argus/bind/.backup"
+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="$1"
+  [[ "$ip" =~ ^([0-9]{1,3}\.){3}[0-9]{1,3}$ ]] || return 1
+  IFS='.' read -r a b c d <<<"$ip"
+  for n in "$a" "$b" "$c" "$d"; do
+    (( n >= 0 && n <= 255 )) || return 1
+  done
+  return 0
+}
+
+get_current_ip() {
+  local name="$1"
+  sed -n -E "s/^${name}[[:space:]]+IN[[:space:]]+A[[:space:]]+([0-9.]+)[[:space:]]*$/\1/p" "$ZONE_DB" | head -n1
+}
+
+upsert_record() {
+  local name="$1"
+  local new_ip="$2"
+  local ts
+  ts="$(date +%Y%m%d-%H%M%S)"
+  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 "$name" || true)"
+
+  if [[ -z "$cur_ip" ]]; then
+    # Ensure the file ends with a newline before adding new record
+    if [[ -s "$ZONE_DB" ]] && [[ $(tail -c1 "$ZONE_DB" | wc -l) -eq 0 ]]; then
+      echo "" >> "$ZONE_DB"
+    fi
+    printf "%-20s IN A %s\n" "$name" "$new_ip" >> "$ZONE_DB"
+    echo "[ADD] ${name} -> ${new_ip}"
+    changed=1
+  elif [[ "$cur_ip" != "$new_ip" ]]; then
+    awk -v n="$name" -v ip="$new_ip" '
+      {
+        if ($1==n && $2=="IN" && $3=="A") {
+          printf "%-20s IN A %s\n", n, ip
+        } else {
+          print
+        }
+      }
+    ' "$ZONE_DB" > "${ZONE_DB}.tmp" && mv "${ZONE_DB}.tmp" "$ZONE_DB"
+    echo "[UPDATE] ${name}: ${cur_ip} -> ${new_ip}"
+    changed=1
+  else
+    echo "[SKIP] ${name} unchanged (${new_ip})"
+  fi
+
+  if [[ $changed -eq 1 ]]; then
+    return 0
+  fi
+  return 1
+}
+
+while true; do
+  exec 9>"$LOCKFILE"
+  if flock -n 9; then
+    shopt -s nullglob
+    NEED_RELOAD=0
+
+  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)"
+
+      if [[ -z "$ip" ]] || ! is_ipv4 "$ip"; then
+        echo "[WARN] $f 未找到有效 IPv4，跳过"
+        continue
+      fi
+
+      if upsert_record "$name" "$ip"; then
+        NEED_RELOAD=1
+      fi
+    done
+
+    if [[ $NEED_RELOAD -eq 1 ]]; then
+      echo "[INFO] 检测到 db.argus.com 变更，执行 reload-bind9.sh"
+      bash "$RELOAD_SCRIPT"
+    fi
+
+    flock -u 9
+  else
+    echo "[INFO] 已有同步任务在运行，跳过本轮"
+  fi
+
+  sleep "$SLEEP_SECONDS"
+done

src/bind/build/db.argus.com

@@ -0,0 +1,16 @@
+$TTL    604800
+@       IN      SOA     ns1.argus.com. admin.argus.com. (
+                              2         ; Serial
+                         604800         ; Refresh
+                          86400         ; Retry
+                        2419200         ; Expire
+                         604800 )       ; Negative Cache TTL
+
+; 定义 DNS 服务器
+@       IN      NS      ns1.argus.com.
+
+; 定义 ns1 主机
+ns1     IN      A       127.0.0.1
+
+; 定义 web 指向 12.4.5.6
+web     IN      A       12.4.5.6

src/bind/build/dns-monitor.sh

@@ -0,0 +1,71 @@
+#!/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"
+
+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 "$DNS_BACKUP" ]; then
+            # 比较文件内容
+            if ! cmp -s "$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/bind/build/named.conf.local

@@ -0,0 +1,4 @@
+zone "argus.com" {
+    type master;
+    file "/etc/bind/db.argus.com";
+};

src/bind/build/reload-bind9.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+
+echo "Reloading BIND9 configuration..."
+
+# Check if configuration files are valid
+echo "Checking named.conf.local syntax..."
+if ! named-checkconf /etc/bind/named.conf.local; then
+    echo "ERROR: named.conf.local has syntax errors!"
+    exit 1
+fi
+
+echo "Checking zone file syntax..."
+if ! named-checkzone argus.com /etc/bind/db.argus.com; then
+    echo "ERROR: db.argus.com has syntax errors!"
+    exit 1
+fi
+
+# Reload BIND9 via supervisor
+echo "Reloading BIND9 service..."
+supervisorctl restart bind9
+
+if [ $? -eq 0 ]; then
+    echo "BIND9 reloaded successfully!"
+else
+    echo "ERROR: Failed to reload BIND9!"
+    exit 1
+fi

src/bind/build/startup.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+# Set /private permissions to 777 as requested
+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
+    cp /etc/bind/named.conf.local /private/argus/bind/named.conf.local
+fi
+
+if [ ! -f /private/argus/bind/db.argus.com ]; then
+    cp /etc/bind/db.argus.com /private/argus/bind/db.argus.com
+fi
+
+# Copy update-dns.sh to /private/argus/etc/
+cp /usr/local/bin/update-dns.sh /private/argus/etc/update-dns.sh
+chown bind:bind /private/argus/etc/update-dns.sh
+chmod a+x /private/argus/etc/update-dns.sh
+
+# Create symlinks to use persistent configs
+ln -sf /private/argus/bind/named.conf.local /etc/bind/named.conf.local
+ln -sf /private/argus/bind/db.argus.com /etc/bind/db.argus.com
+
+# Set proper ownership
+chown bind:bind /private/argus/bind/named.conf.local /private/argus/bind/db.argus.com
+
+# 记录容器ip地址更新到dns.conf
+IP=`ifconfig | grep -A 1 eth0 | grep inet | awk '{print $2}'`
+echo current IP: ${IP}
+echo ${IP} > /private/argus/etc/dns.conf
+
+# Create supervisor log directory
+mkdir -p /var/log/supervisor
+
+# Start supervisor
+exec /usr/bin/supervisord -c /etc/supervisor/conf.d/supervisord.conf

src/bind/build/supervisord.conf

@@ -0,0 +1,37 @@
+[unix_http_server]
+file=/var/run/supervisor.sock
+chmod=0700
+
+[supervisord]
+nodaemon=true
+user=root
+logfile=/var/log/supervisor/supervisord.log
+pidfile=/var/run/supervisord.pid
+
+[rpcinterface:supervisor]
+supervisor.rpcinterface_factory = supervisor.rpcinterface:make_main_rpcinterface
+
+[supervisorctl]
+serverurl=unix:///var/run/supervisor.sock
+
+[program:bind9]
+command=/usr/sbin/named -g -c /etc/bind/named.conf -u bind
+user=bind
+autostart=true
+autorestart=true
+stderr_logfile=/var/log/supervisor/bind9.err.log
+stdout_logfile=/var/log/supervisor/bind9.out.log
+priority=10
+
+[program:argus-dns-sync]
+command=/usr/local/bin/argus_dns_sync.sh
+autostart=true
+autorestart=true
+startsecs=3
+stopsignal=TERM
+user=root
+stdout_logfile=/var/log/argus_dns_sync.out.log
+stderr_logfile=/var/log/argus_dns_sync.err.log
+; 根据环境调整环境变量（可选）
+; environment=RNDC_RELOAD="yes"
+

src/bind/build/update-dns.sh

@@ -0,0 +1,31 @@
+#!/bin/sh
+# update-dns.sh
+# 从 /private/argus/etc/dns.conf 读取 IP，写入 /etc/resolv.conf
+
+DNS_CONF="/private/argus/etc/dns.conf"
+RESOLV_CONF="/etc/resolv.conf"
+
+# 检查配置文件是否存在
+if [ ! -f "$DNS_CONF" ]; then
+  echo "配置文件不存在: $DNS_CONF" >&2
+  exit 1
+fi
+
+# 生成 resolv.conf 内容
+{
+  while IFS= read -r ip; do
+    # 跳过空行和注释
+    case "$ip" in
+      \#*) continue ;;
+      "") continue ;;
+    esac
+    echo "nameserver $ip"
+  done < "$DNS_CONF"
+} > "$RESOLV_CONF".tmp
+
+# 替换写入 /etc/resolv.conf
+cat "$RESOLV_CONF".tmp > "$RESOLV_CONF"
+rm -f "$RESOLV_CONF".tmp
+
+echo "已更新 $RESOLV_CONF"
+

src/bind/tests/docker-compose.yml

@@ -0,0 +1,16 @@
+services:
+  bind9:
+    image: argus-bind9:latest
+    container_name: argus-bind9-test
+    ports:
+      - "${HOST_DNS_PORT:-1053}:53/tcp"
+      - "${HOST_DNS_PORT:-1053}:53/udp"
+    volumes:
+      - ./private:/private
+    restart: unless-stopped
+    networks:
+      - bind-test-network
+
+networks:
+  bind-test-network:
+    driver: bridge

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

@@ -0,0 +1,118 @@
+#!/bin/bash
+
+# End-to-end test for BIND9 DNS server
+# This script runs all tests in sequence to validate the complete functionality
+# Usage: ./00_e2e_test.sh
+
+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"
+echo "=========================================="
+
+# Track test results
+total_tests=0
+passed_tests=0
+failed_tests=0
+
+# Function to run a test step
+run_test_step() {
+    local step_name="$1"
+    local script_name="$2"
+    local description="$3"
+
+    echo ""
+    echo "[$step_name] $description"
+    echo "$(printf '=%.0s' {1..50})"
+
+    ((total_tests++))
+
+    if [ ! -f "$SCRIPT_DIR/$script_name" ]; then
+        echo "✗ Test script not found: $script_name"
+        ((failed_tests++))
+        return 1
+    fi
+
+    # Make sure script is executable
+    chmod +x "$SCRIPT_DIR/$script_name"
+
+    # Run the test
+    echo "Executing: $SCRIPT_DIR/$script_name"
+    if "$SCRIPT_DIR/$script_name"; then
+        echo "✓ $step_name completed successfully"
+        ((passed_tests++))
+        return 0
+    else
+        echo "✗ $step_name failed"
+        ((failed_tests++))
+        return 1
+    fi
+}
+
+# Cleanup any previous test environment (but preserve the Docker image)
+echo ""
+echo "[SETUP] Cleaning up any previous test environment..."
+if [ -f "$SCRIPT_DIR/05_cleanup.sh" ]; then
+    chmod +x "$SCRIPT_DIR/05_cleanup.sh"
+    "$SCRIPT_DIR/05_cleanup.sh" || true
+fi
+
+echo ""
+echo "Starting BIND9 DNS server end-to-end test sequence..."
+
+# Test sequence
+run_test_step "TEST-01" "01_start_container.sh" "Start BIND9 container" || true
+
+run_test_step "TEST-02" "02_dig_test.sh" "Initial DNS resolution test" || true
+
+run_test_step "TEST-03" "03_reload_test.sh" "Configuration reload with IP modification" || true
+
+run_test_step "TEST-03.5" "03.5_dns_sync_test.sh" "DNS auto-sync functionality test" || true
+
+run_test_step "TEST-04" "04_persistence_test.sh" "Configuration persistence after restart" || true
+
+# Final cleanup (but preserve logs for review)
+echo ""
+echo "[CLEANUP] Cleaning up test environment..."
+run_test_step "CLEANUP" "05_cleanup.sh" "Clean up containers and networks" || true
+
+# Test summary
+echo ""
+echo "=========================================="
+echo "TEST SUMMARY"
+echo "=========================================="
+echo "Total tests: $total_tests"
+echo "Passed: $passed_tests"
+echo "Failed: $failed_tests"
+
+if [ $failed_tests -eq 0 ]; then
+    echo ""
+    echo "✅ ALL TESTS PASSED!"
+    echo ""
+    echo "BIND9 DNS server functionality validated:"
+    echo "  ✓ Container startup and basic functionality"
+    echo "  ✓ DNS resolution for configured domains"
+    echo "  ✓ Configuration modification and reload"
+    echo "  ✓ DNS auto-sync from IP files"
+    echo "  ✓ Configuration persistence across restarts"
+    echo "  ✓ Cleanup and resource management"
+    echo ""
+    echo "The BIND9 DNS server is ready for production use."
+    exit 0
+else
+    echo ""
+    echo "❌ SOME TESTS FAILED!"
+    echo ""
+    echo "Please review the test output above to identify and fix issues."
+    echo "You may need to:"
+    echo "  - Check Docker installation and permissions"
+    echo "  - Verify network connectivity"
+    echo "  - Review BIND9 configuration files"
+    echo "  - Check system resources and port availability"
+    exit 1
+fi

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

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+# Start BIND9 test container
+# Usage: ./01_start_container.sh
+
+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 "$TEST_DIR"
+
+echo "Starting BIND9 test container..."
+
+# Ensure private directory exists with proper permissions
+mkdir -p private/argus/bind
+mkdir -p private/argus/etc
+chmod 777 private
+
+# Start the container
+docker compose up -d
+
+echo "Waiting for container to be ready..."
+sleep 5
+
+# Check if container is running
+if docker compose ps | grep -q "Up"; then
+    echo "✓ Container started successfully"
+    echo "Container status:"
+    docker compose ps
+else
+    echo "✗ Failed to start container"
+    docker compose logs
+    exit 1
+fi
+
+echo ""
+echo "BIND9 test environment is ready!"
+echo "DNS server listening on localhost:${HOST_DNS_PORT}"

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

@@ -0,0 +1,75 @@
+#!/bin/bash
+
+# Test DNS resolution using dig
+# Usage: ./02_dig_test.sh
+
+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() {
+    local hostname="$1"
+    local expected_ip="$2"
+    local description="$3"
+
+    echo ""
+    echo "Testing: $description"
+    echo "Query: $hostname.argus.com"
+    echo "Expected IP: $expected_ip"
+
+    # Perform dig query
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$result" = "QUERY_FAILED" ]; then
+        echo "✗ DNS query failed"
+        return 1
+    elif [ "$result" = "$expected_ip" ]; then
+        echo "✓ DNS query successful: $result"
+        return 0
+    else
+        echo "✗ DNS query returned unexpected result: $result"
+        return 1
+    fi
+}
+
+# Check if dig is available
+if ! command -v dig &> /dev/null; then
+    echo "Installing dig (dnsutils)..."
+    apt-get update && apt-get install -y dnsutils
+fi
+
+# Check if container is running
+if ! docker compose ps | grep -q "Up"; then
+    echo "Error: BIND9 container is not running"
+    echo "Please start the container first with: ./01_start_container.sh"
+    exit 1
+fi
+
+echo "=== DNS Resolution Tests ==="
+
+# Test cases based on current configuration
+failed_tests=0
+
+# Test ns1.argus.com -> 127.0.0.1
+if ! test_dns_query "ns1" "127.0.0.1" "Name server resolution"; then
+    ((failed_tests++))
+fi
+
+# Test web.argus.com -> 12.4.5.6
+if ! test_dns_query "web" "12.4.5.6" "Web server resolution"; then
+    ((failed_tests++))
+fi
+
+echo ""
+echo "=== Test Summary ==="
+if [ $failed_tests -eq 0 ]; then
+    echo "✓ All DNS tests passed!"
+    exit 0
+else
+    echo "✗ $failed_tests test(s) failed"
+    exit 1
+fi

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

@@ -0,0 +1,259 @@
+#!/bin/bash
+
+# Test DNS auto-sync functionality using argus_dns_sync.sh
+# This test validates the automatic DNS record updates from IP files
+# Usage: ./03.5_dns_sync_test.sh
+
+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
+    echo "Error: BIND9 container is not running"
+    echo "Please start the container first with: ./01_start_container.sh"
+    exit 1
+fi
+
+# Check if dig is available
+if ! command -v dig &> /dev/null; then
+    echo "Installing dig (dnsutils)..."
+    apt-get update && apt-get install -y dnsutils
+fi
+
+# Function to test DNS query
+test_dns_query() {
+    local hostname="$1"
+    local expected_ip="$2"
+    local description="$3"
+
+    echo "Testing: $description"
+    echo "Query: $hostname.argus.com -> Expected: $expected_ip"
+
+    # Wait a moment for DNS cache
+    sleep 2
+
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$result" = "$expected_ip" ]; then
+        echo "✓ $result"
+        return 0
+    else
+        echo "✗ Got: $result, Expected: $expected_ip"
+        return 1
+    fi
+}
+
+# Function to wait for sync to complete
+wait_for_sync() {
+    local timeout=15
+    local elapsed=0
+    echo "Waiting for DNS sync to complete (max ${timeout}s)..."
+
+    while [ $elapsed -lt $timeout ]; do
+        if docker compose exec bind9 test -f /var/lock/argus_dns_sync.lock; then
+            echo "Sync process is running..."
+        else
+            echo "Sync completed"
+            sleep 2  # Extra wait for DNS propagation
+            return 0
+        fi
+        sleep 2
+        elapsed=$((elapsed + 2))
+    done
+
+    echo "Warning: Sync may still be running after ${timeout}s"
+    return 0
+}
+
+echo ""
+echo "Step 1: Preparing test environment..."
+
+# Ensure required directories exist
+docker compose exec bind9 mkdir -p /private/argus/etc
+docker compose exec bind9 mkdir -p /private/argus/bind/.backup
+
+# Backup original configuration if it exists
+docker compose exec bind9 test -f /private/argus/bind/db.argus.com && \
+    docker compose exec bind9 cp /private/argus/bind/db.argus.com /private/argus/bind/db.argus.com.backup.test || true
+
+# Ensure initial configuration is available (may already be symlinked)
+docker compose exec bind9 test -f /private/argus/bind/db.argus.com || \
+    docker compose exec bind9 cp /etc/bind/db.argus.com /private/argus/bind/db.argus.com
+
+echo "✓ Test environment prepared"
+
+echo ""
+echo "Step 2: Testing initial DNS configuration..."
+
+# Get current IP for web.argus.com (may have been changed by previous tests)
+current_web_ip=$(dig @localhost -p "$HOST_DNS_PORT" web.argus.com A +short 2>/dev/null || echo "UNKNOWN")
+echo "Current web.argus.com IP: $current_web_ip"
+
+# Test that DNS is working (regardless of specific IP)
+if [ "$current_web_ip" = "UNKNOWN" ] || [ -z "$current_web_ip" ]; then
+    echo "DNS resolution not working for web.argus.com"
+    exit 1
+fi
+
+echo "✓ DNS resolution is working"
+
+echo ""
+echo "Step 3: Creating IP files for auto-sync..."
+
+# Create test IP files in the watch directory
+echo "Creating test1.argus.com with IP 10.0.0.100"
+docker compose exec bind9 bash -c 'echo "10.0.0.100" > /private/argus/etc/test1.argus.com'
+
+echo "Creating test2.argus.com with IP 10.0.0.200"
+docker compose exec bind9 bash -c 'echo "test2 service running on 10.0.0.200" > /private/argus/etc/test2.argus.com'
+
+echo "Creating api.argus.com with IP 192.168.1.50"
+docker compose exec bind9 bash -c 'echo "API server: 192.168.1.50 port 8080" > /private/argus/etc/api.argus.com'
+
+echo "✓ IP files created"
+
+echo ""
+echo "Step 4: Checking DNS sync process..."
+
+# Check if DNS sync process is already running (via supervisord)
+if docker compose exec bind9 pgrep -f argus_dns_sync.sh > /dev/null; then
+    echo "✓ DNS sync process already running (via supervisord)"
+else
+    echo "Starting DNS sync process manually..."
+    # Start the DNS sync process in background if not running
+    docker compose exec -d bind9 /usr/local/bin/argus_dns_sync.sh
+    echo "✓ DNS sync process started manually"
+fi
+
+# Wait for first sync cycle
+wait_for_sync
+
+echo ""
+echo "Step 5: Testing auto-synced DNS records..."
+
+failed_tests=0
+
+# Test new DNS records created by auto-sync
+if ! test_dns_query "test1" "10.0.0.100" "Auto-synced test1.argus.com"; then
+    ((failed_tests++))
+fi
+
+if ! test_dns_query "test2" "10.0.0.200" "Auto-synced test2.argus.com"; then
+    ((failed_tests++))
+fi
+
+if ! test_dns_query "api" "192.168.1.50" "Auto-synced api.argus.com"; then
+    ((failed_tests++))
+fi
+
+# Verify original records still work (use current IP from earlier)
+if ! test_dns_query "web" "$current_web_ip" "Original web.argus.com still working"; then
+    ((failed_tests++))
+fi
+
+if ! test_dns_query "ns1" "127.0.0.1" "Original ns1.argus.com still working"; then
+    ((failed_tests++))
+fi
+
+echo ""
+echo "Step 6: Testing IP update functionality..."
+
+# Update an existing IP file
+echo "Updating test1.argus.com IP from 10.0.0.100 to 10.0.0.150"
+docker compose exec bind9 bash -c 'echo "10.0.0.150" > /private/argus/etc/test1.argus.com'
+
+# Wait for sync
+wait_for_sync
+
+# Test updated record
+if ! test_dns_query "test1" "10.0.0.150" "Updated test1.argus.com IP"; then
+    ((failed_tests++))
+fi
+
+echo ""
+echo "Step 7: Testing invalid IP handling..."
+
+# Create file with invalid IP
+echo "Creating invalid.argus.com with invalid IP"
+docker compose exec bind9 bash -c 'echo "this is not an IP address" > /private/argus/etc/invalid.argus.com'
+
+# Wait for sync (should skip invalid IP)
+wait_for_sync
+
+# Verify invalid record was not added (should fail to resolve)
+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
+    echo "✗ Invalid IP was processed: $result"
+    ((failed_tests++))
+fi
+
+echo ""
+echo "Step 8: Verifying backup functionality..."
+
+# Check if backups were created
+backup_count=$(docker compose exec bind9 ls -1 /private/argus/bind/.backup/ | wc -l || echo "0")
+if [ "$backup_count" -gt 0 ]; then
+    echo "✓ Configuration backups created ($backup_count files)"
+    # Show latest backup
+    docker compose exec bind9 ls -la /private/argus/bind/.backup/ | tail -1
+else
+    echo "✗ No backup files found"
+    ((failed_tests++))
+fi
+
+echo ""
+echo "Step 9: Cleanup..."
+
+# Note: We don't stop the DNS sync process since it's managed by supervisord
+echo "Note: DNS sync process will continue running (managed by supervisord)"
+
+# Clean up test files
+docker compose exec bind9 rm -f /private/argus/etc/test1.argus.com
+docker compose exec bind9 rm -f /private/argus/etc/test2.argus.com
+docker compose exec bind9 rm -f /private/argus/etc/api.argus.com
+docker compose exec bind9 rm -f /private/argus/etc/invalid.argus.com
+
+# Restore original configuration if backup exists
+docker compose exec bind9 test -f /private/argus/bind/db.argus.com.backup.test && \
+    docker compose exec bind9 cp /private/argus/bind/db.argus.com.backup.test /private/argus/bind/db.argus.com && \
+    docker compose exec bind9 rm /private/argus/bind/db.argus.com.backup.test || true
+
+# Reload original configuration
+docker compose exec bind9 /usr/local/bin/reload-bind9.sh
+
+echo "✓ Cleanup completed"
+
+echo ""
+echo "=== DNS Auto-Sync Test Summary ==="
+if [ $failed_tests -eq 0 ]; then
+    echo "✅ All DNS auto-sync tests passed!"
+    echo ""
+    echo "Validated functionality:"
+    echo "  ✓ Automatic DNS record creation from IP files"
+    echo "  ✓ IP address extraction from various file formats"
+    echo "  ✓ Dynamic DNS record updates"
+    echo "  ✓ Invalid IP address handling"
+    echo "  ✓ Configuration backup mechanism"
+    echo "  ✓ Preservation of existing DNS records"
+    echo ""
+    echo "The DNS auto-sync functionality is working correctly!"
+    exit 0
+else
+    echo "❌ $failed_tests DNS auto-sync test(s) failed!"
+    echo ""
+    echo "Please check:"
+    echo "  - argus_dns_sync.sh script configuration"
+    echo "  - File permissions in /private/argus/etc/"
+    echo "  - BIND9 reload functionality"
+    echo "  - Network connectivity and DNS resolution"
+    exit 1
+fi

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

@@ -0,0 +1,115 @@
+#!/bin/bash
+
+# Test DNS configuration reload with IP modification
+# Usage: ./03_reload_test.sh
+
+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
+    echo "Error: BIND9 container is not running"
+    echo "Please start the container first with: ./01_start_container.sh"
+    exit 1
+fi
+
+# Check if dig is available
+if ! command -v dig &> /dev/null; then
+    echo "Installing dig (dnsutils)..."
+    apt-get update && apt-get install -y dnsutils
+fi
+
+# Function to test DNS query
+test_dns_query() {
+    local hostname="$1"
+    local expected_ip="$2"
+    local description="$3"
+
+    echo "Testing: $description"
+    echo "Query: $hostname.argus.com -> Expected: $expected_ip"
+
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$result" = "$expected_ip" ]; then
+        echo "✓ $result"
+        return 0
+    else
+        echo "✗ Got: $result, Expected: $expected_ip"
+        return 1
+    fi
+}
+
+echo ""
+echo "Step 1: Testing initial DNS configuration..."
+
+# Test initial configuration
+if ! test_dns_query "web" "12.4.5.6" "Initial web.argus.com resolution"; then
+    echo "Initial DNS test failed"
+    exit 1
+fi
+
+echo ""
+echo "Step 2: Modifying DNS configuration..."
+
+# Backup original configuration
+cp "$TEST_DIR/private/argus/bind/db.argus.com" "$TEST_DIR/private/argus/bind/db.argus.com.backup" 2>/dev/null || true
+
+# Create new configuration with modified IP
+DB_FILE="$TEST_DIR/private/argus/bind/db.argus.com"
+
+# Check if persistent config exists, if not use from container
+if [ ! -f "$DB_FILE" ]; then
+    echo "Persistent config not found, copying from container..."
+    docker compose exec bind9 cp /etc/bind/db.argus.com /private/argus/bind/db.argus.com
+    docker compose exec bind9 chown bind:bind /private/argus/bind/db.argus.com
+fi
+
+# Modify the IP address (12.4.5.6 -> 192.168.1.100)
+sed -i 's/12\.4\.5\.6/192.168.1.100/g' "$DB_FILE"
+
+# Increment serial number for DNS cache invalidation
+current_serial=$(grep -o "2[[:space:]]*;" "$DB_FILE" | grep -o "2")
+new_serial=$((current_serial + 1))
+sed -i "s/2[[:space:]]*;/${new_serial}         ;/" "$DB_FILE"
+
+echo "Modified configuration:"
+echo "- Changed web.argus.com IP: 12.4.5.6 -> 192.168.1.100"
+echo "- Updated serial number: $current_serial -> $new_serial"
+
+echo ""
+echo "Step 3: Reloading BIND9 configuration..."
+
+# Reload BIND9 configuration
+docker compose exec bind9 /usr/local/bin/reload-bind9.sh
+
+echo "Configuration reloaded"
+
+# Wait a moment for changes to take effect
+sleep 3
+
+echo ""
+echo "Step 4: Testing modified DNS configuration..."
+
+# Test modified configuration
+if ! test_dns_query "web" "192.168.1.100" "Modified web.argus.com resolution"; then
+    echo "Modified DNS test failed"
+    exit 1
+fi
+
+# Also verify ns1 still works
+if ! test_dns_query "ns1" "127.0.0.1" "ns1.argus.com still working"; then
+    echo "ns1 DNS test failed after reload"
+    exit 1
+fi
+
+echo ""
+echo "✓ DNS configuration reload test completed successfully!"
+echo "✓ IP address changed from 12.4.5.6 to 192.168.1.100"
+echo "✓ Configuration persisted and reloaded correctly"

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

@@ -0,0 +1,118 @@
+#!/bin/bash
+
+# Test configuration persistence after container restart
+# Usage: ./04_persistence_test.sh
+
+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
+    echo "Installing dig (dnsutils)..."
+    apt-get update && apt-get install -y dnsutils
+fi
+
+# Function to test DNS query
+test_dns_query() {
+    local hostname="$1"
+    local expected_ip="$2"
+    local description="$3"
+
+    echo "Testing: $description"
+    echo "Query: $hostname.argus.com -> Expected: $expected_ip"
+
+    result=$(dig @localhost -p "$HOST_DNS_PORT" "$hostname".argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$result" = "$expected_ip" ]; then
+        echo "✓ $result"
+        return 0
+    else
+        echo "✗ Got: $result, Expected: $expected_ip"
+        return 1
+    fi
+}
+
+echo ""
+echo "Step 1: Stopping current container..."
+
+# Stop the container
+docker compose down
+
+echo "Container stopped"
+
+echo ""
+echo "Step 2: Verifying persistent configuration exists..."
+
+# Check if modified configuration exists
+DB_FILE="$TEST_DIR/private/argus/bind/db.argus.com"
+
+if [ ! -f "$DB_FILE" ]; then
+    echo "✗ Persistent configuration file not found: $DB_FILE"
+    exit 1
+fi
+
+# Check if the modified IP is in the configuration
+if grep -q "192.168.1.100" "$DB_FILE"; then
+    echo "✓ Modified IP (192.168.1.100) found in persistent configuration"
+else
+    echo "✗ Modified IP not found in persistent configuration"
+    echo "Configuration content:"
+    cat "$DB_FILE"
+    exit 1
+fi
+
+echo ""
+echo "Step 3: Restarting container with persistent configuration..."
+
+# Start the container again
+docker compose up -d
+
+echo "Waiting for container to be ready..."
+sleep 5
+
+# Check if container is running
+if ! docker compose ps | grep -q "Up"; then
+    echo "✗ Failed to restart container"
+    docker compose logs
+    exit 1
+fi
+
+echo "✓ Container restarted successfully"
+
+echo ""
+echo "Step 4: Testing DNS resolution after restart..."
+
+# Wait a bit more for DNS to be fully ready
+sleep 5
+
+# Test that the modified configuration is still active
+if ! test_dns_query "web" "192.168.1.100" "Persistent web.argus.com resolution"; then
+    echo "✗ Persistent configuration test failed"
+    exit 1
+fi
+
+# Also verify ns1 still works
+if ! test_dns_query "ns1" "127.0.0.1" "ns1.argus.com still working"; then
+    echo "✗ ns1 DNS test failed after restart"
+    exit 1
+fi
+
+echo ""
+echo "Step 5: Verifying configuration files are linked correctly..."
+
+# Check that the persistent files are properly linked
+echo "Checking file links in container:"
+docker compose exec bind9 ls -la /etc/bind/named.conf.local /etc/bind/db.argus.com
+
+echo ""
+echo "✓ Configuration persistence test completed successfully!"
+echo "✓ Modified IP (192.168.1.100) persisted after container restart"
+echo "✓ Configuration files properly linked to persistent storage"
+echo "✓ DNS resolution working correctly with persisted configuration"

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

@@ -0,0 +1,90 @@
+#!/bin/bash
+
+# Clean up test environment and containers
+# Usage: ./05_cleanup.sh [--full]
+
+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
+while [[ $# -gt 0 ]]; do
+    case $1 in
+        --full)
+            FULL_CLEANUP=true
+            shift
+            ;;
+        *)
+            echo "Unknown option: $1"
+            echo "Usage: $0 [--full]"
+            echo "  --full: Also remove persistent data "
+            exit 1
+            ;;
+    esac
+done
+
+cd "$TEST_DIR"
+
+echo "=== Cleaning up BIND9 test environment ==="
+
+echo ""
+echo "Step 1: Stopping and removing containers..."
+
+# Stop and remove containers
+docker compose down -v
+
+echo "✓ Containers stopped and removed"
+
+echo ""
+echo "Step 2: Removing Docker networks..."
+
+# Clean up networks
+docker network prune -f > /dev/null 2>&1 || true
+
+echo "✓ Docker networks cleaned"
+
+if [ "$FULL_CLEANUP" = true ]; then
+    echo ""
+    echo "Step 3: Removing persistent data..."
+
+    # Remove persistent data directory
+    if [ -d "private" ]; then
+        rm -rf private
+        echo "✓ Persistent data directory removed"
+    else
+        echo "✓ No persistent data directory found"
+    fi
+
+else
+    echo ""
+    echo "Step 3: Preserving persistent data and Docker image..."
+    echo "✓ Persistent data preserved in: private/"
+    echo "✓ Docker image 'argus-bind9:latest' preserved"
+    echo ""
+    echo "To perform full cleanup including persistent data and image, run:"
+    echo "  $0 --full"
+fi
+
+echo ""
+echo "=== Cleanup Summary ==="
+echo "✓ Containers stopped and removed"
+echo "✓ Docker networks cleaned"
+
+if [ "$FULL_CLEANUP" = true ]; then
+    echo "✓ Persistent data removed"
+    echo ""
+    echo "Full cleanup completed! Test environment completely removed."
+else
+    echo "✓ Persistent data preserved"
+    echo "✓ Docker image preserved"
+    echo ""
+    echo "Basic cleanup completed! Run './01_start_container.sh' to restart testing."
+fi
+
+echo ""
+echo "Test environment cleanup finished."

src/log/.gitignore

@@ -0,0 +1,5 @@
+
+private/
+
+
+images/

src/log/README.md

@@ -0,0 +1,8 @@
+
+测试log模块开发
+
+elasticsearch: 部署镜像构建及启动脚本（解决账号问题、挂载目录、使用supervisor守护）
+kibana: 镜像构建
+fluent-bit: 安装包，脚本准备， 交付给大鹏统一组织客户端侧安装流程
+init: EK初始化脚本：数据视图创建脚本等
+

src/log/elasticsearch/build/Dockerfile

@@ -0,0 +1,75 @@
+FROM docker.elastic.co/elasticsearch/elasticsearch:8.13.4
+
+# 切换到 root 用户进行系统级安装
+USER root
+
+ARG ARGUS_BUILD_UID=2133
+ARG ARGUS_BUILD_GID=2015
+
+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
+
+# 配置内网 apt 源 (如果指定了内网选项)
+RUN if [ "$USE_INTRANET" = "true" ]; then \
+        echo "Configuring intranet apt sources..." && \
+        cp /etc/apt/sources.list /etc/apt/sources.list.bak && \
+        echo "deb [trusted=yes] http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
+        echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
+    fi
+
+# 安装 supervisor, net-tools, vim
+RUN apt-get update && \
+    apt-get install -y supervisor net-tools inetutils-ping vim && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# 配置部署时使用的apt源
+RUN if [ "$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/log/elasticsearch/build/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+
+# 复制启动脚本
+COPY src/log/elasticsearch/build/start-es-supervised.sh /usr/local/bin/start-es-supervised.sh
+RUN chmod +x /usr/local/bin/start-es-supervised.sh
+
+# 复制DNS监控脚本
+COPY src/log/elasticsearch/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 管理用户切换
+USER root
+
+# 暴露端口
+EXPOSE 9200 9300
+
+# 使用 supervisor 作为入口点
+CMD ["/usr/bin/supervisord", "-c", "/etc/supervisor/conf.d/supervisord.conf"]

src/log/elasticsearch/build/dns-monitor.sh

@@ -0,0 +1 @@
+../../../bind/build/dns-monitor.sh

src/log/elasticsearch/build/start-es-supervised.sh

@@ -0,0 +1,32 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Elasticsearch under supervisor..."
+
+# 创建数据目录并设置权限（如果不存在）
+mkdir -p /private/argus/log/elasticsearch
+
+# 创建软链接到Elasticsearch预期的数据目录
+if [ -L /usr/share/elasticsearch/data ]; then
+    rm /usr/share/elasticsearch/data
+elif [ -d /usr/share/elasticsearch/data ]; then
+    rm -rf /usr/share/elasticsearch/data
+fi
+
+ln -sf /private/argus/log/elasticsearch /usr/share/elasticsearch/data
+
+# 记录容器ip地址
+DOMAIN=es.log.argus.com
+IP=`ifconfig | grep -A 1 eth0 | grep inet | awk '{print $2}'`
+echo current IP: ${IP}
+echo ${IP} > /private/argus/etc/${DOMAIN}
+
+echo "[INFO] Data directory linked: /usr/share/elasticsearch/data -> /private/argus/log/elasticsearch"
+
+# 设置环境变量（ES配置通过docker-compose传递）
+export ES_JAVA_OPTS="${ES_JAVA_OPTS:-"-Xms512m -Xmx512m"}"
+
+echo "[INFO] Starting Elasticsearch process..."
+
+# 启动原始的Elasticsearch entrypoint
+exec /usr/local/bin/docker-entrypoint.sh elasticsearch

src/log/elasticsearch/build/supervisord.conf

@@ -0,0 +1,39 @@
+[supervisord]
+nodaemon=true
+logfile=/var/log/supervisor/supervisord.log
+pidfile=/var/run/supervisord.pid
+user=root
+
+[program:elasticsearch]
+command=/usr/local/bin/start-es-supervised.sh
+user=elasticsearch
+stdout_logfile=/var/log/supervisor/elasticsearch.log
+stderr_logfile=/var/log/supervisor/elasticsearch_error.log
+autorestart=true
+startretries=3
+startsecs=30
+stopwaitsecs=30
+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/log/fluent-bit/build/etc/fluent-bit.conf

@@ -0,0 +1,37 @@
+[SERVICE]
+    Daemon       Off
+    Parsers_File parsers.conf
+    HTTP_Server  On
+    HTTP_Listen  0.0.0.0
+    HTTP_Port    2020
+    storage.path /buffers
+    storage.sync normal
+    storage.checksum on
+    storage.backlog.mem_limit 128M
+    # 备注：该镜像默认未开启 Hot Reload，修改配置后请重启容器。
+
+@INCLUDE inputs.d/*.conf
+
+[FILTER]
+    Name   parser
+    Match  app.*
+    Key_Name log
+    Parser timestamp_parser
+    Reserve_Data On
+    Preserve_Key On
+    Unescape_Key On
+
+[FILTER]
+    Name   record_modifier
+    Match  *
+    Record cluster  ${CLUSTER}
+    Record rack     ${RACK}
+    Record host     ${HOSTNAME}
+
+[FILTER]
+    Name   lua
+    Match  app.*
+    script inject_labels.lua
+    call   add_labels
+
+@INCLUDE outputs.d/*.conf

src/log/fluent-bit/build/etc/inject_labels.lua

@@ -0,0 +1,15 @@
+function add_labels(tag, ts, record)
+  record["job_id"] = os.getenv("FB_JOB_ID") or record["job_id"] or "unknown"
+  record["user"]   = os.getenv("FB_USER")   or record["user"]   or "unknown"
+  record["model"]  = os.getenv("FB_MODEL")  or record["model"]  or "unknown"
+  record["gpu_id"] = os.getenv("FB_GPU_ID") or record["gpu_id"] or "na"
+  local p = record["log_path"] or ""
+  if string.find(p, "/logs/infer/") then
+    record["role"] = "infer"
+  elseif string.find(p, "/logs/train/") then
+    record["role"] = "train"
+  else
+    record["role"] = record["role"] or "app"
+  end
+  return 1, ts, record
+end

src/log/fluent-bit/build/etc/inputs.d/10-train.conf

@@ -0,0 +1,10 @@
+[INPUT]
+    Name              tail
+    Path              /logs/train/*.log
+    Tag               app.train
+    Path_Key          log_path
+    Refresh_Interval  5
+    DB                /buffers/train.db
+    Skip_Long_Lines   On
+    storage.type      filesystem
+    multiline.parser  python,go,java

src/log/fluent-bit/build/etc/inputs.d/20-infer.conf

@@ -0,0 +1,10 @@
+[INPUT]
+    Name              tail
+    Path              /logs/infer/*.log
+    Tag               app.infer
+    Path_Key          log_path
+    Refresh_Interval  5
+    DB                /buffers/infer.db
+    Skip_Long_Lines   On
+    storage.type      filesystem
+    multiline.parser  python,go,java

src/log/fluent-bit/build/etc/outputs.d/10-es.conf

@@ -0,0 +1,24 @@
+# 重要：使用 Logstash_Format + Logstash_Prefix，生成 train-*/infer-* 索引
+[OUTPUT]
+    Name                es
+    Match               app.train
+    Host                ${ES_HOST}
+    Port                ${ES_PORT}
+    Logstash_Format     On
+    Logstash_Prefix     train
+    Replace_Dots        On
+    Generate_ID         On
+    Retry_Limit         False
+    Suppress_Type_Name  On
+
+[OUTPUT]
+    Name                es
+    Match               app.infer
+    Host                ${ES_HOST}
+    Port                ${ES_PORT}
+    Logstash_Format     On
+    Logstash_Prefix     infer
+    Replace_Dots        On
+    Generate_ID         On
+    Retry_Limit         False
+    Suppress_Type_Name  On

src/log/fluent-bit/build/etc/parsers.conf

@@ -0,0 +1,29 @@
+[MULTILINE_PARSER]
+    Name   python
+    Type   regex
+    Flush  2
+    Rule   "start_state"  "/^\d{4}-\d{2}-\d{2}[\sT]/"  "cont"
+    Rule   "cont"         "/^\s+|^Traceback|^\tat\s+/" "cont"
+
+[MULTILINE_PARSER]
+    Name   go
+    Type   regex
+    Flush  2
+    Rule   "start_state"  "/^[0-9]{4}\/[0-9]{2}\/[0-9]{2}/" "cont"
+    Rule   "cont"         "/^\s+|^\t/" "cont"
+
+[MULTILINE_PARSER]
+    Name   java
+    Type   regex
+    Flush  2
+    Rule   "start_state"  "/^\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}/" "cont"
+    Rule   "cont"         "/^\s+at\s+|^\t.../" "cont"
+
+[PARSER]
+    Name   timestamp_parser
+    Format regex
+    Regex  ^(?<timestamp>\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2})\s+(?<level>\w+)\s+(?<message>.*)$
+    Time_Key    timestamp
+    Time_Format %Y-%m-%d %H:%M:%S
+    Time_Offset +0800
+    Time_Keep On

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

@@ -0,0 +1,96 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Fluent Bit setup in Ubuntu container (offline-first)..."
+
+export DEBIAN_FRONTEND=noninteractive
+
+# Stage bundle to /tmp (read-only mount under /private)
+echo "[INFO] Staging fluent-bit bundle..."
+rm -rf /tmp/flb && mkdir -p /tmp/flb
+cp -r /private/etc /tmp/flb/
+mkdir -p /tmp/flb/packages
+cp -r /private/packages/* /tmp/flb/packages/ 2>/dev/null || true
+
+# Helper: check and install a local deb if not already satisfied
+ensure_lib() {
+  local soname="$1"; shift
+  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
+
+echo "[INFO] Fluent Bit version:"
+/opt/fluent-bit/bin/fluent-bit --version || { echo "[ERROR] fluent-bit not installed or libraries missing" >&2; exit 1; }
+
+# Place configuration
+mkdir -p /etc/fluent-bit
+cp -r /tmp/flb/etc/* /etc/fluent-bit/
+
+# Create logs/buffers dirs
+mkdir -p /logs/train /logs/infer /buffers
+chmod 755 /logs/train /logs/infer /buffers
+
+# Wait for Elasticsearch via bash /dev/tcp to avoid curl dependency
+echo "[INFO] Waiting for Elasticsearch to be ready (tcp ${ES_HOST}:${ES_PORT})..."
+for i in $(seq 1 120); do
+  if exec 3<>/dev/tcp/${ES_HOST}/${ES_PORT}; then
+    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] 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 --config=/etc/fluent-bit/fluent-bit.conf

src/log/kibana/build/Dockerfile

@@ -0,0 +1,79 @@
+FROM docker.elastic.co/kibana/kibana:8.13.4
+
+# 切换到 root 用户进行系统级安装
+USER root
+
+ARG ARGUS_BUILD_UID=2133
+ARG ARGUS_BUILD_GID=2015
+
+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
+
+# 配置内网 apt 源 (如果指定了内网选项)
+RUN if [ "$USE_INTRANET" = "true" ]; then \
+        echo "Configuring intranet apt sources..." && \
+        cp /etc/apt/sources.list /etc/apt/sources.list.bak && \
+        echo "deb [trusted=yes]  http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
+        echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
+    fi
+
+# 安装 supervisor, net-tools, vim
+RUN apt-get update && \
+    apt-get install -y supervisor net-tools inetutils-ping vim && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# 配置部署时使用的apt源
+RUN if [ "$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/log/kibana/build/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+
+# 复制启动脚本
+COPY src/log/kibana/build/start-kibana-supervised.sh /usr/local/bin/start-kibana-supervised.sh
+COPY src/log/kibana/build/kibana-post-start.sh /usr/local/bin/kibana-post-start.sh
+RUN chmod +x /usr/local/bin/start-kibana-supervised.sh /usr/local/bin/kibana-post-start.sh
+
+# 复制DNS监控脚本
+COPY src/log/kibana/build/dns-monitor.sh /usr/local/bin/dns-monitor.sh
+RUN chmod +x /usr/local/bin/dns-monitor.sh
+
+# kibana需要用到 /root/.config/puppeteer 路径
+RUN chmod 777 /root
+
+# 保持 root 用户，由 supervisor 管理用户切换
+USER root
+
+# 暴露端口
+EXPOSE 5601
+
+# 使用 supervisor 作为入口点
+CMD ["/usr/bin/supervisord", "-c", "/etc/supervisor/conf.d/supervisord.conf"]

src/log/kibana/build/dns-monitor.sh

@@ -0,0 +1 @@
+../../../bind/build/dns-monitor.sh

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

@@ -0,0 +1,133 @@
+#!/bin/bash
+set -euo pipefail
+
+ES_HOST="${ELASTICSEARCH_HOSTS:-http://es:9200}"
+KB_HOST="${KB_HOST:-http://127.0.0.1:5601}"
+
+echo "[INFO] Starting Kibana post-start configuration..."
+
+# 等待 Elasticsearch 可用
+wait_for_elasticsearch() {
+    echo "[INFO] Waiting for Elasticsearch..."
+    local max_attempts=60
+    local attempt=1
+
+    while [ $attempt -le $max_attempts ]; do
+        if curl -fs "$ES_HOST/_cluster/health" >/dev/null 2>&1; then
+            echo "[OK] Elasticsearch is available"
+            return 0
+        fi
+        echo "    Waiting for ES... ($attempt/$max_attempts)"
+        sleep 5
+        ((attempt++))
+    done
+
+    echo "[ERROR] Elasticsearch timeout"
+    return 1
+}
+
+# 等待 Kibana 可用
+wait_for_kibana() {
+    echo "[INFO] Waiting for Kibana..."
+    local max_attempts=120
+    local attempt=1
+
+    while [ $attempt -le $max_attempts ]; do
+        if curl -fs "$KB_HOST/api/status" >/dev/null 2>&1; then
+            local status=$(curl -s "$KB_HOST/api/status" | grep -o '"level":"available"' || echo "")
+            if [ -n "$status" ]; then
+                echo "[OK] Kibana is available"
+                return 0
+            fi
+            echo "    Waiting for Kibana... ($attempt/$max_attempts, status: $status)"
+        else
+            echo "    Waiting for Kibana... ($attempt/$max_attempts, connection failed)"
+        fi
+        sleep 5
+        ((attempt++))
+    done
+
+    echo "[ERROR] Kibana timeout"
+    return 1
+}
+
+# 幂等设置索引副本数为0
+fix_replicas_idempotent() {
+    echo "[INFO] Checking and fixing index replicas..."
+
+    # 获取所有 train-* 和 infer-* 索引
+    local indices=$(curl -s "$ES_HOST/_cat/indices/train-*,infer-*?h=index" 2>/dev/null || echo "")
+
+    if [ -z "$indices" ]; then
+        echo "[INFO] No train-*/infer-* indices found, skipping replica adjustment"
+        return 0
+    fi
+
+    for idx in $indices; do
+        # 检查当前副本数
+        local current_replicas=$(curl -s "$ES_HOST/$idx/_settings" | grep -o '"number_of_replicas":"[^"]*"' | cut -d'"' -f4 || echo "")
+
+        if [ "$current_replicas" != "0" ]; then
+            echo "[INFO] Setting replicas to 0 for index: $idx (current: $current_replicas)"
+            curl -fsS -X PUT "$ES_HOST/$idx/_settings" \
+                -H 'Content-Type: application/json' \
+                -d '{"index":{"number_of_replicas":0}}' >/dev/null || {
+                echo "[WARN] Failed to set replicas for $idx"
+                continue
+            }
+            echo "[OK] Updated replicas for $idx"
+        else
+            echo "[INFO] Index $idx already has 0 replicas, skipping"
+        fi
+    done
+}
+
+# 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..."
+
+    create_or_ensure_data_view "train" "train-*"
+    create_or_ensure_data_view "infer" "infer-*"
+}
+
+# 主逻辑
+main() {
+    # 等待服务可用
+    wait_for_elasticsearch || exit 1
+    wait_for_kibana || exit 1
+
+    # 执行幂等配置
+    fix_replicas_idempotent
+    create_data_views_idempotent
+
+    echo "[INFO] Kibana post-start configuration completed"
+}
+
+# 运行主逻辑
+main

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

@@ -0,0 +1,37 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Kibana under supervisor..."
+
+mkdir -p /private/argus/log/kibana
+
+# 创建软链接到Kibana预期的数据目录
+if [ -L /usr/share/kibana/data ]; then
+    rm /usr/share/kibana/data
+elif [ -d /usr/share/kibana/data ]; then
+    rm -rf /usr/share/kibana/data
+fi
+
+ln -sf /private/argus/log/kibana /usr/share/kibana/data
+
+echo "[INFO] Data directory linked: /usr/share/kibana/data -> /private/argus/log/kibana"
+
+# 记录容器ip地址
+DOMAIN=kibana.log.argus.com
+IP=`ifconfig | grep -A 1 eth0 | grep inet | awk '{print $2}'`
+echo current IP: ${IP}
+echo ${IP} > /private/argus/etc/${DOMAIN}
+
+# 设置环境变量
+export ELASTICSEARCH_HOSTS="${ELASTICSEARCH_HOSTS:-"http://es:9200"}"
+
+echo "[INFO] Connecting to Elasticsearch at: $ELASTICSEARCH_HOSTS"
+
+# 启动后台配置任务
+echo "[INFO] Starting background post-start configuration..."
+/usr/local/bin/kibana-post-start.sh &
+
+echo "[INFO] Starting Kibana process..."
+
+# 启动原始的Kibana entrypoint
+exec /usr/local/bin/kibana-docker

src/log/kibana/build/supervisord.conf

@@ -0,0 +1,39 @@
+[supervisord]
+nodaemon=true
+logfile=/var/log/supervisor/supervisord.log
+pidfile=/var/run/supervisord.pid
+user=root
+
+[program:kibana]
+command=/usr/local/bin/start-kibana-supervised.sh
+user=kibana
+stdout_logfile=/var/log/supervisor/kibana.log
+stderr_logfile=/var/log/supervisor/kibana_error.log
+autorestart=true
+startretries=3
+startsecs=30
+stopwaitsecs=30
+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