[#2] master模块离线构建方式

build/build_images.sh

@@ -0,0 +1,138 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 帮助信息
+show_help() {
+    cat << EOF
+ARGUS Unified Build System - Image Build Tool
+
+Usage: $0 [OPTIONS]
+
+Options:
+  --intranet    Use intranet mirror for Ubuntu 22.04 packages
+  -h, --help    Show this help message
+
+Examples:
+  $0                # Build with default sources
+  $0 --intranet     # Build with intranet mirror
+
+EOF
+}
+
+# 解析命令行参数
+use_intranet=false
+
+while [[ $# -gt 0 ]]; do
+    case $1 in
+        --intranet)
+            use_intranet=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 "$root"
+
+echo "======================================="
+echo "ARGUS Unified Build System"
+echo "======================================="
+
+if [[ "$use_intranet" == true ]]; then
+    echo "🌐 Mode: Intranet (Using internal mirror: 10.68.64.1)"
+    build_args="--build-arg USE_INTRANET=true"
+else
+    echo "🌐 Mode: Public (Using default package sources)"
+    build_args=""
+fi
+
+echo "📁 Build context: $root"
+echo ""
+
+# 构建镜像的函数
+build_image() {
+    local image_name=$1
+    local dockerfile_path=$2
+    local tag=$3
+
+    echo "🔄 Building $image_name image..."
+    echo "   Dockerfile: $dockerfile_path"
+    echo "   Tag: $tag"
+
+    if docker build $build_args -f "$dockerfile_path" -t "$tag" .; then
+        echo "✅ $image_name image built successfully"
+        return 0
+    else
+        echo "❌ Failed to build $image_name image"
+        return 1
+    fi
+}
+
+# 构建所有镜像
+images_built=()
+build_failed=false
+
+# 构建 Elasticsearch 镜像
+if build_image "Elasticsearch" "src/log/elasticsearch/build/Dockerfile" "argus-elasticsearch:latest"; then
+    images_built+=("argus-elasticsearch:latest")
+else
+    build_failed=true
+fi
+
+echo ""
+
+# 构建 Kibana 镜像
+if build_image "Kibana" "src/log/kibana/build/Dockerfile" "argus-kibana:latest"; then
+    images_built+=("argus-kibana:latest")
+else
+    build_failed=true
+fi
+
+echo ""
+
+# 构建 BIND9 镜像
+if build_image "BIND9" "src/bind/build/Dockerfile" "argus-bind9:latest"; then
+    images_built+=("argus-bind9:latest")
+else
+    build_failed=true
+fi
+
+echo ""
+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
+
+echo ""
+echo "🚀 Next steps:"
+echo "   cd src/log && ./scripts/save_images.sh     # Export log images"
+echo "   cd src/bind && ./scripts/save_images.sh    # Export bind images"
+echo "   cd src/log/tests && ./scripts/02_up.sh     # Start log services"
+echo ""

build/save_images.sh

@@ -0,0 +1,222 @@
+#!/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"
+)
+
+# 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 ""

src/.gitignore

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

src/agent/.gitignore

@@ -0,0 +1,3 @@
+build/
+*.egg-info/
+__pycache__/

src/agent/README.md

@@ -0,0 +1,53 @@
+# 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
+```
+
+脚本会在本地创建虚拟环境、通过 `pyproject.toml` 安装依赖，并调用 PyInstaller 产出单文件可执行程序。构建过程中生成的 `build/`、`dist/` 会被 `.gitignore` 忽略。
+
+## 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。 |
+
+派生路径：
+
+- 节点信息：`/private/argus/agent/<hostname>/node.json`
+- 子模块健康目录：`/private/argus/agent/health/<hostname>/`
+
+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 +1,111 @@
+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
+    env, user, instance = _parse_hostname(hostname)
+    meta = {
+        "hostname": hostname,
+        "ip": _detect_ip_address(),
+        "env": env,
+        "user": user,
+        "instance": 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 +1,74 @@
+from __future__ import annotations
+
+import os
+import socket
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Final
+
+from .version import VERSION
+
+DEFAULT_REPORT_INTERVAL_SECONDS: Final[int] = 60
+
+
+@dataclass(frozen=True)
+class AgentConfig:
+    hostname: 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_config() -> AgentConfig:
+    """从环境变量推导配置，移除了外部配置文件依赖。"""
+
+    hostname = _resolve_hostname()
+    node_file = f"/private/argus/agent/{hostname}/node.json"
+    health_dir = f"/private/argus/agent/health/{hostname}/"
+
+    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,
+        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/build_binary.sh

@@ -0,0 +1,42 @@
+#!/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"
+VENV_DIR="$BUILD_ROOT/venv"
+
+mkdir -p "$PYINSTALLER_BUILD"
+mkdir -p "$DIST_DIR"
+
+if [[ ! -d "$VENV_DIR" ]]; then
+  python3 -m venv "$VENV_DIR"
+fi
+
+# shellcheck disable=SC1091
+source "$VENV_DIR/bin/activate"
+
+pip install --upgrade pip
+pip install "$MODULE_ROOT"
+pip install "pyinstaller==6.6.0"
+
+rm -rf "$PYINSTALLER_BUILD"/*
+rm -f "$DIST_DIR/argus-agent"
+
+pyinstaller \
+  --clean \
+  --onefile \
+  --name argus-agent \
+  --distpath "$DIST_DIR" \
+  --workpath "$PYINSTALLER_BUILD/work" \
+  --specpath "$PYINSTALLER_BUILD/spec" \
+  --add-data "$MODULE_ROOT/pyproject.toml:." \
+  "$MODULE_ROOT/entry.py"
+
+chmod +x "$DIST_DIR/argus-agent"
+
+deactivate
+
+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 +1,37 @@
+services:
+  master:
+    image: argus-master:dev
+    container_name: argus-master-agent-e2e
+    environment:
+      - OFFLINE_THRESHOLD_SECONDS=6
+      - ONLINE_THRESHOLD_SECONDS=2
+      - SCHEDULER_INTERVAL_SECONDS=1
+    ports:
+      - "32300:3000"
+    volumes:
+      - ./private/argus/master:/private/argus/master
+      - ./private/argus/metric/prometheus:/private/argus/metric/prometheus
+
+  agent:
+    image: ubuntu:24.04
+    container_name: argus-agent-e2e
+    hostname: dev-e2euser-e2einst-pod-0
+    depends_on:
+      - master
+    environment:
+      - MASTER_ENDPOINT=http://master:3000
+      - REPORT_INTERVAL_SECONDS=2
+    volumes:
+      - ./private/argus/agent/dev-e2euser-e2einst-pod-0:/private/argus/agent/dev-e2euser-e2einst-pod-0
+      - ./private/argus/agent/health/dev-e2euser-e2einst-pod-0:/private/argus/agent/health/dev-e2euser-e2einst-pod-0
+      - ../dist/argus-agent:/usr/local/bin/argus-agent:ro
+    entrypoint:
+      - /usr/local/bin/argus-agent
+
+networks:
+  default:
+    driver: bridge
+    ipam:
+      driver: default
+      config:
+        - subnet: 172.28.0.0/16

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

@@ -0,0 +1,22 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+SCRIPTS=(
+  "01_bootstrap.sh"
+  "02_up.sh"
+  "03_wait_and_assert_registration.sh"
+  "04_write_health_files.sh"
+  "05_assert_status_on_master.sh"
+  "06_restart_agent_and_reregister.sh"
+  "07_down.sh"
+)
+
+for script in "${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 +1,42 @@
+#!/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 "$AGENT_ROOT/../master" && pwd)"
+PRIVATE_ROOT="$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/health/$AGENT_HOSTNAME"
+MASTER_PRIVATE_DIR="$PRIVATE_ROOT/argus/master"
+METRIC_PRIVATE_DIR="$PRIVATE_ROOT/argus/metric/prometheus"
+
+mkdir -p "$AGENT_CONFIG_DIR"
+mkdir -p "$AGENT_HEALTH_DIR"
+mkdir -p "$MASTER_PRIVATE_DIR"
+mkdir -p "$METRIC_PRIVATE_DIR"
+mkdir -p "$TMP_ROOT"
+
+touch "$AGENT_HEALTH_DIR/.keep"
+
+pushd "$MASTER_ROOT" >/dev/null
+./scripts/build_images.sh --tag argus-master:dev
+popd >/dev/null
+
+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 "[INFO] Agent E2E bootstrap complete"

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

@@ -0,0 +1,37 @@
+#!/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"
+
+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: $AGENT_BINARY" >&2
+  exit 1
+fi
+
+compose() {
+  if docker compose version >/dev/null 2>&1; then
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
+
+docker network rm tests_default >/dev/null 2>&1 || true
+
+pushd "$TEST_ROOT" >/dev/null
+compose down --remove-orphans || true
+compose up -d
+popd >/dev/null
+
+echo "[INFO] Master+Agent stack started"

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

@@ -0,0 +1,65 @@
+#!/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"
+NODE_FILE="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME/node.json"
+
+mkdir -p "$TMP_ROOT"
+
+node_id=""
+for _ in {1..30}; do
+  sleep 2
+  response=$(curl -sS "$API_BASE/nodes" || true)
+  if [[ -z "$response" ]]; then
+    continue
+  fi
+  list_file="$TMP_ROOT/nodes_list.json"
+  echo "$response" > "$list_file"
+  node_id=$(python3 - "$list_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    nodes = json.load(handle)
+print(nodes[0]["id"] if nodes else "")
+PY
+)
+  if [[ -n "$node_id" ]]; then
+    break
+  fi
+ done
+
+if [[ -z "$node_id" ]]; then
+  echo "[ERROR] Agent did not register within timeout" >&2
+  exit 1
+fi
+
+echo "$node_id" > "$TMP_ROOT/node_id"
+
+if [[ ! -f "$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
+
+detail_file="$TMP_ROOT/initial_detail.json"
+curl -sS "$API_BASE/nodes/$node_id" -o "$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 $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/health/dev-e2euser-e2einst-pod-0"
+
+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_assert_status_on_master.sh

@@ -0,0 +1,53 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+TMP_ROOT="$TEST_ROOT/tmp"
+API_BASE="http://localhost:32300/api/v1/master"
+NODE_ID="$(cat "$TMP_ROOT/node_id")"
+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" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    nodes = json.load(handle)
+assert len(nodes) == 1, nodes
+entry = nodes[0]
+assert entry["node_id"], entry
+PY
+
+echo "[INFO] Master reflects agent health and nodes.json entries"

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

@@ -0,0 +1,122 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+TMP_ROOT="$TEST_ROOT/tmp"
+API_BASE="http://localhost:32300/api/v1/master"
+NODE_ID="$(cat "$TMP_ROOT/node_id")"
+AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
+NETWORK_NAME="tests_default"
+NEW_AGENT_IP="172.28.0.200"
+
+if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
+  echo "[ERROR] Agent binary path missing; rerun bootstrap" >&2
+  exit 1
+fi
+
+AGENT_BINARY="$(cat "$TMP_ROOT/agent_binary_path")"
+if [[ ! -x "$AGENT_BINARY" ]]; then
+  echo "[ERROR] Agent binary not executable: $AGENT_BINARY" >&2
+  exit 1
+fi
+
+compose() {
+  if docker compose version >/dev/null 2>&1; then
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+before_file="$TMP_ROOT/before_restart.json"
+curl -sS "$API_BASE/nodes/$NODE_ID" -o "$before_file"
+prev_last_updated=$(python3 - "$before_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+print(node.get("last_updated", ""))
+PY
+)
+prev_ip=$(python3 - "$before_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+print(node["meta_data"].get("ip", ""))
+PY
+)
+initial_ip=$(cat "$TMP_ROOT/initial_ip")
+if [[ "$prev_ip" != "$initial_ip" ]]; then
+  echo "[ERROR] Expected initial IP $initial_ip, got $prev_ip" >&2
+  exit 1
+fi
+
+pushd "$TEST_ROOT" >/dev/null
+compose rm -sf agent
+popd >/dev/null
+
+docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
+
+AGENT_DIR="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME"
+HEALTH_DIR="$TEST_ROOT/private/argus/agent/health/$AGENT_HOSTNAME"
+
+# 先以 sleep 方式启动容器，确保我们掌握注册时的网络状态
+if ! docker run -d \
+  --name argus-agent-e2e \
+  --hostname "$AGENT_HOSTNAME" \
+  --network "$NETWORK_NAME" \
+  --ip "$NEW_AGENT_IP" \
+  -v "$AGENT_DIR:/private/argus/agent/$AGENT_HOSTNAME" \
+  -v "$HEALTH_DIR:/private/argus/agent/health/$AGENT_HOSTNAME" \
+  -v "$AGENT_BINARY:/usr/local/bin/argus-agent:ro" \
+  -e MASTER_ENDPOINT=http://master:3000 \
+  -e REPORT_INTERVAL_SECONDS=2 \
+  ubuntu:24.04 \
+  sleep 300 >/dev/null; then
+  echo "[ERROR] Failed to start agent container with custom IP" >&2
+  exit 1
+fi
+
+# 在容器内启动真实 agent 进程
+if ! docker exec -d argus-agent-e2e /usr/local/bin/argus-agent; then
+  echo "[ERROR] Failed to spawn agent process inside container" >&2
+  exit 1
+fi
+
+success=false
+detail_file="$TMP_ROOT/post_restart.json"
+for _ in {1..20}; do
+  sleep 3
+  if ! curl -sS "$API_BASE/nodes/$NODE_ID" -o "$detail_file"; then
+    continue
+  fi
+  if python3 - "$detail_file" "$prev_last_updated" "$NODE_ID" "$prev_ip" "$NEW_AGENT_IP" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+prev_last_updated = sys.argv[2]
+expected_id = sys.argv[3]
+old_ip = sys.argv[4]
+expected_ip = sys.argv[5]
+last_updated = node.get("last_updated")
+current_ip = node["meta_data"].get("ip")
+assert node["id"] == expected_id
+if current_ip != expected_ip:
+    raise SystemExit(1)
+if current_ip == old_ip:
+    raise SystemExit(1)
+if not last_updated or last_updated == prev_last_updated:
+    raise SystemExit(1)
+PY
+  then
+    success=true
+    break
+  fi
+done
+
+if [[ "$success" != true ]]; then
+  echo "[ERROR] Agent did not report expected new IP $NEW_AGENT_IP after restart" >&2
+  exit 1
+fi
+
+echo "[INFO] Agent restart produced successful re-registration with IP change"

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

@@ -0,0 +1,24 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+
+compose() {
+  if docker compose version >/dev/null 2>&1; then
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
+
+pushd "$TEST_ROOT" >/dev/null
+compose down --remove-orphans
+popd >/dev/null
+
+rm -rf "$TEST_ROOT/private"
+rm -rf "$TEST_ROOT/tmp"
+
+echo "[INFO] Agent E2E environment cleaned up"

src/bind/.gitignore

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

src/bind/build/Dockerfile

@@ -0,0 +1,66 @@
+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
+
+# 配置内网 apt 源 (如果指定了内网选项)
+RUN if [ "$USE_INTRANET" = "true" ]; then \
+        echo "Configuring intranet apt sources..." && \
+        cp /etc/apt/sources.list /etc/apt/sources.list.bak && \
+        echo "deb [trusted=yes] http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
+        echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
+    fi
+
+# Update package list and install required packages
+RUN apt-get update && \
+    apt-get install -y \
+    bind9 \
+    bind9utils \
+    bind9-doc \
+    supervisor \
+    net-tools \
+    inetutils-ping \
+    vim \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# 配置部署时使用的apt源
+RUN if [ "$USE_INTRANET" = "true" ]; then \
+	echo "deb [trusted=yes] https://10.92.132.52/mirrors/ubuntu2204/ jammy main" > /etc/apt/sources.list; \
+    fi
+
+# 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 +1,100 @@
+#!/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"
+
+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.$ts.bak"
+
+  local cur_ip
+  cur_ip="$(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
+
+  return $changed
+}
+
+while true; do
+  exec 9>"$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 +1,68 @@
+#!/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/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 +1,40 @@
+#!/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
+
+# 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:
+      - "53:53/tcp"
+      - "53:53/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 +1,115 @@
+#!/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)"
+
+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 +1,38 @@
+#!/bin/bash
+
+# Start BIND9 test container
+# Usage: ./01_start_container.sh
+
+set -e
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_DIR="$(dirname "$SCRIPT_DIR")"
+
+cd "$TEST_DIR"
+
+echo "Starting BIND9 test container..."
+
+# Ensure private directory exists with proper permissions
+mkdir -p private
+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:53"

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

@@ -0,0 +1,72 @@
+#!/bin/bash
+
+# Test DNS resolution using dig
+# Usage: ./02_dig_test.sh
+
+set -e
+
+echo "Testing DNS resolution with dig..."
+
+# 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 $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$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 +1,256 @@
+#!/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
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_DIR="$(dirname "$SCRIPT_DIR")"
+
+echo "=== DNS Auto-Sync Functionality Test ==="
+
+# 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 $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$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 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=$(dig @localhost invalid.argus.com A +short 2>/dev/null || echo "NO_RESULT")
+if [ "$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 +1,112 @@
+#!/bin/bash
+
+# Test DNS configuration reload with IP modification
+# Usage: ./03_reload_test.sh
+
+set -e
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_DIR="$(dirname "$SCRIPT_DIR")"
+
+echo "=== DNS Configuration Reload Test ==="
+
+# 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 $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$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 +1,115 @@
+#!/bin/bash
+
+# Test configuration persistence after container restart
+# Usage: ./04_persistence_test.sh
+
+set -e
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_DIR="$(dirname "$SCRIPT_DIR")"
+
+echo "=== Configuration Persistence Test ==="
+
+# 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 $hostname.argus.com A +short 2>/dev/null || echo "QUERY_FAILED")
+
+    if [ "$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 +1,87 @@
+#!/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")"
+
+# 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 +1,56 @@
+FROM docker.elastic.co/elasticsearch/elasticsearch:8.13.4
+
+# 切换到 root 用户进行系统级安装
+USER root
+
+# 修改elasticsearch用户的UID和GID
+RUN usermod -u 2133 elasticsearch && \
+    groupmod -g 2015 elasticsearch && \
+    chown -R elasticsearch:elasticsearch /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 +1,47 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Fluent Bit setup in Ubuntu container..."
+
+# 安装必要的工具
+echo "[INFO] Installing required packages..."
+export DEBIAN_FRONTEND=noninteractive
+apt-get update -qq
+apt-get install -y -qq curl
+
+# 解压bundle到/tmp
+echo "[INFO] Extracting fluent-bit bundle..."
+cp -r /private/etc /tmp
+cp -r /private/packages /tmp
+cd /tmp
+
+# 安装 Fluent Bit 从 deb 包
+echo "[INFO] Installing Fluent Bit from deb package..."
+dpkg -i /tmp/packages/fluent-bit_3.1.9_amd64.deb || true
+apt-get install -f -y -qq  # 解决依赖问题
+
+# 验证 Fluent Bit 可以运行
+echo "[INFO] Fluent Bit version:"
+/opt/fluent-bit/bin/fluent-bit --version
+
+# 创建配置目录
+mkdir -p /etc/fluent-bit
+cp -r /tmp/etc/* /etc/fluent-bit/
+
+# 创建日志和缓冲区目录
+mkdir -p /logs/train /logs/infer /buffers
+chmod 755 /logs/train /logs/infer /buffers
+
+# 等待 Elasticsearch 就绪
+echo "[INFO] Waiting for Elasticsearch to be ready..."
+while ! curl -fs http://${ES_HOST}:${ES_PORT}/_cluster/health >/dev/null 2>&1; do
+    echo "  Waiting for ES at ${ES_HOST}:${ES_PORT}..."
+    sleep 5
+done
+echo "[INFO] Elasticsearch is ready"
+
+# 启动 Fluent Bit
+echo "[INFO] Starting Fluent Bit with configuration from /etc/fluent-bit/"
+echo "[INFO] Command: /opt/fluent-bit/bin/fluent-bit --config=/etc/fluent-bit/fluent-bit.conf"
+exec /opt/fluent-bit/bin/fluent-bit \
+    --config=/etc/fluent-bit/fluent-bit.conf

src/log/kibana/build/Dockerfile

@@ -0,0 +1,60 @@
+FROM docker.elastic.co/kibana/kibana:8.13.4
+
+# 切换到 root 用户进行系统级安装
+USER root
+
+# 修改kibana用户的UID和GID
+RUN usermod -u 2133 kibana && \
+    groupmod -g 2015 kibana && \
+    chown -R kibana:kibana /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 +1,146 @@
+#!/bin/bash
+set -euo pipefail
+
+ES_HOST="${ELASTICSEARCH_HOSTS:-http://es:9200}"
+KB_HOST="http://localhost: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 $idx already has 0 replicas, skipping"
+        fi
+    done
+}
+
+# 幂等创建数据视图
+create_data_views_idempotent() {
+    echo "[INFO] Checking and creating data views..."
+
+    # 检查是否存在匹配的索引
+    local train_indices=$(curl -s "$ES_HOST/_cat/indices/train-*?h=index" 2>/dev/null | wc -l || echo "0")
+    local infer_indices=$(curl -s "$ES_HOST/_cat/indices/infer-*?h=index" 2>/dev/null | wc -l || echo "0")
+
+    # 创建 train 数据视图
+    if [ "$train_indices" -gt 0 ]; then
+        # 检查数据视图是否已存在
+        local train_exists=$(curl -s "$KB_HOST/api/data_views" -H 'kbn-xsrf: true' 2>/dev/null | grep '"title":"train-\*"' | wc -l )
+
+        if [ "$train_exists" -eq 0 ]; then
+            echo "[INFO] Creating data view for train-* indices"
+            curl -fsS -X POST "$KB_HOST/api/data_views/data_view" \
+                -H 'kbn-xsrf: true' \
+                -H 'Content-Type: application/json' \
+                -d '{"data_view":{"name":"train","title":"train-*","timeFieldName":"@timestamp"}}' \
+                >/dev/null && echo "[OK] Created train data view" || echo "[WARN] Failed to create train data view"
+        else
+            echo "[INFO] Train data view already exists, skipping"
+        fi
+    else
+        echo "[INFO] No train-* indices found, skipping train data view creation"
+    fi
+
+    # 创建 infer 数据视图
+    if [ "$infer_indices" -gt 0 ]; then
+        # 检查数据视图是否已存在
+        local infer_exists=$(curl -s "$KB_HOST/api/data_views" -H 'kbn-xsrf: true' 2>/dev/null | grep '"title":"infer-\*"' | wc -l )
+
+        if [ "$infer_exists" -eq 0 ]; then
+            echo "[INFO] Creating data view for infer-* indices"
+            curl -fsS -X POST "$KB_HOST/api/data_views/data_view" \
+                -H 'kbn-xsrf: true' \
+                -H 'Content-Type: application/json' \
+                -d '{"data_view":{"name":"infer","title":"infer-*","timeFieldName":"@timestamp"}}' \
+                >/dev/null && echo "[OK] Created infer data view" || echo "[WARN] Failed to create infer data view"
+        else
+            echo "[INFO] Infer data view already exists, skipping"
+        fi
+    else
+        echo "[INFO] No infer-* indices found, skipping infer data view creation"
+    fi
+}
+
+# 主逻辑
+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

src/log/tests/docker-compose.yml

@@ -0,0 +1,85 @@
+version: "3.8"
+services:
+  es:
+    build:
+      context: ../elasticsearch/build
+      dockerfile: Dockerfile
+    image: argus-elasticsearch:latest
+    environment:
+      - discovery.type=single-node
+      - xpack.security.enabled=false
+      - ES_JAVA_OPTS=-Xms512m -Xmx512m
+    volumes:
+      - ./private/argus/:/private/argus/
+    ports: ["9200:9200"]
+    healthcheck:
+      test: ["CMD-SHELL", "curl -fs http://localhost:9200 >/dev/null || exit 1"]
+      interval: 10s
+      timeout: 5s
+      retries: 30
+
+  kibana:
+    build:
+      context: ../kibana/build
+      dockerfile: Dockerfile
+    image: argus-kibana:latest
+    environment:
+      - ELASTICSEARCH_HOSTS=http://es.log.argus.com:9200
+    volumes:
+      - ./private/argus/:/private/argus/
+    ports: ["5601:5601"]
+    depends_on:
+      es:
+        condition: service_healthy
+
+  fluent-bit-host01:
+    image: ubuntu:22.04
+    environment:
+      - CLUSTER=local
+      - RACK=dev
+      - HOSTNAME=host01
+      - ES_HOST=es
+      - ES_PORT=9200
+    volumes:
+      - ../fluent-bit/build:/private/
+    ports: ["2020:2020"]
+    depends_on:
+      es:
+        condition: service_healthy
+    command: /private/start-fluent-bit.sh
+    healthcheck:
+      test: ["CMD-SHELL", "curl -fs http://localhost:2020/api/v2/metrics >/dev/null || exit 1"]
+      interval: 15s
+      timeout: 10s
+      retries: 30
+
+  fluent-bit-host02:
+    image: ubuntu:22.04
+    environment:
+      - CLUSTER=local
+      - RACK=dev
+      - HOSTNAME=host02
+      - ES_HOST=es
+      - ES_PORT=9200
+    volumes:
+      - ../fluent-bit/build:/private/
+    ports: ["2021:2020"]
+    depends_on:
+      es:
+        condition: service_healthy
+    command: /private/start-fluent-bit.sh
+    healthcheck:
+      test: ["CMD-SHELL", "curl -fs http://localhost:2020/api/v2/metrics >/dev/null || exit 1"]
+      interval: 15s
+      timeout: 10s
+      retries: 30
+
+  bind9:
+      image: argus-bind9:latest
+      ports:
+        - "53:53/tcp"
+        - "53:53/udp"
+      volumes:
+        - ./private/argus:/private/argus/
+      restart: unless-stopped
+

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

@@ -0,0 +1,30 @@
+#!/usr/bin/env bash
+set -euo pipefail
+root="$(cd "$(dirname "${BASH_SOURCE[0]}")/../" && pwd)"
+
+# 创建新的private目录结构 (基于argus目录结构)
+echo "[INFO] Creating private directory structure for supervisor-based containers..."
+mkdir -p "$root/private/argus/log/elasticsearch"
+mkdir -p "$root/private/argus/log/kibana"
+mkdir -p "$root/private/argus/etc/"
+
+
+# 设置数据目录权限（ES 和 Kibana 容器都使用 UID 1000）
+echo "[INFO] Setting permissions for data directories..."
+sudo chown -R 2133:2015 "$root/private/argus/log/elasticsearch" 2>/dev/null || true
+sudo chown -R 2133:2015 "$root/private/argus/log/kibana" 2>/dev/null || true
+sudo chown -R 2133:2015 "$root/private/argus/etc" 2>/dev/null || true
+
+echo "[INFO] Supervisor-based containers will manage their own scripts and configurations"
+
+# 检查fluent-bit相关文件是否存在
+if [[ ! -f "$root/../fluent-bit/fluent-bit-bundle.tar.gz" ]]; then
+    echo "[WARN] fluent-bit/fluent-bit-bundle.tar.gz 不存在，请确保已创建该文件"
+fi
+
+if [[ ! -f "$root/../fluent-bit/start-fluent-bit.sh" ]]; then
+    echo "[WARN] fluent-bit/start-fluent-bit.sh 不存在，请确保已创建该启动脚本"
+fi
+
+echo "[OK] 初始化完成: private/argus/log/{elasticsearch,kibana}"
+echo "[INFO] Fluent-bit files should be in fluent-bit/ directory"

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

@@ -0,0 +1,10 @@
+#!/usr/bin/env bash
+set -euo pipefail
+cd "$(dirname "$0")/.."
+compose_cmd="docker compose"
+if ! $compose_cmd version >/dev/null 2>&1; then
+  if command -v docker-compose >/dev/null 2>&1; then compose_cmd="docker-compose"; else
+    echo "需要 Docker Compose，请安装后重试" >&2; exit 1; fi
+fi
+$compose_cmd -p logging-mvp up -d --remove-orphans
+echo "[OK] 服务已启动：ES http://localhost:9200  Kibana http://localhost:5601  Fluent-Bit host01 http://localhost:2020  Fluent-Bit host02 http://localhost:2021"

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

@@ -0,0 +1,31 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 获取fluent-bit-host01容器名称
+container_name="logging-mvp-fluent-bit-host01-1"
+
+# 检查容器是否存在并运行
+if ! docker ps | grep -q "$container_name"; then
+    echo "[ERROR] Fluent Bit容器 $container_name 未运行"
+    exit 1
+fi
+
+# 创建日志目录
+docker exec "$container_name" mkdir -p /logs/train /logs/infer
+
+# 写入训练日志 (host01)
+docker exec "$container_name" sh -c "printf '%s INFO [host01] training step=1 loss=1.23 model=bert\n' \"\$(date '+%F %T')\" >> /logs/train/train-demo.log"
+docker exec "$container_name" sh -c "printf '%s INFO [host01] training step=2 loss=1.15 model=bert\n' \"\$(date '+%F %T')\" >> /logs/train/train-demo.log"
+
+# 写入推理日志 (host01)
+docker exec "$container_name" sh -c "printf '%s ERROR [host01] inference failed on batch=1\n' \"\$(date '+%F %T')\" >> /logs/infer/infer-demo.log"
+docker exec "$container_name" sh -c "cat <<'STACK' >> /logs/infer/infer-demo.log
+Traceback (most recent call last):
+  File \"inference.py\", line 15, in <module>
+    raise RuntimeError(\"CUDA out of memory on host01\")
+RuntimeError: CUDA out of memory on host01
+STACK"
+
+echo "[OK] 已通过docker exec写入测试日志到 host01 容器内："
+echo " - /logs/train/train-demo.log"
+echo " - /logs/infer/infer-demo.log"

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

@@ -0,0 +1,27 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 获取fluent-bit-host02容器名称
+container_name="logging-mvp-fluent-bit-host02-1"
+
+# 检查容器是否存在并运行
+if ! docker ps | grep -q "$container_name"; then
+    echo "[ERROR] Fluent Bit容器 $container_name 未运行"
+    exit 1
+fi
+
+# 创建日志目录
+docker exec "$container_name" mkdir -p /logs/train /logs/infer
+
+# 写入训练日志 (host02)
+docker exec "$container_name" sh -c "printf '%s INFO [host02] training step=1 loss=1.45 model=gpt\n' \"\$(date '+%F %T')\" >> /logs/train/train-demo.log"
+docker exec "$container_name" sh -c "printf '%s INFO [host02] training step=2 loss=1.38 model=gpt\n' \"\$(date '+%F %T')\" >> /logs/train/train-demo.log"
+docker exec "$container_name" sh -c "printf '%s INFO [host02] training step=3 loss=1.32 model=gpt\n' \"\$(date '+%F %T')\" >> /logs/train/train-demo.log"
+
+# 写入推理日志 (host02)  
+docker exec "$container_name" sh -c "printf '%s WARN [host02] inference slow on batch=5 latency=2.3s\n' \"\$(date '+%F %T')\" >> /logs/infer/infer-demo.log"
+docker exec "$container_name" sh -c "printf '%s INFO [host02] inference completed batch=6 latency=0.8s\n' \"\$(date '+%F %T')\" >> /logs/infer/infer-demo.log"
+
+echo "[OK] 已通过docker exec写入测试日志到 host02 容器内："
+echo " - /logs/train/train-demo.log"
+echo " - /logs/infer/infer-demo.log"

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

@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+set -euo pipefail
+ES="${ES:-http://localhost:9200}"
+echo "[i] 查询 ES 端点：$ES"
+curl -fsS "$ES/_cat/indices?v" | egrep 'train-|infer-|logstash' || true
+printf "train-* 计数："; curl -fsS "$ES/train-*/_count" | sed -E 's/.*"count":([0-9]+).*/\1/'; echo
+printf "infer-* 计数："; curl -fsS "$ES/infer-*/_count" | sed -E 's/.*"count":([0-9]+).*/\1/'; echo

src/log/tests/scripts/05_down.sh

@@ -0,0 +1,21 @@
+#!/usr/bin/env bash
+set -euo pipefail
+cd "$(dirname "$0")/.."
+compose_cmd="docker compose"
+if ! $compose_cmd version >/dev/null 2>&1; then
+  if command -v docker-compose >/dev/null 2>&1; then compose_cmd="docker-compose"; else
+    echo "需要 Docker Compose，请安装后重试" >&2; exit 1; fi
+fi
+$compose_cmd -p logging-mvp down
+echo "[OK] 已停止所有容器"
+
+# 清理private目录内容
+echo "[INFO] 清理private目录内容..."
+cd "$(dirname "$0")/.."
+if [ -d "private" ]; then
+    # 删除private目录及其所有内容
+    rm -rf private
+    echo "[OK] 已清理private目录"
+else
+    echo "[INFO] private目录不存在，无需清理"
+fi

src/log/tests/scripts/06_dns_test.sh

@@ -0,0 +1,208 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+echo "======================================="
+echo "ARGUS DNS监控功能测试"
+echo "======================================="
+echo ""
+
+# 记录测试开始时间
+test_start_time=$(date +%s)
+
+# 函数：显示测试步骤
+show_step() {
+    echo ""
+    echo "🔄 Step $1: $2"
+    echo "----------------------------------------"
+}
+
+# 函数：验证步骤结果
+verify_step() {
+    if [ $? -eq 0 ]; then
+        echo "✅ $1 - SUCCESS"
+    else
+        echo "❌ $1 - FAILED"
+        exit 1
+    fi
+}
+
+# 函数：等待服务就绪
+wait_for_services() {
+    echo "[INFO] Waiting for services to be ready..."
+    local max_attempts=60
+    local attempt=1
+
+    while [ $attempt -le $max_attempts ]; do
+        if curl -fs http://localhost:9200/_cluster/health >/dev/null 2>&1 && \
+           curl -fs http://localhost:5601/api/status >/dev/null 2>&1; then
+            echo "[OK] Services are ready!"
+            return 0
+        fi
+        echo "    Waiting for services... ($attempt/$max_attempts)"
+        sleep 5
+        ((attempt++))
+    done
+
+    echo "[ERROR] Services not ready after $max_attempts attempts"
+    return 1
+}
+
+# 函数：检查容器中的/etc/resolv.conf
+check_resolv_conf() {
+    local service_name=$1
+    local expected_dns=$2
+
+    echo "[INFO] 检查 $service_name 容器的 /etc/resolv.conf..."
+
+    local resolv_content=$(docker exec "${service_name}" cat /etc/resolv.conf 2>/dev/null || echo "")
+    if echo "$resolv_content" | grep -q "nameserver $expected_dns"; then
+        echo "✅ $service_name resolv.conf contains nameserver $expected_dns"
+        return 0
+    else
+        echo "❌ $service_name resolv.conf does not contain nameserver $expected_dns"
+        echo "实际内容:"
+        echo "$resolv_content"
+        return 1
+    fi
+}
+
+# 函数：检查DNS监控日志
+check_dns_monitor_logs() {
+    local service_name=$1
+
+    echo "[INFO] 检查 $service_name 的DNS监控日志..."
+
+    local dns_logs=$(docker exec "$service_name" tail -n 20 /var/log/supervisor/dns-monitor.log 2>/dev/null || echo "")
+    if [ -n "$dns_logs" ]; then
+        echo "✅ $service_name DNS监控日志存在"
+        echo "最近的日志:"
+        echo "$dns_logs"
+        return 0
+    else
+        echo "❌ $service_name DNS监控日志为空或不存在"
+        return 1
+    fi
+}
+
+# 函数：确保目录结构存在
+ensure_directories() {
+    echo "[INFO] 确保目录结构存在..."
+    # 确保目录存在
+    mkdir -p ./private/argus/etc/
+    echo "✅ 目录结构准备完成（注：使用真实的update-dns.sh脚本）"
+}
+
+# 开始DNS监控测试
+show_step "1" "Bootstrap - Initialize environment"
+./scripts/01_bootstrap.sh
+verify_step "Bootstrap"
+
+# 确保目录结构
+ensure_directories
+
+show_step "2" "Startup - Start all services"
+./scripts/02_up.sh
+verify_step "Service startup"
+
+# 等待服务完全就绪
+wait_for_services || exit 1
+
+show_step "3" "Create initial DNS configuration"
+# 创建初始的DNS配置文件 - 只有一个IP
+echo "[INFO] 创建初始的dns.conf文件 (8.8.8.8)..."
+cat > ./private/argus/etc/dns.conf << 'EOF'
+8.8.8.8
+EOF
+
+echo "✅ 初始dns.conf文件创建成功 (8.8.8.8)"
+verify_step "Initial DNS configuration creation"
+
+# 等待DNS监控检测到配置文件
+echo "[INFO] 等待DNS监控检测并处理初始配置..."
+sleep 15
+
+show_step "4" "Verify initial DNS configuration processing"
+# 检查两个容器的DNS监控日志
+check_dns_monitor_logs "logging-mvp-es-1"
+verify_step "Elasticsearch DNS monitor logs"
+
+check_dns_monitor_logs "logging-mvp-kibana-1"
+verify_step "Kibana DNS monitor logs"
+
+# 检查resolv.conf是否包含新的DNS服务器
+check_resolv_conf "logging-mvp-es-1" "8.8.8.8"
+verify_step "Elasticsearch resolv.conf initial check"
+
+check_resolv_conf "logging-mvp-kibana-1" "8.8.8.8"
+verify_step "Kibana resolv.conf initial check"
+
+show_step "5" "Modify DNS configuration and test auto-update"
+# 修改DNS配置文件 - 改为另一个IP
+echo "[INFO] 修改dns.conf文件，改为1.1.1.1..."
+cat > ./private/argus/etc/dns.conf << 'EOF'
+1.1.1.1
+EOF
+
+echo "✅ dns.conf文件更新成功，改为1.1.1.1"
+
+# 等待DNS监控检测到配置变化
+echo "[INFO] 等待DNS监控检测配置变化并执行更新..."
+sleep 15
+
+show_step "6" "Verify DNS configuration auto-update"
+# 再次检查DNS监控日志，应该看到配置变化检测
+echo "[INFO] 检查DNS监控是否检测到配置变化..."
+
+# 检查elasticsearch容器
+echo "[INFO] 检查elasticsearch容器的DNS监控日志（最近30行）..."
+docker exec logging-mvp-es-1 tail -n 30 /var/log/supervisor/dns-monitor.log || true
+
+# 检查kibana容器
+echo "[INFO] 检查kibana容器的DNS监控日志（最近30行）..."
+docker exec logging-mvp-kibana-1 tail -n 30 /var/log/supervisor/dns-monitor.log || true
+
+# 验证新的DNS服务器是否被添加到resolv.conf
+check_resolv_conf "logging-mvp-es-1" "1.1.1.1"
+verify_step "Elasticsearch resolv.conf after update"
+
+check_resolv_conf "logging-mvp-kibana-1" "1.1.1.1"
+verify_step "Kibana resolv.conf after update"
+
+show_step "7" "Final verification - Check DNS configuration"
+# 最终验证DNS配置
+echo "[INFO] 最终验证elasticsearch容器的resolv.conf..."
+docker exec logging-mvp-es-1 cat /etc/resolv.conf
+
+echo "[INFO] 最终验证kibana容器的resolv.conf..."
+docker exec logging-mvp-kibana-1 cat /etc/resolv.conf
+
+echo "[INFO] 最终dns.conf内容:"
+cat ./private/argus/etc/dns.conf
+
+verify_step "Final DNS configuration verification"
+
+show_step "8" "Cleanup - Stop all services"
+./scripts/05_down.sh
+verify_step "Service cleanup"
+
+# 清理测试文件
+rm -f ./private/argus/etc/dns.conf
+# 注：不删除update-dns.sh，因为这是真实的脚本
+
+# 计算总测试时间
+test_end_time=$(date +%s)
+total_time=$((test_end_time - test_start_time))
+
+echo ""
+echo "======================================="
+echo "🎉 DNS监控功能测试完成!"
+echo "======================================="
+echo "📊 测试总结:"
+echo "   • 总耗时: ${total_time}秒"
+echo "   • 初始DNS配置: 8.8.8.8"
+echo "   • 更新DNS配置: 1.1.1.1"
+echo "   • DNS监控脚本正常工作"
+echo "   • 容器resolv.conf自动覆盖更新成功"
+echo ""
+echo "✅ DNS自动更新功能测试通过!"
+echo ""

src/log/tests/scripts/e2e_test.sh

@@ -0,0 +1,169 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+echo "======================================="
+echo "ARGUS Log System End-to-End Test"
+echo "======================================="
+echo ""
+
+# 记录测试开始时间
+test_start_time=$(date +%s)
+
+# 函数：获取ES中的日志计数
+get_log_count() {
+    local train_count=$(curl -s "http://localhost:9200/train-*/_count" 2>/dev/null | grep -o '"count":[0-9]*' | cut -d':' -f2 || echo "0")
+    local infer_count=$(curl -s "http://localhost:9200/infer-*/_count" 2>/dev/null | grep -o '"count":[0-9]*' | cut -d':' -f2 || echo "0")
+    echo "$((train_count + infer_count))"
+}
+
+# 函数：等待服务就绪
+wait_for_services() {
+    echo "[INFO] Waiting for all services to be ready..."
+    local max_attempts=60
+    local attempt=1
+
+    while [ $attempt -le $max_attempts ]; do
+        if curl -fs http://localhost:9200/_cluster/health >/dev/null 2>&1 && \
+           curl -fs http://localhost:5601/api/status >/dev/null 2>&1 && \
+           curl -fs http://localhost:2020/api/v2/metrics >/dev/null 2>&1 && \
+           curl -fs http://localhost:2021/api/v2/metrics >/dev/null 2>&1; then
+            echo "[OK] All services are ready!"
+            return 0
+        fi
+        echo "    Waiting for services... ($attempt/$max_attempts)"
+        sleep 5
+        ((attempt++))
+    done
+
+    echo "[ERROR] Services not ready after $max_attempts attempts"
+    return 1
+}
+
+# 函数：显示测试步骤
+show_step() {
+    echo ""
+    echo "🔄 Step $1: $2"
+    echo "----------------------------------------"
+}
+
+# 函数：验证步骤结果
+verify_step() {
+    if [ $? -eq 0 ]; then
+        echo "✅ $1 - SUCCESS"
+    else
+        echo "❌ $1 - FAILED"
+        exit 1
+    fi
+}
+
+# 开始端到端测试
+show_step "1" "Bootstrap - Initialize environment"
+./scripts/01_bootstrap.sh
+verify_step "Bootstrap"
+
+show_step "2" "Startup - Start all services"
+./scripts/02_up.sh
+verify_step "Service startup"
+
+# 等待服务完全就绪
+wait_for_services || exit 1
+
+# 记录发送测试数据前的日志计数
+initial_count=$(get_log_count)
+echo "[INFO] Initial log count: $initial_count"
+
+show_step "3a" "Send test data - Host01"
+./scripts/03_send_test_host01.sh
+verify_step "Test data sending (host01)"
+
+show_step "3b" "Send test data - Host02"
+./scripts/03_send_test_host02.sh
+verify_step "Test data sending (host02)"
+
+# 等待数据被处理
+echo "[INFO] Waiting for data to be processed..."
+sleep 10
+
+show_step "4" "Verify data - Query Elasticsearch"
+./scripts/04_query_es.sh
+verify_step "Data verification"
+
+# 记录发送测试数据后的日志计数
+final_count=$(get_log_count)
+echo "[INFO] Final log count: $final_count"
+
+# 验证日志数量是否增加
+if [ "$final_count" -gt "$initial_count" ]; then
+    added_logs=$((final_count - initial_count))
+    echo "✅ Log count verification - SUCCESS: Added $added_logs logs (from $initial_count to $final_count)"
+else
+    echo "❌ Log count verification - FAILED: Expected count to increase, but got $initial_count -> $final_count"
+    exit 1
+fi
+
+# 验证预期的最小日志数量（每个主机应该发送一些日志）
+expected_min_logs=4  # 至少应该有几条日志
+if [ "$final_count" -ge "$expected_min_logs" ]; then
+    echo "✅ Minimum log threshold - SUCCESS: $final_count logs (>= $expected_min_logs expected)"
+else
+    echo "❌ Minimum log threshold - FAILED: Only $final_count logs (>= $expected_min_logs expected)"
+    exit 1
+fi
+
+# 检查服务健康状态
+show_step "Health" "Check service health"
+echo "[INFO] Checking service health..."
+
+# 检查 Elasticsearch 健康状态
+es_health=$(curl -s "http://localhost:9200/_cluster/health" | grep -o '"status":"[^"]*"' | cut -d'"' -f4)
+if [ "$es_health" = "green" ] || [ "$es_health" = "yellow" ]; then
+    echo "✅ Elasticsearch health: $es_health"
+else
+    echo "❌ Elasticsearch health: $es_health"
+fi
+
+# 检查 Kibana 状态
+if curl -fs "http://localhost:5601/api/status" >/dev/null 2>&1; then
+    kb_status="available"
+    echo "✅ Kibana status: $kb_status"
+else
+    kb_status="unavailable"
+    echo "⚠️  Kibana status: $kb_status"
+fi
+
+# 检查 Fluent-Bit 指标
+fb_host01_uptime=$(curl -s "http://localhost:2020/api/v2/metrics" | grep "fluentbit_uptime" | head -1 | grep -o "[0-9]\+$" || echo "0")
+fb_host02_uptime=$(curl -s "http://localhost:2021/api/v2/metrics" | grep "fluentbit_uptime" | head -1 | grep -o "[0-9]\+$" || echo "0")
+
+if [ "$fb_host01_uptime" -gt 0 ] && [ "$fb_host02_uptime" -gt 0 ]; then
+    echo "✅ Fluent-Bit services: host01 uptime=${fb_host01_uptime}s, host02 uptime=${fb_host02_uptime}s"
+else
+    echo "⚠️  Fluent-Bit services: host01 uptime=${fb_host01_uptime}s, host02 uptime=${fb_host02_uptime}s"
+fi
+
+verify_step "Service health check"
+
+show_step "5" "Cleanup - Stop all services"
+./scripts/05_down.sh
+verify_step "Service cleanup"
+
+# 计算总测试时间
+test_end_time=$(date +%s)
+total_time=$((test_end_time - test_start_time))
+
+echo ""
+echo "======================================="
+echo "🎉 END-TO-END TEST COMPLETED SUCCESSFULLY!"
+echo "======================================="
+echo "📊 Test Summary:"
+echo "   • Initial logs: $initial_count"
+echo "   • Final logs: $final_count"
+echo "   • Added logs: $added_logs"
+echo "   • Total time: ${total_time}s"
+echo "   • ES health: $es_health"
+echo "   • Kibana status: $kb_status"
+echo "   • DNS resolv: ✅ Passed (ES domain verified)"
+echo "   • All services started and stopped successfully"
+echo ""
+echo "✅ The ARGUS log system is working correctly!"
+echo ""

src/master/Dockerfile

@@ -0,0 +1,32 @@
+FROM python:3.11-slim
+
+SHELL ["/bin/bash", "-c"]
+
+ARG PIP_INDEX_URL=
+ARG USE_OFFLINE=0
+ENV PIP_NO_CACHE_DIR=1 \
+    PYTHONUNBUFFERED=1 \
+    PYTHONPATH=/app
+
+WORKDIR /app
+
+COPY requirements.txt ./
+COPY offline_wheels/ /opt/offline_wheels/
+
+RUN set -euxo pipefail \
+    && if [[ "$USE_OFFLINE" == "1" ]]; then \
+         python -m pip install --no-index --find-links /opt/offline_wheels -r requirements.txt; \
+       else \
+         python -m pip install --upgrade pip \
+         && if [[ -n "$PIP_INDEX_URL" ]]; then \
+              PIP_INDEX_URL="$PIP_INDEX_URL" python -m pip install -r requirements.txt; \
+            else \
+              python -m pip install -r requirements.txt; \
+            fi; \
+       fi
+
+COPY app ./app
+
+EXPOSE 3000
+
+CMD ["gunicorn", "--bind", "0.0.0.0:3000", "app:create_app()"]

src/master/README.md

@@ -0,0 +1,171 @@
+# Argus Master 模块
+
+Argus Master 是基于 Flask + SQLite 的节点管理服务，负责：
+
+- 接收 agent 的注册与重注册请求，分配/校验节点 ID。
+- 存储节点元数据、配置、健康状态，并根据上报时间计算在线状态。
+- 输出仅包含在线节点的 `nodes.json`，供其他模块（如 metric）消费。
+- 提供查询、配置更新、统计等 REST API。
+
+## 构建与运行
+
+```bash
+cd src/master
+./scripts/build_images.sh            # 生成 argus-master:dev 镜像
+```
+
+如需离线构建，先在有网环境运行准备脚本：
+
+```bash
+cd src/master
+./scripts/prepare_offline_wheels.sh --pip-version 25.2  # 可选 --clean
+```
+
+脚本会把 `requirements.txt` 及 pip 指定版本全部下载到 `offline_wheels/`。随后将源码目录（含该子目录）与基础镜像一并拷贝到内网，执行：
+
+```bash
+cd src/master
+./scripts/build_images.sh --offline --tag argus-master:dev
+```
+
+若内网缺少 `python:3.11-slim`，请提前在外网 `docker save` 后通过离线介质 `docker load`。
+
+本仓库提供的端到端测试会使用 `src/master/tests/docker-compose.yml` 启动示例环境：
+
+```bash
+cd src/master/tests
+./scripts/01_up_master.sh            # 构建镜像并启动容器，监听 http://localhost:31300
+```
+
+服务日志与数据默认写入 `tests/private/argus/master/`（或自定义的挂载目录）。
+
+## 运行时环境变量
+
+| 变量 | 默认值 | 说明 |
+| --- | --- | --- |
+| `DB_PATH` | `/private/argus/master/db.sqlite3` | SQLite 数据库存放路径。目录会在启动时自动创建。 |
+| `METRIC_NODES_JSON_PATH` | `/private/argus/metric/prometheus/nodes.json` | `nodes.json` 输出位置，仅包含在线节点。采用原子写入避免部分文件。 |
+| `OFFLINE_THRESHOLD_SECONDS` | `180` | 若距离最近一次上报时间超过该值，调度器会将节点标记为 `offline`。 |
+| `ONLINE_THRESHOLD_SECONDS` | `120` | 若最新上报时间距当前不超过该值，则标记为 `online`。范围处于两个阈值之间时保持原状态。 |
+| `SCHEDULER_INTERVAL_SECONDS` | `30` | 调度器检查节点状态与刷新 `nodes.json` 的周期。 |
+| `NODE_ID_PREFIX` | `A` | 新节点 ID 的前缀，实际 ID 形如 `A1`、`A2`。 |
+| `AUTH_MODE` | `disabled` | 预留的认证开关，当前固定为禁用。 |
+
+## REST API 详解
+
+基础路径：`/api/v1/master`，全部返回 JSON。
+
+### 1. `GET /nodes`
+- **用途**：获取所有节点的简要信息。
+- **响应示例**：
+  ```json
+  [
+    {"id": "A1", "name": "dev-user-inst-pod-0", "status": "online", "type": "agent", "version": "1.1.0"}
+  ]
+  ```
+
+### 2. `GET /nodes/{id}`
+- **用途**：获取节点详情（包含配置、健康、持久化时间戳等）。
+- **错误**：`404` 表示节点不存在。
+
+### 3. `POST /nodes`
+- **用途**：注册或重注册节点。
+- **请求体**：
+  ```json
+  {
+    "id": "A1",            // 可选，重注册时携带
+    "name": "dev-user-inst-pod-0",
+    "type": "agent",
+    "version": "1.1.0",
+    "meta_data": {
+      "hostname": "dev-user-inst-pod-0",
+      "ip": "10.0.0.10",
+      "env": "dev",
+      "user": "testuser",
+      "instance": "testinst",
+      "cpu_number": 4,
+      "memory_in_bytes": 2147483648,
+      "gpu_number": 0
+    }
+  }
+  ```
+- **成功返回**：
+  - 新节点：`201 Created`，返回完整节点对象。
+  - 重注册：`200 OK`，返回更新后的节点对象。
+- **错误情况**：
+  - `404 Not Found`：携带的 ID 在 Master 中不存在。
+  - `500 Internal Server Error`：携带的 ID 与已有名称不匹配。
+  - `400 Bad Request`：请求体缺字段或类型不正确。
+
+### 4. `PUT /nodes/{id}/status`
+- **用途**：Agent 上报状态。Master 记录 `last_report`（服务器时间）与 `agent_last_report`（上报内时间），并更新 `health` 字段。
+- **请求体示例**：
+  ```json
+  {
+    "timestamp": "2025-09-24T03:24:59Z",
+    "health": {
+      "log-fluentbit": {"status": "healthy"},
+      "metric-node-exporter": {"status": "healthy"}
+    }
+  }
+  ```
+- **响应**：`200 OK`，返回最新节点对象。`404` 表示节点不存在。
+
+### 5. `PUT /nodes/{id}/config`
+- **用途**：局部更新节点配置与标签。
+- **请求体示例**：
+  ```json
+  {
+    "config": {"log_level": "debug"},
+    "label": ["gpu", "exp001"]
+  }
+  ```
+- **说明**：字段可任选其一；未提供的配置保持原值。更新标签会触发 `nodes.json` 重新生成。
+- **错误**：`404` 表示节点不存在；`400` 表示请求体不合法。
+
+### 6. `GET /nodes/statistics`
+- **用途**：统计节点总数及按状态分布。
+- **响应示例**：
+  ```json
+  {
+    "total": 2,
+    "status_statistics": [
+      {"status": "online", "count": 1},
+      {"status": "offline", "count": 1}
+    ]
+  }
+  ```
+
+### 7. 健康探针
+- `GET /healthz`：进程存活检查。
+- `GET /readyz`：数据库可用性检查（会尝试访问 `DB_PATH`）。
+
+
+如需验证离线镜像，可使用自动化脚本：
+```bash
+cd src/master/tests
+./scripts/00_e2e_test_offline.sh    # 构建离线镜像并执行完整 E2E
+```
+
+## 端到端测试场景
+
+执行 `src/master/tests/scripts/00_e2e_test.sh` 会串联以下用例（脚本 01–10）：
+
+1. **01_up_master**：构建镜像、启动容器、初始化目录与卷。 
+2. **02_verify_ready_and_nodes_json**：轮询 `/readyz`，校验初始 `nodes.json` 为 `[]`。
+3. **03_register_via_curl**：模拟 agent 注册，保存返回的节点 ID，并确认节点出现在列表接口中。
+4. **04_reregister_and_error_cases**：覆盖重注册成功、携带未知 ID 的 `404`、ID/名称不匹配触发 `500` 等场景。
+5. **05_status_report_via_curl**：上报健康信息并验证状态自动从 `initialized`→`online`→`offline`→`online` 的转换。
+6. **06_config_update_and_nodes_json**：更新配置/标签，检查 `nodes.json` 中的标签同步，并确保离线节点不会出现在文件里。
+7. **07_stats_single_node**：等待节点掉线，验证统计接口与 `nodes.json` 为空列表。
+8. **08_multi_node_stats**：注册第二节点，使一在线一离线，校验统计聚合和 `nodes.json` 仅包含在线节点。
+9. **09_restart_persistence**：重启 master 容器，确认节点数据、统计结果与 `nodes.json` 在持久化目录中保持不变。
+10. **10_down**：停止并清理容器、网络与临时目录。
+
+## 相关持久化文件
+
+- SQLite：默认位于 `DB_PATH`，包含 `nodes` 与 `kv` 两张表。
+- `nodes.json`：由调度器周期生成，仅保留状态为 `online` 的节点信息。
+- 测试用例中的 `tests/private/`、`tests/tmp/` 会随脚本自动清理，避免污染后续运行。
+
+如需在生产环境运行，可将镜像推送到私有仓库，或参考测试 Compose 配置自行部署；只需确保上述环境变量在容器内正确设置即可。

src/master/app/__init__.py

@@ -0,0 +1,41 @@
+from __future__ import annotations
+
+import atexit
+import logging
+
+from flask import Flask
+
+from .config import AppConfig, load_config
+from .routes import register_routes
+from .scheduler import StatusScheduler
+from .storage import Storage
+
+
+def create_app(config: AppConfig | None = None) -> Flask:
+    app_config = config or load_config()
+    storage = Storage(app_config.db_path, app_config.node_id_prefix)
+    scheduler = StatusScheduler(storage, app_config)
+
+    app = Flask(__name__)
+    app.config["APP_CONFIG"] = app_config
+    app.config["STORAGE"] = storage
+    app.config["SCHEDULER"] = scheduler
+
+    register_routes(app, storage, scheduler, app_config)
+
+    scheduler.start()
+
+    def _cleanup() -> None:
+        logging.getLogger("argus.master").info("Shutting down master app")
+        try:
+            scheduler.stop()
+        except Exception:  # pragma: no cover - defensive
+            logging.getLogger("argus.master").exception("Failed to stop scheduler")
+        try:
+            storage.close()
+        except Exception:  # pragma: no cover - defensive
+            logging.getLogger("argus.master").exception("Failed to close storage")
+
+    atexit.register(_cleanup)
+
+    return app

src/master/app/config.py

@@ -0,0 +1,40 @@
+from __future__ import annotations
+
+import os
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True)
+class AppConfig:
+    db_path: str
+    metric_nodes_json_path: str
+    offline_threshold_seconds: int
+    online_threshold_seconds: int
+    scheduler_interval_seconds: int
+    node_id_prefix: str
+    auth_mode: str
+
+
+def _get_int_env(name: str, default: int) -> int:
+    raw = os.environ.get(name)
+    if raw is None or raw.strip() == "":
+        return default
+    try:
+        return int(raw)
+    except ValueError as exc:
+        raise ValueError(f"Environment variable {name} must be an integer, got {raw!r}") from exc
+
+
+def load_config() -> AppConfig:
+    """读取环境变量生成配置对象，方便统一管理运行参数。"""
+    return AppConfig(
+        db_path=os.environ.get("DB_PATH", "/private/argus/master/db.sqlite3"),
+        metric_nodes_json_path=os.environ.get(
+            "METRIC_NODES_JSON_PATH", "/private/argus/metric/prometheus/nodes.json"
+        ),
+        offline_threshold_seconds=_get_int_env("OFFLINE_THRESHOLD_SECONDS", 180),
+        online_threshold_seconds=_get_int_env("ONLINE_THRESHOLD_SECONDS", 120),
+        scheduler_interval_seconds=_get_int_env("SCHEDULER_INTERVAL_SECONDS", 30),
+        node_id_prefix=os.environ.get("NODE_ID_PREFIX", "A"),
+        auth_mode=os.environ.get("AUTH_MODE", "disabled"),
+    )

src/master/app/models.py

@@ -0,0 +1,171 @@
+from __future__ import annotations
+
+import json
+from dataclasses import dataclass
+from typing import Any, Dict, Iterable, Mapping
+
+from .util import parse_iso
+
+
+class ValidationError(Exception):
+    """Raised when user payload fails validation."""
+
+
+@dataclass
+class Node:
+    id: str
+    name: str
+    type: str
+    version: str | None
+    status: str
+    config: Dict[str, Any]
+    labels: Iterable[str]
+    meta_data: Dict[str, Any]
+    health: Dict[str, Any]
+    register_time: str | None
+    last_report: str | None
+    agent_last_report: str | None
+    last_updated: str | None
+
+
+def serialize_node_row(row: Mapping[str, Any]) -> Dict[str, Any]:
+    def _json_or_default(value: str | None, default: Any) -> Any:
+        if value is None or value == "":
+            return default
+        try:
+            return json.loads(value)
+        except json.JSONDecodeError:
+            return default
+
+    config = _json_or_default(row["config_json"], {})
+    labels = _json_or_default(row["labels_json"], [])
+    meta = _json_or_default(row["meta_json"], {})
+    health = _json_or_default(row["health_json"], {})
+    return {
+        "id": row["id"],
+        "name": row["name"],
+        "type": row["type"],
+        "version": row["version"],
+        "status": row["status"],
+        "config": config if isinstance(config, dict) else {},
+        "label": list(labels) if isinstance(labels, list) else [],
+        "meta_data": meta if isinstance(meta, dict) else {},
+        "health": health if isinstance(health, dict) else {},
+        "register_time": row["register_time"],
+        "last_report": row["last_report"],
+        "agent_last_report": row["agent_last_report"],
+        "last_updated": row["last_updated"],
+    }
+
+
+def serialize_node_summary(row: Mapping[str, Any]) -> Dict[str, Any]:
+    return {
+        "id": row["id"],
+        "name": row["name"],
+        "status": row["status"],
+        "type": row["type"],
+        "version": row["version"],
+    }
+
+
+def validate_registration_payload(payload: Mapping[str, Any]) -> Dict[str, Any]:
+    if not isinstance(payload, Mapping):
+        raise ValidationError("Request body must be a JSON object")
+
+    name = payload.get("name")
+    if not isinstance(name, str) or not name.strip():
+        raise ValidationError("Field 'name' is required and must be a non-empty string")
+
+    node_type = payload.get("type", "agent")
+    if not isinstance(node_type, str) or not node_type:
+        raise ValidationError("Field 'type' must be a string")
+
+    version = payload.get("version")
+    if version is not None and not isinstance(version, str):
+        raise ValidationError("Field 'version' must be a string if provided")
+
+    meta = payload.get("meta_data")
+    if not isinstance(meta, Mapping):
+        raise ValidationError("Field 'meta_data' must be an object")
+
+    required_meta = ["hostname", "ip", "env", "user", "instance", "cpu_number", "memory_in_bytes", "gpu_number"]
+    for key in required_meta:
+        if key not in meta:
+            raise ValidationError(f"meta_data.{key} is required")
+
+    cpu_number = meta["cpu_number"]
+    memory_in_bytes = meta["memory_in_bytes"]
+    gpu_number = meta["gpu_number"]
+    if not isinstance(cpu_number, int) or cpu_number < 0:
+        raise ValidationError("meta_data.cpu_number must be a non-negative integer")
+    if not isinstance(memory_in_bytes, int) or memory_in_bytes < 0:
+        raise ValidationError("meta_data.memory_in_bytes must be a non-negative integer")
+    if not isinstance(gpu_number, int) or gpu_number < 0:
+        raise ValidationError("meta_data.gpu_number must be a non-negative integer")
+
+    node_id = payload.get("id")
+    if node_id is not None and (not isinstance(node_id, str) or not node_id.strip()):
+        raise ValidationError("Field 'id' must be a non-empty string when provided")
+
+    return {
+        "id": node_id,
+        "name": name,
+        "type": node_type,
+        "version": version,
+        "meta_data": dict(meta),
+    }
+
+
+def validate_status_payload(payload: Mapping[str, Any]) -> Dict[str, Any]:
+    if not isinstance(payload, Mapping):
+        raise ValidationError("Request body must be a JSON object")
+
+    timestamp = payload.get("timestamp")
+    if not isinstance(timestamp, str) or not timestamp:
+        raise ValidationError("Field 'timestamp' is required and must be a string")
+
+    parsed = parse_iso(timestamp)
+    if parsed is None:
+        raise ValidationError("Field 'timestamp' must be an ISO8601 datetime string")
+
+    health = payload.get("health", {})
+    if not isinstance(health, Mapping):
+        raise ValidationError("Field 'health' must be an object if provided")
+
+    sanitized_health: Dict[str, Any] = {}
+    for key, value in health.items():
+        if not isinstance(key, str):
+            raise ValidationError("Keys in 'health' must be strings")
+        if not isinstance(value, (Mapping, list, str, int, float, bool)) and value is not None:
+            raise ValidationError("Values in 'health' must be JSON-compatible")
+        sanitized_health[key] = value
+
+    return {
+        "timestamp": timestamp,
+        "parsed_timestamp": parsed,
+        "health": sanitized_health,
+    }
+
+
+def validate_config_payload(payload: Mapping[str, Any]) -> Dict[str, Any]:
+    if not isinstance(payload, Mapping):
+        raise ValidationError("Request body must be a JSON object")
+
+    result: Dict[str, Any] = {}
+    if "config" in payload:
+        config = payload["config"]
+        if not isinstance(config, Mapping):
+            raise ValidationError("Field 'config' must be an object")
+        result["config"] = dict(config)
+
+    if "label" in payload:
+        labels = payload["label"]
+        if not isinstance(labels, list) or not all(isinstance(item, str) for item in labels):
+            raise ValidationError("Field 'label' must be an array of strings")
+        result["label"] = list(labels)
+
+    if not result:
+        raise ValidationError("At least one of 'config' or 'label' must be provided")
+
+    return result
+

src/master/app/nodes_api.py

@@ -0,0 +1,155 @@
+from __future__ import annotations
+
+import logging
+from http import HTTPStatus
+from typing import Any, Mapping
+
+from flask import Blueprint, jsonify, request
+
+from .models import (
+    ValidationError,
+    validate_config_payload,
+    validate_registration_payload,
+    validate_status_payload,
+)
+from .scheduler import StatusScheduler
+from .storage import Storage
+from .util import to_iso, utcnow
+
+
+def create_nodes_blueprint(storage: Storage, scheduler: StatusScheduler) -> Blueprint:
+    bp = Blueprint("nodes", __name__)
+    logger = logging.getLogger("argus.master.api")
+
+    def _json_error(message: str, status: HTTPStatus, code: str) -> Any:
+        response = jsonify({"error": message, "code": code})
+        response.status_code = status
+        return response
+
+    @bp.errorhandler(ValidationError)
+    def _handle_validation_error(err: ValidationError):
+        return _json_error(str(err), HTTPStatus.BAD_REQUEST, "invalid_request")
+
+    @bp.get("/nodes")
+    def list_nodes():
+        nodes = storage.list_nodes()
+        return jsonify(nodes)
+
+    @bp.get("/nodes/<node_id>")
+    def get_node(node_id: str):
+        node = storage.get_node(node_id)
+        if node is None:
+            return _json_error("Node not found", HTTPStatus.NOT_FOUND, "not_found")
+        return jsonify(node)
+
+    @bp.post("/nodes")
+    def register_node():
+        payload = _get_json()
+        data = validate_registration_payload(payload)
+        now = utcnow()
+        now_iso = to_iso(now)
+        node_id = data["id"]
+        name = data["name"]
+        node_type = data["type"]
+        version = data["version"]
+        meta = data["meta_data"]
+
+        if node_id:
+            # 携带 id 说明是重注册，需要校验名称一致性
+            existing_row = storage.get_node_raw(node_id)
+            if existing_row is None:
+                return _json_error("Node not found", HTTPStatus.NOT_FOUND, "not_found")
+            if existing_row["name"] != name:
+                return _json_error(
+                    "Node id and name mismatch during re-registration",
+                    HTTPStatus.INTERNAL_SERVER_ERROR,
+                    "id_name_mismatch",
+                )
+            updated = storage.update_node_meta(
+                node_id,
+                node_type=node_type,
+                version=version,
+                meta_data=meta,
+                last_updated_iso=now_iso,
+            )
+            scheduler.trigger_nodes_json_refresh()
+            return jsonify(updated), HTTPStatus.OK
+
+        # No id provided → search by name
+        existing_by_name = storage.get_node_by_name(name)
+        if existing_by_name:
+            # 同名节点已存在，视为无 id 重注册
+            updated = storage.update_node_meta(
+                existing_by_name["id"],
+                node_type=node_type,
+                version=version,
+                meta_data=meta,
+                last_updated_iso=now_iso,
+            )
+            scheduler.trigger_nodes_json_refresh()
+            return jsonify(updated), HTTPStatus.OK
+
+        new_id = storage.allocate_node_id()
+        created = storage.create_node(
+            new_id,
+            name,
+            node_type,
+            version,
+            meta,
+            status="initialized",
+            register_time_iso=now_iso,
+            last_updated_iso=now_iso,
+        )
+        scheduler.trigger_nodes_json_refresh()
+        return jsonify(created), HTTPStatus.CREATED
+
+    @bp.put("/nodes/<node_id>/config")
+    def update_node_config(node_id: str):
+        payload = _get_json()
+        updates = validate_config_payload(payload)
+        try:
+            updated = storage.update_config_and_labels(
+                node_id,
+                config=updates.get("config"),
+                labels=updates.get("label"),
+            )
+        except KeyError:
+            return _json_error("Node not found", HTTPStatus.NOT_FOUND, "not_found")
+
+        if "label" in updates:
+            scheduler.trigger_nodes_json_refresh()
+        return jsonify(updated)
+
+    @bp.get("/nodes/statistics")
+    def node_statistics():
+        stats = storage.get_statistics()
+        return jsonify(stats)
+
+    @bp.put("/nodes/<node_id>/status")
+    def update_status(node_id: str):
+        payload = _get_json()
+        data = validate_status_payload(payload)
+        try:
+            # master 负责写入 last_report，状态由调度器计算
+            updated = storage.update_last_report(
+                node_id,
+                server_timestamp_iso=to_iso(utcnow()),
+                agent_timestamp_iso=data["timestamp"],
+                health=data["health"],
+            )
+        except KeyError:
+            return _json_error("Node not found", HTTPStatus.NOT_FOUND, "not_found")
+
+        scheduler.trigger_nodes_json_refresh()
+        return jsonify(updated)
+
+    return bp
+
+
+def _get_json() -> Mapping[str, Any]:
+    data = request.get_json(silent=True)
+    if data is None:
+        raise ValidationError("Request body must be valid JSON")
+    if not isinstance(data, Mapping):
+        raise ValidationError("Request body must be a JSON object")
+    return data

src/master/app/routes.py

@@ -0,0 +1,24 @@
+from __future__ import annotations
+
+from flask import Flask, jsonify
+
+from .config import AppConfig
+from .nodes_api import create_nodes_blueprint
+from .scheduler import StatusScheduler
+from .storage import Storage
+
+
+def register_routes(app: Flask, storage: Storage, scheduler: StatusScheduler, config: AppConfig) -> None:
+    app.register_blueprint(create_nodes_blueprint(storage, scheduler), url_prefix="/api/v1/master")
+
+    @app.get("/healthz")
+    def healthz():
+        return jsonify({"status": "ok"})
+
+    @app.get("/readyz")
+    def readyz():
+        try:
+            storage.list_nodes()  # simple readiness probe
+        except Exception as exc:  # pragma: no cover - defensive
+            return jsonify({"status": "error", "error": str(exc)}), 500
+        return jsonify({"status": "ok"})

src/master/app/scheduler.py

@@ -0,0 +1,90 @@
+from __future__ import annotations
+
+import logging
+import threading
+from typing import Optional
+
+from .config import AppConfig
+from .storage import Storage
+from .util import atomic_write_json, parse_iso, to_iso, utcnow
+
+
+class StatusScheduler:
+    def __init__(self, storage: Storage, config: AppConfig, logger: Optional[logging.Logger] = None) -> None:
+        self._storage = storage
+        self._config = config
+        self._logger = logger or logging.getLogger("argus.master.scheduler")
+        self._stop_event = threading.Event()
+        self._thread = threading.Thread(target=self._run, name="status-scheduler", daemon=True)
+        self._nodes_json_lock = threading.Lock()
+        self._pending_nodes_json = threading.Event()
+
+    def start(self) -> None:
+        """启动后台线程，定期刷新节点状态与 nodes.json。"""
+        if not self._thread.is_alive():
+            self._logger.info("Starting scheduler thread")
+            self._thread.start()
+
+    def stop(self) -> None:
+        self._stop_event.set()
+        self._pending_nodes_json.set()
+        self._thread.join(timeout=5)
+
+    def trigger_nodes_json_refresh(self) -> None:
+        self._pending_nodes_json.set()
+
+    def generate_nodes_json(self) -> None:
+        with self._nodes_json_lock:
+            online_nodes = self._storage.get_online_nodes()
+            atomic_write_json(self._config.metric_nodes_json_path, online_nodes)
+            self._logger.info("nodes.json updated", extra={"count": len(online_nodes)})
+
+    # ------------------------------------------------------------------
+    # internal loop
+    # ------------------------------------------------------------------
+
+    def _run(self) -> None:
+        # 确保启动时 nodes.json 会立即生成
+        self._pending_nodes_json.set()
+        while not self._stop_event.is_set():
+            changed = self._reconcile_statuses()
+            if changed or self._pending_nodes_json.is_set():
+                try:
+                    self.generate_nodes_json()
+                finally:
+                    self._pending_nodes_json.clear()
+            self._stop_event.wait(self._config.scheduler_interval_seconds)
+
+    def _reconcile_statuses(self) -> bool:
+        """根据 last_report 与当前时间对比，决定是否切换状态。"""
+        any_status_changed = False
+        now = utcnow()
+        rows = self._storage.fetch_nodes_for_scheduler()
+        for row in rows:
+            node_id = row["id"]
+            last_report_iso = row["last_report"]
+            current_status = row["status"]
+            last_report_dt = parse_iso(last_report_iso)
+            if last_report_dt is None:
+                # No report yet; treat as initialized until report arrives
+                continue
+            delta_seconds = (now - last_report_dt).total_seconds()
+            new_status = current_status
+            if delta_seconds > self._config.offline_threshold_seconds:
+                new_status = "offline"
+            elif delta_seconds <= self._config.online_threshold_seconds:
+                new_status = "online"
+            # Between thresholds: keep current status (sticky)
+            if new_status != current_status:
+                any_status_changed = True
+                self._logger.info(
+                    "Updating node status",
+                    extra={
+                        "node_id": node_id,
+                        "previous": current_status,
+                        "new": new_status,
+                        "delta_seconds": delta_seconds,
+                    },
+                )
+                self._storage.update_status(node_id, new_status, last_updated_iso=to_iso(now))
+        return any_status_changed

src/master/app/storage.py

@@ -0,0 +1,332 @@
+from __future__ import annotations
+
+import json
+import sqlite3
+import threading
+from typing import Any, Dict, Iterable, List, Mapping, Optional, Tuple
+
+from .models import serialize_node_row, serialize_node_summary
+from .util import ensure_parent, to_iso, utcnow
+
+
+class Storage:
+    def __init__(self, db_path: str, node_id_prefix: str) -> None:
+        self._db_path = db_path
+        self._node_id_prefix = node_id_prefix
+        ensure_parent(db_path)
+        self._lock = threading.Lock()
+        self._conn = sqlite3.connect(db_path, detect_types=sqlite3.PARSE_DECLTYPES, check_same_thread=False)
+        self._conn.row_factory = sqlite3.Row
+        with self._lock:
+            self._conn.execute("PRAGMA foreign_keys = ON;")
+        self._ensure_schema()
+
+    # ------------------------------------------------------------------
+    # schema & helpers
+    # ------------------------------------------------------------------
+
+    def _ensure_schema(self) -> None:
+        """初始化表结构，确保服务启动时数据库结构就绪。"""
+        with self._lock:
+            self._conn.executescript(
+                """
+                CREATE TABLE IF NOT EXISTS nodes (
+                    id TEXT PRIMARY KEY,
+                    name TEXT NOT NULL UNIQUE,
+                    type TEXT NOT NULL,
+                    version TEXT,
+                    status TEXT NOT NULL,
+                    config_json TEXT,
+                    labels_json TEXT,
+                    meta_json TEXT,
+                    health_json TEXT,
+                    register_time TEXT,
+                    last_report TEXT,
+                    agent_last_report TEXT,
+                    last_updated TEXT
+                );
+
+                CREATE TABLE IF NOT EXISTS kv (
+                    key TEXT PRIMARY KEY,
+                    value TEXT NOT NULL
+                );
+
+                CREATE INDEX IF NOT EXISTS idx_nodes_status ON nodes(status);
+                CREATE INDEX IF NOT EXISTS idx_nodes_name ON nodes(name);
+                """
+            )
+            self._conn.commit()
+
+    def close(self) -> None:
+        with self._lock:
+            self._conn.close()
+
+    # ------------------------------------------------------------------
+    # Node ID allocation
+    # ------------------------------------------------------------------
+
+    def allocate_node_id(self) -> str:
+        """在 kv 表里维护自增序列，为新节点生成形如 A1 的 ID。"""
+        with self._lock:
+            cur = self._conn.execute("SELECT value FROM kv WHERE key = ?", ("node_id_seq",))
+            row = cur.fetchone()
+            if row is None:
+                next_id = 1
+                self._conn.execute("INSERT INTO kv(key, value) VALUES(?, ?)", ("node_id_seq", str(next_id)))
+            else:
+                next_id = int(row["value"]) + 1
+                self._conn.execute("UPDATE kv SET value = ? WHERE key = ?", (str(next_id), "node_id_seq"))
+            self._conn.commit()
+        return f"{self._node_id_prefix}{next_id}"
+
+    # ------------------------------------------------------------------
+    # Query helpers
+    # ------------------------------------------------------------------
+
+    def list_nodes(self) -> List[Dict[str, Any]]:
+        with self._lock:
+            cur = self._conn.execute(
+                "SELECT id, name, status, type, version FROM nodes ORDER BY id ASC"
+            )
+            rows = cur.fetchall()
+        return [serialize_node_summary(row) for row in rows]
+
+    def get_node(self, node_id: str) -> Optional[Dict[str, Any]]:
+        with self._lock:
+            cur = self._conn.execute("SELECT * FROM nodes WHERE id = ?", (node_id,))
+            row = cur.fetchone()
+        if row is None:
+            return None
+        return serialize_node_row(row)
+
+    def get_node_raw(self, node_id: str) -> Optional[sqlite3.Row]:
+        with self._lock:
+            cur = self._conn.execute("SELECT * FROM nodes WHERE id = ?", (node_id,))
+            row = cur.fetchone()
+        return row
+
+    def get_node_by_name(self, name: str) -> Optional[Dict[str, Any]]:
+        with self._lock:
+            cur = self._conn.execute("SELECT * FROM nodes WHERE name = ?", (name,))
+            row = cur.fetchone()
+        if row is None:
+            return None
+        return serialize_node_row(row)
+
+    # ------------------------------------------------------------------
+    # Mutation helpers
+    # ------------------------------------------------------------------
+
+    def create_node(
+        self,
+        node_id: str,
+        name: str,
+        node_type: str,
+        version: str | None,
+        meta_data: Mapping[str, Any],
+        status: str,
+        register_time_iso: str,
+        last_updated_iso: str,
+    ) -> Dict[str, Any]:
+        """插入节点初始记录，默认 config/label/health 为空。"""
+        now_iso = last_updated_iso
+        with self._lock:
+            self._conn.execute(
+                """
+                INSERT INTO nodes (
+                    id, name, type, version, status, config_json, labels_json, meta_json,
+                    health_json, register_time, last_report, agent_last_report, last_updated
+                ) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+                """,
+                (
+                    node_id,
+                    name,
+                    node_type,
+                    version,
+                    status,
+                    json.dumps({}),
+                    json.dumps([]),
+                    json.dumps(dict(meta_data)),
+                    json.dumps({}),
+                    register_time_iso,
+                    None,
+                    None,
+                    now_iso,
+                ),
+            )
+            self._conn.commit()
+
+        created = self.get_node(node_id)
+        if created is None:
+            raise RuntimeError("Failed to read back created node")
+        return created
+
+    def update_node_meta(
+        self,
+        node_id: str,
+        *,
+        name: Optional[str] = None,
+        node_type: Optional[str] = None,
+        version: Optional[str | None] = None,
+        meta_data: Optional[Mapping[str, Any]] = None,
+        last_updated_iso: Optional[str] = None,
+    ) -> Dict[str, Any]:
+        """重注册时更新节点静态信息，缺省字段保持不变。"""
+        updates: List[str] = []
+        params: List[Any] = []
+        if name is not None:
+            updates.append("name = ?")
+            params.append(name)
+        if node_type is not None:
+            updates.append("type = ?")
+            params.append(node_type)
+        if version is not None:
+            updates.append("version = ?")
+            params.append(version)
+        if meta_data is not None:
+            updates.append("meta_json = ?")
+            params.append(json.dumps(dict(meta_data)))
+        if last_updated_iso is not None:
+            updates.append("last_updated = ?")
+            params.append(last_updated_iso)
+
+        if not updates:
+            result = self.get_node(node_id)
+            if result is None:
+                raise KeyError(node_id)
+            return result
+
+        params.append(node_id)
+        with self._lock:
+            self._conn.execute(
+                f"UPDATE nodes SET {', '.join(updates)} WHERE id = ?",
+                tuple(params),
+            )
+            self._conn.commit()
+        updated = self.get_node(node_id)
+        if updated is None:
+            raise KeyError(node_id)
+        return updated
+
+    def update_config_and_labels(
+        self, node_id: str, *, config: Optional[Mapping[str, Any]] = None, labels: Optional[Iterable[str]] = None
+    ) -> Dict[str, Any]:
+        """部分更新 config/label，并刷新 last_updated 时间戳。"""
+        updates: List[str] = []
+        params: List[Any] = []
+        if config is not None:
+            updates.append("config_json = ?")
+            params.append(json.dumps(dict(config)))
+        if labels is not None:
+            updates.append("labels_json = ?")
+            params.append(json.dumps(list(labels)))
+        updates.append("last_updated = ?")
+        params.append(to_iso(utcnow()))
+        params.append(node_id)
+        with self._lock:
+            self._conn.execute(
+                f"UPDATE nodes SET {', '.join(updates)} WHERE id = ?",
+                tuple(params),
+            )
+            if self._conn.total_changes == 0:
+                self._conn.rollback()
+                raise KeyError(node_id)
+            self._conn.commit()
+        updated = self.get_node(node_id)
+        if updated is None:
+            raise KeyError(node_id)
+        return updated
+
+    def update_last_report(
+        self,
+        node_id: str,
+        *,
+        server_timestamp_iso: str,
+        agent_timestamp_iso: str,
+        health: Mapping[str, Any],
+    ) -> Dict[str, Any]:
+        """记录最新上报时间和健康信息，用于后续状态计算。"""
+        with self._lock:
+            self._conn.execute(
+                """
+                UPDATE nodes
+                SET last_report = ?,
+                    agent_last_report = ?,
+                    health_json = ?,
+                    last_updated = ?
+                WHERE id = ?
+                """,
+                (
+                    server_timestamp_iso,
+                    agent_timestamp_iso,
+                    json.dumps(health),
+                    server_timestamp_iso,
+                    node_id,
+                ),
+            )
+            if self._conn.total_changes == 0:
+                self._conn.rollback()
+                raise KeyError(node_id)
+            self._conn.commit()
+        updated = self.get_node(node_id)
+        if updated is None:
+            raise KeyError(node_id)
+        return updated
+
+    def update_status(self, node_id: str, status: str, *, last_updated_iso: str) -> None:
+        with self._lock:
+            self._conn.execute(
+                "UPDATE nodes SET status = ?, last_updated = ? WHERE id = ?",
+                (status, last_updated_iso, node_id),
+            )
+            self._conn.commit()
+
+    # ------------------------------------------------------------------
+    # Reporting helpers
+    # ------------------------------------------------------------------
+
+    def get_statistics(self) -> Dict[str, Any]:
+        """统计节点总数及按状态聚合的数量。"""
+        with self._lock:
+            cur = self._conn.execute("SELECT COUNT(*) AS total FROM nodes")
+            total_row = cur.fetchone()
+            cur = self._conn.execute("SELECT status, COUNT(*) AS count FROM nodes GROUP BY status")
+            status_rows = cur.fetchall()
+        return {
+            "total": total_row["total"] if total_row else 0,
+            "status_statistics": [
+                {"status": row["status"], "count": row["count"]}
+                for row in status_rows
+            ],
+        }
+
+    def fetch_nodes_for_scheduler(self) -> List[sqlite3.Row]:
+        with self._lock:
+            cur = self._conn.execute(
+                "SELECT id, last_report, status FROM nodes"
+            )
+            return cur.fetchall()
+
+    def get_online_nodes(self) -> List[Dict[str, Any]]:
+        """返回在线节点列表，用于生成 nodes.json。"""
+        with self._lock:
+            cur = self._conn.execute(
+                "SELECT id, meta_json, labels_json, name FROM nodes WHERE status = ? ORDER BY id ASC",
+                ("online",),
+            )
+            rows = cur.fetchall()
+
+        result: List[Dict[str, Any]] = []
+        for row in rows:
+            meta = json.loads(row["meta_json"]) if row["meta_json"] else {}
+            labels = json.loads(row["labels_json"]) if row["labels_json"] else []
+            result.append(
+                {
+                    "node_id": row["id"],
+                    "user_id": meta.get("user"),
+                    "ip": meta.get("ip"),
+                    "hostname": meta.get("hostname", row["name"]),
+                    "labels": labels if isinstance(labels, list) else [],
+                }
+            )
+        return result

src/master/app/util.py

@@ -0,0 +1,51 @@
+from __future__ import annotations
+
+import json
+import os
+import tempfile
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any, Iterable
+
+
+ISO_FORMAT = "%Y-%m-%dT%H:%M:%SZ"
+
+
+def utcnow() -> datetime:
+    """获取当前 UTC 时间，统一时间基准。"""
+    return datetime.now(timezone.utc)
+
+
+def to_iso(dt: datetime | None) -> str | None:
+    if dt is None:
+        return None
+    return dt.astimezone(timezone.utc).replace(microsecond=0).strftime(ISO_FORMAT)
+
+
+def parse_iso(value: str | None) -> datetime | None:
+    if not value:
+        return None
+    try:
+        if value.endswith("Z"):
+            return datetime.strptime(value, ISO_FORMAT).replace(tzinfo=timezone.utc)
+        # Fallback for ISO strings with offset
+        return datetime.fromisoformat(value).astimezone(timezone.utc)
+    except ValueError:
+        return None
+
+
+def ensure_parent(path: str) -> None:
+    """确保目标文件所在目录存在。"""
+    Path(path).parent.mkdir(parents=True, exist_ok=True)
+
+
+def atomic_write_json(path: str, data: Iterable[Any] | Any) -> None:
+    """原子化写 JSON，避免被其它进程读到半成品。"""
+    ensure_parent(path)
+    directory = Path(path).parent
+    with tempfile.NamedTemporaryFile("w", dir=directory, delete=False) as tmp:
+        json.dump(data, tmp, separators=(",", ":"))
+        tmp.flush()
+        os.fsync(tmp.fileno())
+        temp_path = tmp.name
+    os.replace(temp_path, path)

src/master/requirements.txt

@@ -0,0 +1,2 @@
+Flask==2.3.3
+gunicorn==21.2.0

src/master/scripts/build_images.sh

@@ -0,0 +1,63 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+usage() {
+  cat >&2 <<'USAGE'
+Usage: $0 [--intranet] [--offline] [--tag <image_tag>]
+
+Options:
+  --intranet           使用指定的 PyPI 镜像源（默认清华镜像）。
+  --offline            完全离线构建，依赖 offline_wheels/ 目录中的离线依赖包。
+  --tag <image_tag>    自定义镜像标签，默认 argus-master:dev。
+USAGE
+}
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+IMAGE_TAG="${IMAGE_TAG:-argus-master:dev}"
+BUILD_ARGS=()
+OFFLINE_MODE=0
+
+while [[ "$#" -gt 0 ]]; do
+  case "$1" in
+    --intranet)
+      INTRANET_INDEX="${INTRANET_INDEX:-https://pypi.tuna.tsinghua.edu.cn/simple}"
+      BUILD_ARGS+=("--build-arg" "PIP_INDEX_URL=${INTRANET_INDEX}")
+      shift
+      ;;
+    --offline)
+      OFFLINE_MODE=1
+      BUILD_ARGS+=("--build-arg" "USE_OFFLINE=1")
+      shift
+      ;;
+    --tag)
+      [[ $# -ge 2 ]] || { usage; exit 1; }
+      IMAGE_TAG="$2"
+      shift 2
+      ;;
+    -h|--help)
+      usage
+      exit 0
+      ;;
+    *)
+      echo "Unknown option: $1" >&2
+      usage
+      exit 1
+      ;;
+  esac
+ done
+
+if [[ "$OFFLINE_MODE" -eq 1 ]]; then
+  WHEELS_DIR="$PROJECT_ROOT/offline_wheels"
+  if [[ ! -d "$WHEELS_DIR" ]]; then
+    echo "[ERROR] offline_wheels 目录不存在: $WHEELS_DIR" >&2
+    exit 1
+  fi
+  if ! find "$WHEELS_DIR" -maxdepth 1 -type f -name '*.whl' -print -quit >/dev/null; then
+    echo "[WARN] offline_wheels 目录为空，请确保已提前下载所需的 wheel 包" >&2
+  fi
+fi
+
+echo "[INFO] Building image $IMAGE_TAG"
+docker build "${BUILD_ARGS[@]}" -t "$IMAGE_TAG" "$PROJECT_ROOT"
+echo "[OK] Image $IMAGE_TAG built"

src/master/scripts/load_images.sh

@@ -0,0 +1,39 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+usage() {
+  echo "Usage: $0 [--file <tar_path>]" >&2
+}
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+DEFAULT_INPUT="$PROJECT_ROOT/images/argus-master-dev.tar"
+IMAGE_TAR="$DEFAULT_INPUT"
+
+while [[ "$#" -gt 0 ]]; do
+  case "$1" in
+    --file)
+      [[ $# -ge 2 ]] || { usage; exit 1; }
+      IMAGE_TAR="$2"
+      shift 2
+      ;;
+    -h|--help)
+      usage
+      exit 0
+      ;;
+    *)
+      echo "Unknown option: $1" >&2
+      usage
+      exit 1
+      ;;
+  esac
+ done
+
+if [[ ! -f "$IMAGE_TAR" ]]; then
+  echo "[ERROR] Image tarball not found: $IMAGE_TAR" >&2
+  exit 1
+fi
+
+echo "[INFO] Loading image from $IMAGE_TAR"
+docker image load -i "$IMAGE_TAR"
+echo "[OK] Image loaded"

src/master/scripts/prepare_offline_wheels.sh

@@ -0,0 +1,97 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+usage() {
+  cat >&2 <<'USAGE'
+Usage: $0 [--pip-version <version>] [--clean] [--local]
+
+Options:
+  --pip-version <version>   额外下载指定版本的 pip wheel（例如 25.2）。
+  --clean                   清理 offline_wheels/*.whl 后重新下载。
+  --local                   使用本地 python 执行下载（默认通过 docker python:3.11-slim）。
+USAGE
+}
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+REQUIREMENTS_FILE="$PROJECT_ROOT/requirements.txt"
+WHEEL_DIR="$PROJECT_ROOT/offline_wheels"
+PIP_VERSION=""
+CLEAN=0
+USE_LOCAL=0
+
+while [[ $# -gt 0 ]]; do
+  case "$1" in
+    --pip-version)
+      [[ $# -ge 2 ]] || { usage; exit 1; }
+      PIP_VERSION="$2"
+      shift 2
+      ;;
+    --clean)
+      CLEAN=1
+      shift
+      ;;
+    --local)
+      USE_LOCAL=1
+      shift
+      ;;
+    -h|--help)
+      usage
+      exit 0
+      ;;
+    *)
+      echo "Unknown option: $1" >&2
+      usage
+      exit 1
+      ;;
+  esac
+ done
+
+if [[ ! -f "$REQUIREMENTS_FILE" ]]; then
+  echo "[ERROR] requirements.txt not found at $REQUIREMENTS_FILE" >&2
+  exit 1
+fi
+
+mkdir -p "$WHEEL_DIR"
+
+if [[ "$CLEAN" -eq 1 ]]; then
+  echo "[INFO] Cleaning existing wheels in $WHEEL_DIR"
+  find "$WHEEL_DIR" -maxdepth 1 -type f -name '*.whl' -delete
+fi
+
+run_with_python() {
+  local cmd=("python" "-m" "pip" "$@")
+  eval "${cmd[@]}"
+}
+
+if [[ "$USE_LOCAL" -eq 1 ]]; then
+  PYTHON_BIN=${PYTHON_BIN:-python3}
+  if ! command -v "$PYTHON_BIN" >/dev/null 2>&1; then
+    echo "[ERROR] $PYTHON_BIN not found" >&2
+    exit 1
+  fi
+  echo "[INFO] Using local python ($PYTHON_BIN) to download wheels"
+  "$PYTHON_BIN" -m pip download -r "$REQUIREMENTS_FILE" -d "$WHEEL_DIR"
+  if [[ -n "$PIP_VERSION" ]]; then
+    "$PYTHON_BIN" -m pip download "pip==${PIP_VERSION}" -d "$WHEEL_DIR"
+  fi
+else
+  if ! command -v docker >/dev/null 2>&1; then
+    echo "[ERROR] docker not found; rerun with --local or安装 docker" >&2
+    exit 1
+  fi
+  echo "[INFO] Using docker image python:3.11-slim 下载 wheel"
+  docker run --rm \
+    -v "$WHEEL_DIR":/wheels \
+    -v "$REQUIREMENTS_FILE":/tmp/requirements.txt \
+    python:3.11-slim \
+    bash -c "set -euo pipefail && python -m pip install --upgrade pip && python -m pip download -r /tmp/requirements.txt -d /wheels"
+  if [[ -n "$PIP_VERSION" ]]; then
+    docker run --rm \
+      -v "$WHEEL_DIR":/wheels \
+      python:3.11-slim \
+      bash -c "set -euo pipefail && python -m pip download pip==${PIP_VERSION} -d /wheels"
+  fi
+fi
+
+echo "[INFO] Offline wheels prepared at $WHEEL_DIR"

src/master/scripts/save_images.sh

@@ -0,0 +1,41 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+usage() {
+  echo "Usage: $0 [--tag <image_tag>] [--output <tar_path>]" >&2
+}
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+DEFAULT_OUTPUT="$PROJECT_ROOT/images/argus-master-dev.tar"
+IMAGE_TAG="${IMAGE_TAG:-argus-master:dev}"
+OUTPUT_PATH="$DEFAULT_OUTPUT"
+
+while [[ "$#" -gt 0 ]]; do
+  case "$1" in
+    --tag)
+      [[ $# -ge 2 ]] || { usage; exit 1; }
+      IMAGE_TAG="$2"
+      shift 2
+      ;;
+    --output)
+      [[ $# -ge 2 ]] || { usage; exit 1; }
+      OUTPUT_PATH="$2"
+      shift 2
+      ;;
+    -h|--help)
+      usage
+      exit 0
+      ;;
+    *)
+      echo "Unknown option: $1" >&2
+      usage
+      exit 1
+      ;;
+  esac
+ done
+
+mkdir -p "$(dirname "$OUTPUT_PATH")"
+echo "[INFO] Saving image $IMAGE_TAG to $OUTPUT_PATH"
+docker image save "$IMAGE_TAG" -o "$OUTPUT_PATH"
+echo "[OK] Image saved"

src/master/tests/.gitignore

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

src/master/tests/docker-compose.yml

@@ -0,0 +1,18 @@
+services:
+  master:
+    image: ${MASTER_IMAGE_TAG:-argus-master:dev}
+    container_name: argus-master-e2e
+    environment:
+      - OFFLINE_THRESHOLD_SECONDS=6
+      - ONLINE_THRESHOLD_SECONDS=2
+      - SCHEDULER_INTERVAL_SECONDS=1
+    ports:
+      - "31300:3000"
+    volumes:
+      - ./private/argus/master:/private/argus/master
+      - ./private/argus/metric/prometheus:/private/argus/metric/prometheus
+    restart: unless-stopped
+
+networks:
+  default:
+    driver: bridge

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

@@ -0,0 +1,25 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+SCRIPTS=(
+  "01_up_master.sh"
+  "02_verify_ready_and_nodes_json.sh"
+  "03_register_via_curl.sh"
+  "04_reregister_and_error_cases.sh"
+  "05_status_report_via_curl.sh"
+  "06_config_update_and_nodes_json.sh"
+  "07_stats_single_node.sh"
+  "08_multi_node_stats.sh"
+  "09_restart_persistence.sh"
+  "10_down.sh"
+)
+
+for script in "${SCRIPTS[@]}"; do
+  echo "[TEST] Running $script"
+  MASTER_IMAGE_TAG="${MASTER_IMAGE_TAG:-argus-master:dev}" "$SCRIPT_DIR/$script"
+  echo "[TEST] $script completed"
+  echo
+done
+
+echo "[TEST] Master module E2E tests completed"

src/master/tests/scripts/00_e2e_test_offline.sh

@@ -0,0 +1,16 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+MODULE_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+MASTER_ROOT="$(cd "$MODULE_ROOT/.." && pwd)"
+
+# 准备离线依赖并构建镜像
+pushd "$MASTER_ROOT" >/dev/null
+./scripts/prepare_offline_wheels.sh --clean --pip-version 25.2
+./scripts/build_images.sh --offline --tag argus-master:offline
+popd >/dev/null
+
+# 使用离线镜像执行既有端到端用例
+MASTER_IMAGE_TAG="argus-master:offline" ./scripts/00_e2e_test.sh
+

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

@@ -0,0 +1,37 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+MODULE_ROOT="$(cd "$TEST_ROOT/.." && pwd)"
+PRIVATE_ROOT="$TEST_ROOT/private"
+TMP_ROOT="$TEST_ROOT/tmp"
+
+mkdir -p "$PRIVATE_ROOT/argus/master"
+mkdir -p "$PRIVATE_ROOT/argus/metric/prometheus"
+mkdir -p "$TMP_ROOT"
+
+# 确保上一次运行留下的容器/数据被清理
+compose() {
+  if docker compose version >/dev/null 2>&1; then
+    docker compose "$@"
+  else
+    docker-compose "$@"
+  fi
+}
+
+pushd "$TEST_ROOT" >/dev/null
+compose down --remove-orphans || true
+popd >/dev/null
+
+rm -rf "$TMP_ROOT" "$PRIVATE_ROOT"
+mkdir -p "$PRIVATE_ROOT/argus/master"
+mkdir -p "$PRIVATE_ROOT/argus/metric/prometheus"
+mkdir -p "$TMP_ROOT"
+
+pushd "$TEST_ROOT" >/dev/null
+compose down --remove-orphans || true
+MASTER_IMAGE_TAG="${MASTER_IMAGE_TAG:-argus-master:dev}" compose up -d
+popd >/dev/null
+
+echo "[INFO] Master container is up on http://localhost:31300"

src/master/tests/scripts/02_verify_ready_and_nodes_json.sh

@@ -0,0 +1,51 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+PRIVATE_ROOT="$TEST_ROOT/private"
+API_BASE="http://localhost:31300"
+NODES_JSON_PATH="$PRIVATE_ROOT/argus/metric/prometheus/nodes.json"
+
+# 等待 readyz 返回 200，确保数据库初始化完成
+for _ in {1..30}; do
+  status=$(curl -s -o /dev/null -w '%{http_code}' "$API_BASE/readyz" || true)
+  if [[ "$status" == "200" ]]; then
+    break
+  fi
+  sleep 1
+ done
+
+if [[ "${status:-}" != "200" ]]; then
+  echo "[ERROR] /readyz 未在预期时间内返回 200，实际=$status" >&2
+  exit 1
+fi
+
+echo "[INFO] /readyz 已通过，就绪检查成功"
+
+# scheduler 启动时会产生空的 nodes.json，这里等待文件出现并校验内容
+for _ in {1..30}; do
+  if [[ -f "$NODES_JSON_PATH" ]]; then
+    break
+  fi
+  sleep 1
+ done
+
+if [[ ! -f "$NODES_JSON_PATH" ]]; then
+  echo "[ERROR] 未在预期时间内生成 $NODES_JSON_PATH" >&2
+  exit 1
+fi
+
+if ! python3 - "$NODES_JSON_PATH" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    data = json.load(handle)
+if data != []:
+    raise SystemExit(f"nodes.json initial content should be [], got {data}")
+PY
+then
+  echo "[ERROR] nodes.json 初始内容不是空数组" >&2
+  exit 1
+fi
+
+echo "[INFO] nodes.json 初始状态校验通过"

src/master/tests/scripts/03_register_via_curl.sh

@@ -0,0 +1,68 @@
+#!/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:31300/api/v1/master"
+
+mkdir -p "$TMP_ROOT"
+
+for _ in {1..30}; do
+  if curl -sf "$API_BASE/healthz" >/dev/null; then
+    break
+  fi
+  sleep 1
+done
+
+payload=$(cat <<'JSON'
+{
+  "name": "dev-testuser-testinst-pod-0",
+  "type": "agent",
+  "meta_data": {
+    "hostname": "dev-testuser-testinst-pod-0",
+    "ip": "10.0.0.10",
+    "env": "dev",
+    "user": "testuser",
+    "instance": "testinst",
+    "cpu_number": 4,
+    "memory_in_bytes": 2147483648,
+    "gpu_number": 0
+  },
+  "version": "1.1.0"
+}
+JSON
+)
+
+body_file="$TMP_ROOT/register_body.json"
+status=$(curl -sS -o "$body_file" -w '%{http_code}' -H 'Content-Type: application/json' -X POST "$API_BASE/nodes" -d "$payload")
+body="$(cat "$body_file")"
+
+if [[ "$status" != "201" ]]; then
+  echo "[ERROR] Unexpected status code: $status" >&2
+  echo "$body" >&2
+  exit 1
+fi
+
+node_id=$(python3 - "$body_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    body = json.load(handle)
+print(body["id"])
+PY
+)
+
+echo "$body" > "$TMP_ROOT/last_response.json"
+echo "$node_id" > "$TMP_ROOT/node_id"
+
+list_file="$TMP_ROOT/nodes_list.json"
+curl -sS "$API_BASE/nodes" -o "$list_file"
+python3 - "$list_file" "$node_id" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    data = json.load(handle)
+node_id = sys.argv[2]
+assert any(item.get("id") == node_id for item in data), "node not in list"
+PY
+
+echo "[INFO] Registered node with id $node_id"

src/master/tests/scripts/04_reregister_and_error_cases.sh

@@ -0,0 +1,116 @@
+#!/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:31300/api/v1/master"
+NODE_ID="$(cat "$TMP_ROOT/node_id")"
+
+# 使用相同 ID 重注册，同时修改部分 meta/version 字段
+payload=$(cat <<JSON
+{
+  "id": "$NODE_ID",
+  "name": "dev-testuser-testinst-pod-0",
+  "type": "agent",
+  "meta_data": {
+    "hostname": "dev-testuser-testinst-pod-0",
+    "ip": "10.0.0.11",
+    "env": "dev",
+    "user": "testuser",
+    "instance": "testinst",
+    "cpu_number": 8,
+    "memory_in_bytes": 2147483648,
+    "gpu_number": 0
+  },
+  "version": "1.2.0"
+}
+JSON
+)
+
+status=$(curl -sS -o "$TMP_ROOT/reregister_response.json" -w '%{http_code}' -H 'Content-Type: application/json' -X POST "$API_BASE/nodes" -d "$payload")
+if [[ "$status" != "200" ]]; then
+  echo "[ERROR] 重注册返回非 200: $status" >&2
+  cat "$TMP_ROOT/reregister_response.json" >&2
+  exit 1
+fi
+
+python3 - "$TMP_ROOT/reregister_response.json" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+assert node["meta_data"]["ip"] == "10.0.0.11", node["meta_data"]
+assert node["meta_data"]["cpu_number"] == 8, node["meta_data"]
+assert node["version"] == "1.2.0", node
+PY
+
+echo "[INFO] 重注册成功，元数据已更新"
+
+# 未知 ID => 404
+unknown_payload=$(cat <<'JSON'
+{
+  "id": "A999",
+  "name": "dev-testuser-testinst-pod-0",
+  "type": "agent",
+  "meta_data": {
+    "hostname": "dev-testuser-testinst-pod-0",
+    "ip": "10.0.0.12",
+    "env": "dev",
+    "user": "testuser",
+    "instance": "testinst",
+    "cpu_number": 4,
+    "memory_in_bytes": 2147483648,
+    "gpu_number": 0
+  },
+  "version": "1.2.0"
+}
+JSON
+)
+
+status=$(curl -sS -o "$TMP_ROOT/unknown_id_response.json" -w '%{http_code}' -H 'Content-Type: application/json' -X POST "$API_BASE/nodes" -d "$unknown_payload")
+if [[ "$status" != "404" ]]; then
+  echo "[ERROR] 未知 ID 应返回 404，实际=$status" >&2
+  cat "$TMP_ROOT/unknown_id_response.json" >&2
+  exit 1
+fi
+
+echo "[INFO] 未知 ID 返回 404 验证通过"
+
+# id 与 name 不匹配 => 500，节点保持原名
+mismatch_payload=$(cat <<JSON
+{
+  "id": "$NODE_ID",
+  "name": "dev-testuser-testinst-pod-0-mismatch",
+  "type": "agent",
+  "meta_data": {
+    "hostname": "dev-testuser-testinst-pod-0-mismatch",
+    "ip": "10.0.0.13",
+    "env": "dev",
+    "user": "testuser",
+    "instance": "testinst",
+    "cpu_number": 4,
+    "memory_in_bytes": 2147483648,
+    "gpu_number": 0
+  },
+  "version": "1.2.0"
+}
+JSON
+)
+
+status=$(curl -sS -o "$TMP_ROOT/mismatch_response.json" -w '%{http_code}' -H 'Content-Type: application/json' -X POST "$API_BASE/nodes" -d "$mismatch_payload")
+if [[ "$status" != "500" ]]; then
+  echo "[ERROR] 名称不匹配应返回 500，实际=$status" >&2
+  cat "$TMP_ROOT/mismatch_response.json" >&2
+  exit 1
+fi
+
+# 验证名称仍保持正确
+curl -sS "$API_BASE/nodes/$NODE_ID" -o "$TMP_ROOT/post_mismatch_detail.json"
+python3 - "$TMP_ROOT/post_mismatch_detail.json" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+assert node["name"] == "dev-testuser-testinst-pod-0", node["name"]
+PY
+
+echo "[INFO] 名称不匹配返回 500，且原始节点未被篡改"

src/master/tests/scripts/05_status_report_via_curl.sh

@@ -0,0 +1,98 @@
+#!/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:31300/api/v1/master"
+
+node_id="$(cat "$TMP_ROOT/node_id")"
+
+payload=$(python3 - <<'PY'
+import json
+from datetime import datetime, timezone
+body = {
+    "timestamp": datetime.now(timezone.utc).replace(microsecond=0).isoformat().replace("+00:00", "Z"),
+    "health": {
+        "log-fluentbit": {"status": "healthy", "timestamp": "2023-10-05T12:05:00Z"},
+        "metric-node-exporter": {"status": "healthy", "timestamp": "2023-10-05T12:05:00Z"}
+    }
+}
+print(json.dumps(body))
+PY
+)
+
+response=$(curl -sS -w '\n%{http_code}' -H 'Content-Type: application/json' -X PUT "$API_BASE/nodes/$node_id/status" -d "$payload")
+body="$(echo "$response" | head -n -1)"
+status="$(echo "$response" | tail -n1)"
+
+if [[ "$status" != "200" ]]; then
+  echo "[ERROR] Status update failed with code $status" >&2
+  echo "$body" >&2
+  exit 1
+fi
+
+echo "$body" > "$TMP_ROOT/last_response.json"
+
+sleep 3
+
+detail_file="$TMP_ROOT/status_detail.json"
+curl -sS "$API_BASE/nodes/$node_id" -o "$detail_file"
+python3 - "$detail_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+assert node["status"] == "online", f"Expected online, got {node['status']}"
+assert "log-fluentbit" in node["health"], node["health"].keys()
+PY
+
+echo "[INFO] Status report successful and node is online"
+
+# 等待超过 offline 阈值，验证会自动转为 offline
+sleep 7
+
+offline_detail_file="$TMP_ROOT/status_offline.json"
+curl -sS "$API_BASE/nodes/$node_id" -o "$offline_detail_file"
+python3 - "$offline_detail_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+assert node["status"] == "offline", f"Expected offline, got {node['status']}"
+PY
+
+echo "[INFO] Node transitioned to offline as expected"
+
+# 再次上报健康，触发状态回到 online
+payload=$(python3 - <<'PY'
+import json
+from datetime import datetime, timezone
+body = {
+    "timestamp": datetime.now(timezone.utc).replace(microsecond=0).isoformat().replace("+00:00", "Z"),
+    "health": {
+        "log-fluentbit": {"status": "healthy", "timestamp": "2023-10-05T12:05:00Z"},
+        "metric-node-exporter": {"status": "healthy", "timestamp": "2023-10-05T12:05:00Z"}
+    }
+}
+print(json.dumps(body))
+PY
+)
+
+curl -sS -o "$TMP_ROOT/second_status_response.json" -w '%{http_code}' -H 'Content-Type: application/json' -X PUT "$API_BASE/nodes/$node_id/status" -d "$payload" > "$TMP_ROOT/second_status_code"
+if [[ $(cat "$TMP_ROOT/second_status_code") != "200" ]]; then
+  echo "[ERROR] Second status update failed" >&2
+  cat "$TMP_ROOT/second_status_response.json" >&2
+  exit 1
+fi
+
+sleep 3
+
+final_detail_file="$TMP_ROOT/status_back_online.json"
+curl -sS "$API_BASE/nodes/$node_id" -o "$final_detail_file"
+python3 - "$final_detail_file" <<'PY'
+import json, sys
+with open(sys.argv[1]) as handle:
+    node = json.load(handle)
+assert node["status"] == "online", f"Expected online after second report, got {node['status']}"
+PY
+
+echo "[INFO] Node transitioned back to online after new status report"