[#2 ] agent端到端测试使用域名完成

[#2 ] master模块使用supervisor守护；增加dns-monitor功能；内网apt源
2025-09-25 07:03:57 +00:00 · 2025-09-25 03:54:47 +00:00
16 changed files with 304 additions and 27 deletions
--- a/src/agent/dist/argus-agent
+++ b/src/agent/dist/argus-agent
--- a/src/agent/tests/docker-compose.yml
+++ b/src/agent/tests/docker-compose.yml
@ -1,7 +1,19 @@
 services:
+  bind:
+    image: ${BIND_IMAGE_TAG:-argus-bind9:e2e}
+    container_name: argus-bind-agent-e2e
+    volumes:
+      - ./private:/private
+    restart: unless-stopped
+    networks:
+      default:
+        ipv4_address: 172.28.0.2
+
  master:
    image: argus-master:dev
    container_name: argus-master-agent-e2e
+    depends_on:
+      - bind
    environment:
      - OFFLINE_THRESHOLD_SECONDS=6
      - ONLINE_THRESHOLD_SECONDS=2
@ -11,6 +23,10 @@ services:
    volumes:
      - ./private/argus/master:/private/argus/master
      - ./private/argus/metric/prometheus:/private/argus/metric/prometheus
+      - ./private/argus/etc:/private/argus/etc
+    networks:
+      default:
+        ipv4_address: 172.28.0.10

  agent:
    image: ubuntu:24.04
@ -18,15 +34,21 @@ services:
    hostname: dev-e2euser-e2einst-pod-0
    depends_on:
      - master
+      - bind
    environment:
-      - MASTER_ENDPOINT=http://master:3000
+      - MASTER_ENDPOINT=http://master.argus.com: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
+      - ./private/argus/etc:/private/argus/etc
      - ../dist/argus-agent:/usr/local/bin/argus-agent:ro
+      - ./scripts/agent_entrypoint.sh:/usr/local/bin/agent-entrypoint.sh:ro
    entrypoint:
-      - /usr/local/bin/argus-agent
+      - /usr/local/bin/agent-entrypoint.sh
+    networks:
+      default:
+        ipv4_address: 172.28.0.20

 networks:
  default:
--- a/src/agent/tests/scripts/01_bootstrap.sh
+++ b/src/agent/tests/scripts/01_bootstrap.sh
@ -5,6 +5,7 @@ SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 AGENT_ROOT="$(cd "$TEST_ROOT/.." && pwd)"
 MASTER_ROOT="$(cd "$AGENT_ROOT/../master" && pwd)"
+REPO_ROOT="$(cd "$AGENT_ROOT/../.." && pwd)"
 PRIVATE_ROOT="$TEST_ROOT/private"
 TMP_ROOT="$TEST_ROOT/tmp"

@ -13,15 +14,27 @@ 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"
+DNS_DIR="$PRIVATE_ROOT/argus/etc"
+BIND_IMAGE_TAG="${BIND_IMAGE_TAG:-argus-bind9:e2e}"
+BIND_ROOT="$(cd "$MASTER_ROOT/../bind" && pwd)"

 mkdir -p "$AGENT_CONFIG_DIR"
 mkdir -p "$AGENT_HEALTH_DIR"
 mkdir -p "$MASTER_PRIVATE_DIR"
 mkdir -p "$METRIC_PRIVATE_DIR"
 mkdir -p "$TMP_ROOT"
+mkdir -p "$DNS_DIR"

 touch "$AGENT_HEALTH_DIR/.keep"

+# 中文提示：准备 bind 模块提供的 update-dns.sh，模拟生产下发
+if [[ -f "$BIND_ROOT/build/update-dns.sh" ]]; then
+  cp "$BIND_ROOT/build/update-dns.sh" "$DNS_DIR/update-dns.sh"
+  chmod +x "$DNS_DIR/update-dns.sh"
+else
+  echo "[WARN] bind update script missing at $BIND_ROOT/build/update-dns.sh"
+fi
+
 pushd "$MASTER_ROOT" >/dev/null
 ./scripts/build_images.sh --tag argus-master:dev
 popd >/dev/null
@ -37,6 +50,12 @@ if [[ ! -x "$AGENT_BINARY" ]]; then
  exit 1
 fi

+# 中文提示：构建测试专用 bind9 镜像，确保解析服务可用
+pushd "$REPO_ROOT" >/dev/null
+docker build -f src/bind/build/Dockerfile -t "$BIND_IMAGE_TAG" .
+popd >/dev/null
+
 echo "$AGENT_BINARY" > "$TMP_ROOT/agent_binary_path"
+echo "$BIND_IMAGE_TAG" > "$TMP_ROOT/bind_image_tag"

 echo "[INFO] Agent E2E bootstrap complete"
--- a/src/agent/tests/scripts/02_up.sh
+++ b/src/agent/tests/scripts/02_up.sh
@ -17,6 +17,11 @@ if [[ ! -x "$AGENT_BINARY" ]]; then
  exit 1
 fi

+BIND_IMAGE_TAG_VALUE="argus-bind9:e2e"
+if [[ -f "$TMP_ROOT/bind_image_tag" ]]; then
+  BIND_IMAGE_TAG_VALUE="$(cat "$TMP_ROOT/bind_image_tag")"
+fi
+
 compose() {
  if docker compose version >/dev/null 2>&1; then
    docker compose "$@"
@ -25,13 +30,13 @@ compose() {
  fi
 }

-docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
+docker container rm -f argus-agent-e2e argus-master-agent-e2e argus-bind-agent-e2e >/dev/null 2>&1 || true

 docker network rm tests_default >/dev/null 2>&1 || true

 pushd "$TEST_ROOT" >/dev/null
 compose down --remove-orphans || true
-compose up -d
+BIND_IMAGE_TAG="$BIND_IMAGE_TAG_VALUE" compose up -d
 popd >/dev/null

 echo "[INFO] Master+Agent stack started"
--- a/src/agent/tests/scripts/06_restart_agent_and_reregister.sh
+++ b/src/agent/tests/scripts/06_restart_agent_and_reregister.sh
@ -9,6 +9,13 @@ NODE_ID="$(cat "$TMP_ROOT/node_id")"
 AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
 NETWORK_NAME="tests_default"
 NEW_AGENT_IP="172.28.0.200"
+ENTRYPOINT_SCRIPT="$SCRIPT_DIR/agent_entrypoint.sh"
+
+# 中文提示：重启场景也需要同样的入口脚本，确保 DNS 注册逻辑一致
+if [[ ! -f "$ENTRYPOINT_SCRIPT" ]]; then
+  echo "[ERROR] agent entrypoint script missing at $ENTRYPOINT_SCRIPT" >&2
+  exit 1
+fi

 if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
  echo "[ERROR] Agent binary path missing; rerun bootstrap" >&2
@ -68,21 +75,17 @@ if ! docker run -d \
  --ip "$NEW_AGENT_IP" \
  -v "$AGENT_DIR:/private/argus/agent/$AGENT_HOSTNAME" \
  -v "$HEALTH_DIR:/private/argus/agent/health/$AGENT_HOSTNAME" \
+  -v "$TEST_ROOT/private/argus/etc:/private/argus/etc" \
  -v "$AGENT_BINARY:/usr/local/bin/argus-agent:ro" \
-  -e MASTER_ENDPOINT=http://master:3000 \
+  -v "$ENTRYPOINT_SCRIPT:/usr/local/bin/agent-entrypoint.sh:ro" \
+  -e MASTER_ENDPOINT=http://master.argus.com:3000 \
  -e REPORT_INTERVAL_SECONDS=2 \
-  ubuntu:24.04 \
-  sleep 300 >/dev/null; then
+  --entrypoint /usr/local/bin/agent-entrypoint.sh \
+  ubuntu:24.04 >/dev/null; then
  echo "[ERROR] Failed to start agent container with custom IP" >&2
  exit 1
 fi

-# 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="$TMP_ROOT/post_restart.json"
 for _ in {1..20}; do
--- a/src/agent/tests/scripts/agent_entrypoint.sh
+++ b/src/agent/tests/scripts/agent_entrypoint.sh
@ -0,0 +1,49 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+LOG_PREFIX="[AGENT-ENTRYPOINT]"
+DNS_SCRIPT="/private/argus/etc/update-dns.sh"
+DNS_CONF="/private/argus/etc/dns.conf"
+TARGET_DOMAIN="master.argus.com"
+
+log() {
+  echo "${LOG_PREFIX} $*"
+}
+
+# 中文提示：等待 bind 下发的 update-dns.sh 脚本
+for _ in {1..30}; do
+  if [[ -x "$DNS_SCRIPT" ]]; then
+    break
+  fi
+  log "等待 update-dns.sh 准备就绪..."
+  sleep 1
+done
+
+if [[ -x "$DNS_SCRIPT" ]]; then
+  log "执行 update-dns.sh 更新容器 DNS"
+  if ! "$DNS_SCRIPT"; then
+    log "update-dns.sh 执行失败，继续尝试默认 DNS"
+  fi
+else
+  log "未获取到 update-dns.sh，使用镜像默认 DNS"
+fi
+
+# 中文提示：记录当前 dns.conf 内容，便于排查
+if [[ -f "$DNS_CONF" ]]; then
+  log "dns.conf 内容: $(tr '\n' ' ' < "$DNS_CONF")"
+else
+  log "dns.conf 暂未生成"
+fi
+
+# 中文提示：尝试解析 master 域名，失败不阻塞但会打日志
+for _ in {1..30}; do
+  if getent hosts "$TARGET_DOMAIN" >/dev/null 2>&1; then
+    MASTER_IP=$(getent hosts "$TARGET_DOMAIN" | awk '{print $1}' | head -n 1)
+    log "master.argus.com 解析成功: $MASTER_IP"
+    break
+  fi
+  sleep 1
+done
+
+log "启动 argus-agent"
+exec /usr/local/bin/argus-agent
--- a/src/master/Dockerfile
+++ b/src/master/Dockerfile
@ -4,29 +4,60 @@ SHELL ["/bin/bash", "-c"]

 ARG PIP_INDEX_URL=
 ARG USE_OFFLINE=0
+ARG USE_INTRANET=false
+
 ENV PIP_NO_CACHE_DIR=1 \
    PYTHONUNBUFFERED=1 \
    PYTHONPATH=/app

+USER root
+
 WORKDIR /app

-COPY requirements.txt ./
-COPY offline_wheels/ /opt/offline_wheels/
+COPY ./src/master/requirements.txt ./requirements.txt
+COPY ./src/master/offline_wheels/ /opt/offline_wheels/

 RUN set -euxo pipefail \
    && if [[ "$USE_OFFLINE" == "1" ]]; then \
+         python -m pip install --no-index --find-links /opt/offline_wheels pip && \
         python -m pip install --no-index --find-links /opt/offline_wheels -r requirements.txt; \
       else \
-         python -m pip install --upgrade pip \
-         && if [[ -n "$PIP_INDEX_URL" ]]; then \
+         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 \
+         else \
              python -m pip install -r requirements.txt; \
-            fi; \
+         fi; \
       fi

-COPY app ./app
+# 配置内网 apt 源并安装常用工具
+RUN if [[ "$USE_INTRANET" == "true" ]]; then \
+        echo "Configuring intranet apt sources" && \
+        if [[ -f /etc/apt/sources.list ]]; then cp /etc/apt/sources.list /etc/apt/sources.list.bak; fi && \
+        mkdir -p /etc/apt && \
+        echo "deb [trusted=yes] http://10.68.64.1/ubuntu2204/ jammy main" > /etc/apt/sources.list && \
+        echo 'Acquire::https::Verify-Peer "false";' > /etc/apt/apt.conf.d/99disable-ssl-check && \
+        echo 'Acquire::https::Verify-Host "false";' >> /etc/apt/apt.conf.d/99disable-ssl-check; \
+    fi && \
+    apt-get update && \
+    apt-get install -y supervisor net-tools inetutils-ping vim && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# 运行期切换到运行所需的 apt 源
+RUN if [[ "$USE_INTRANET" == "true" ]]; then \
+        echo "deb [trusted=yes] https://10.92.132.52/mirrors/ubuntu2204/ jammy main" > /etc/apt/sources.list; \
+    fi
+
+RUN mkdir -p /var/log/supervisor
+
+COPY ./src/master/build/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+COPY ./src/master/build/start-master.sh /usr/local/bin/start-master.sh
+COPY ./src/master/build/dns-monitor.sh /usr/local/bin/dns-monitor.sh
+RUN chmod +x /usr/local/bin/start-master.sh /usr/local/bin/dns-monitor.sh
+
+COPY ./src/master/app ./app

 EXPOSE 3000

-CMD ["gunicorn", "--bind", "0.0.0.0:3000", "app:create_app()"]
+CMD ["/usr/bin/supervisord", "-c", "/etc/supervisor/conf.d/supervisord.conf"]
--- a/src/master/README.md
+++ b/src/master/README.md
@ -51,6 +51,21 @@ cd src/master/tests
 | `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。 |

+## 进程与监控
+
+镜像内通过 `supervisord` 管理进程：
+
+- `master`：执行 `/usr/local/bin/start-master.sh`，默认以 4 个 Gunicorn worker 监听 `0.0.0.0:3000`；可通过环境变量 `GUNICORN_WORKERS`、`GUNICORN_BIND`、`GUNICORN_EXTRA_ARGS` 调整。
+- `dns-monitor`：轮询 `/private/argus/etc/dns.conf`，若发现变更则调用 `/private/argus/etc/update-dns.sh`，日志输出在 `/var/log/supervisor/dns-monitor.log`。
+
+镜像构建阶段会安装 `supervisor`/`net-tools`/`inetutils-ping`/`vim` 等基础工具，并在运行前把 apt 源切换到内网镜像，方便容器内进一步运维。
+
+## 域名注册与 DNS 联动
+
+- Master 容器启动时会主动执行 `/private/argus/etc/update-dns.sh`（若存在），把自身 `/etc/resolv.conf` 指向 bind 服务提供的 DNS；随后解析 `eth0` 的 IPv4 地址并写入 `/private/argus/etc/master.argus.com`。该文件会被 bind 模块的 `argus_dns_sync.sh` 监控，用于生成 `master.argus.com` → 当前容器 IP 的 A 记录。
+- 测试与生产都需要将 bind 下发的 `update-dns.sh`、`dns.conf` 等文件挂载到 `/private/argus/etc/`。在 E2E 场景中，`tests/private/argus/etc` 会由脚本自动准备。
+- 其他模块（如 agent）在启动脚本中只需执行同一份 `update-dns.sh`，即可使用域名访问 master；若域名注册异常，agent 将无法成功上报，可据此快速定位问题。
+
 ## REST API 详解

 基础路径：`/api/v1/master`，全部返回 JSON。
--- a/src/master/build/dns-monitor.sh
+++ b/src/master/build/dns-monitor.sh
@ -0,0 +1 @@
+../../bind/build/dns-monitor.sh
--- a/src/master/build/start-master.sh
+++ b/src/master/build/start-master.sh
@ -0,0 +1,38 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 中文提示：确保共享目录与 DNS 相关脚本存在
+DNS_DIR="/private/argus/etc"
+DNS_SCRIPT="${DNS_DIR}/update-dns.sh"
+MASTER_DOMAIN_FILE="${DNS_DIR}/master.argus.com"
+
+mkdir -p "$DNS_DIR"
+
+if [[ -x "$DNS_SCRIPT" ]]; then
+  echo "[INFO] Running update-dns.sh before master starts"
+  # 中文提示：若脚本存在则执行，保证容器使用 bind 作为 DNS
+  "$DNS_SCRIPT" || echo "[WARN] update-dns.sh execution failed"
+else
+  echo "[WARN] DNS update script not found or not executable: $DNS_SCRIPT"
+fi
+
+# 中文提示：记录 master 当前 IP，供 bind 服务同步
+MASTER_IP=$(ifconfig | grep -A 1 eth0 | grep inet | awk '{print $2}' || true)
+if [[ -n "${MASTER_IP}" ]]; then
+  echo "current IP: ${MASTER_IP}"
+  echo "${MASTER_IP}" > "$MASTER_DOMAIN_FILE"
+else
+  echo "[WARN] Failed to detect master IP via ifconfig"
+fi
+
+WORKERS=${GUNICORN_WORKERS:-4}
+BIND_ADDR=${GUNICORN_BIND:-0.0.0.0:3000}
+EXTRA_OPTS=${GUNICORN_EXTRA_ARGS:-}
+
+if [[ -n "$EXTRA_OPTS" ]]; then
+  read -r -a EXTRA_ARRAY <<< "$EXTRA_OPTS"
+else
+  EXTRA_ARRAY=()
+fi
+
+exec gunicorn --bind "$BIND_ADDR" --workers "$WORKERS" "${EXTRA_ARRAY[@]}" "app:create_app()"
--- a/src/master/build/supervisord.conf
+++ b/src/master/build/supervisord.conf
@ -0,0 +1,39 @@
+[supervisord]
+nodaemon=true
+logfile=/var/log/supervisor/supervisord.log
+pidfile=/var/run/supervisord.pid
+user=root
+
+[program:master]
+command=/usr/local/bin/start-master.sh
+user=root
+stdout_logfile=/var/log/supervisor/master.log
+stderr_logfile=/var/log/supervisor/master_error.log
+autostart=true
+autorestart=true
+startsecs=5
+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
+autostart=true
+autorestart=true
+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
--- a/src/master/scripts/build_images.sh
+++ b/src/master/scripts/build_images.sh
@ -13,21 +13,27 @@ USAGE
 }

 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
+PROJECT_ROOT="$(cd "$SCRIPT_DIR/../../.." && pwd)"
+MODULE_ROOT="$PROJECT_ROOT/src/master"
 IMAGE_TAG="${IMAGE_TAG:-argus-master:dev}"
+DOCKERFILE="src/master/Dockerfile"
 BUILD_ARGS=()
 OFFLINE_MODE=0

+cd "$PROJECT_ROOT"
+
 while [[ "$#" -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=${INTRANET_INDEX}")
+      BUILD_ARGS+=("--build-arg" "USE_INTRANET=true")
      shift
      ;;
    --offline)
      OFFLINE_MODE=1
      BUILD_ARGS+=("--build-arg" "USE_OFFLINE=1")
+      BUILD_ARGS+=("--build-arg" "USE_INTRANET=true")
      shift
      ;;
    --tag)
@ -48,16 +54,19 @@ while [[ "$#" -gt 0 ]]; do
 done

 if [[ "$OFFLINE_MODE" -eq 1 ]]; then
-  WHEELS_DIR="$PROJECT_ROOT/offline_wheels"
+  WHEELS_DIR="$MODULE_ROOT/offline_wheels"
  if [[ ! -d "$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
+    echo "[ERROR] offline_wheels 目录为空，请先在有网环境执行 scripts/prepare_offline_wheels.sh" >&2
+    exit 1
  fi
 fi

+
+
 echo "[INFO] Building image $IMAGE_TAG"
-docker build "${BUILD_ARGS[@]}" -t "$IMAGE_TAG" "$PROJECT_ROOT"
+docker build -f "$DOCKERFILE" "${BUILD_ARGS[@]}" -t "$IMAGE_TAG" "$PROJECT_ROOT"
 echo "[OK] Image $IMAGE_TAG built"
--- a/src/master/tests/docker-compose.yml
+++ b/src/master/tests/docker-compose.yml
@ -11,6 +11,7 @@ services:
    volumes:
      - ./private/argus/master:/private/argus/master
      - ./private/argus/metric/prometheus:/private/argus/metric/prometheus
+      - ./private/argus/etc:/private/argus/etc
    restart: unless-stopped

 networks:
--- a/src/master/tests/scripts/01_up_master.sh
+++ b/src/master/tests/scripts/01_up_master.sh
@ -6,10 +6,14 @@ TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 MODULE_ROOT="$(cd "$TEST_ROOT/.." && pwd)"
 PRIVATE_ROOT="$TEST_ROOT/private"
 TMP_ROOT="$TEST_ROOT/tmp"
+DNS_ROOT="$PRIVATE_ROOT/argus/etc"
+BIND_UPDATE_SCRIPT_SRC="$(cd "$MODULE_ROOT/../bind" && pwd)/build/update-dns.sh"
+BIND_UPDATE_SCRIPT_DEST="$DNS_ROOT/update-dns.sh"

 mkdir -p "$PRIVATE_ROOT/argus/master"
 mkdir -p "$PRIVATE_ROOT/argus/metric/prometheus"
 mkdir -p "$TMP_ROOT"
+mkdir -p "$DNS_ROOT"

 # 确保上一次运行留下的容器/数据被清理
 compose() {
@ -28,6 +32,15 @@ rm -rf "$TMP_ROOT" "$PRIVATE_ROOT"
 mkdir -p "$PRIVATE_ROOT/argus/master"
 mkdir -p "$PRIVATE_ROOT/argus/metric/prometheus"
 mkdir -p "$TMP_ROOT"
+mkdir -p "$DNS_ROOT"
+
+# 中文提示：将 bind 模块自带的 update-dns.sh 下发到共享目录，模拟实际环境
+if [[ -f "$BIND_UPDATE_SCRIPT_SRC" ]]; then
+  cp "$BIND_UPDATE_SCRIPT_SRC" "$BIND_UPDATE_SCRIPT_DEST"
+  chmod +x "$BIND_UPDATE_SCRIPT_DEST"
+else
+  echo "[WARN] bind update script missing at $BIND_UPDATE_SCRIPT_SRC"
+fi

 pushd "$TEST_ROOT" >/dev/null
 compose down --remove-orphans || true
--- a/src/master/tests/scripts/02_verify_ready_and_nodes_json.sh
+++ b/src/master/tests/scripts/02_verify_ready_and_nodes_json.sh
@ -6,6 +6,7 @@ TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
 PRIVATE_ROOT="$TEST_ROOT/private"
 API_BASE="http://localhost:31300"
 NODES_JSON_PATH="$PRIVATE_ROOT/argus/metric/prometheus/nodes.json"
+MASTER_DOMAIN_FILE="$PRIVATE_ROOT/argus/etc/master.argus.com"

 # 等待 readyz 返回 200，确保数据库初始化完成
 for _ in {1..30}; do
@ -49,3 +50,11 @@ then
 fi

 echo "[INFO] nodes.json 初始状态校验通过"
+
+# 中文提示：输出 master 写入的域名文件，失败不影响测试
+if [[ -f "$MASTER_DOMAIN_FILE" ]]; then
+  MASTER_IP=$(<"$MASTER_DOMAIN_FILE")
+  echo "[INFO] master.argus.com 记录: $MASTER_IP"
+else
+  echo "[WARN] 未找到 master.argus.com 记录文件，目录=$MASTER_DOMAIN_FILE"
+fi
--- a/src/master/tests/scripts/09_restart_persistence.sh
+++ b/src/master/tests/scripts/09_restart_persistence.sh
@ -93,7 +93,6 @@ keys = [
    "health",
    "last_report",
    "agent_last_report",
-    "status",
 ]
 for key in keys:
    if before.get(key) != after.get(key):
@ -119,13 +118,37 @@ keys = [
    "health",
    "last_report",
    "agent_last_report",
-    "status",
 ]
 for key in keys:
    if before.get(key) != after.get(key):
        raise AssertionError(f"Key {key} changed after restart: {before.get(key)} -> {after.get(key)}")
 PY

+payload=$(python3 - <<'PY'
+import json
+from datetime import datetime, timezone
+body = {
+    "timestamp": datetime.now(timezone.utc).replace(microsecond=0).isoformat().replace("+00:00", "Z"),
+    "health": {
+        "log-fluentbit": {"status": "healthy"}
+    }
+}
+print(json.dumps(body))
+PY
+)
+
+curl -sS -o "$TMP_ROOT/restart_second_status.json" -w '%{http_code}' \
+  -H 'Content-Type: application/json' -X PUT \
+  "$API_BASE/nodes/$SECOND_NODE_ID/status" -d "$payload" > "$TMP_ROOT/restart_second_status_code"
+
+if [[ $(cat "$TMP_ROOT/restart_second_status_code") != "200" ]]; then
+  echo "[ERROR] Failed to restore second node status post-restart" >&2
+  cat "$TMP_ROOT/restart_second_status.json" >&2
+  exit 1
+fi
+
+sleep 3
+
 # 对比重启前后的 nodes.json 与统计信息，验证持久化一致性
 nodes_json_after="$TMP_ROOT/nodes_json_post_restart.json"
 cp "$PRIVATE_ROOT/argus/metric/prometheus/nodes.json" "$nodes_json_after"
Author	SHA1	Message	Date
yuyr	955d1a17b2	[#2 ] agent端到端测试使用域名完成	2025-09-25 07:03:57 +00:00
yuyr	8687b937d7	[#2 ] master模块使用supervisor守护；增加dns-monitor功能；内网apt源	2025-09-25 03:54:47 +00:00