src/log/.gitignore

@@ -0,0 +1,3 @@
+
+private/
+

src/log/elasticsearch/build/Dockerfile

@@ -0,0 +1,33 @@
+FROM docker.elastic.co/elasticsearch/elasticsearch:8.13.4
+
+# 切换到 root 用户进行系统级安装
+USER root
+
+# 安装 supervisor
+RUN apt-get update && \
+    apt-get install -y supervisor && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# 创建 supervisor 日志目录
+RUN mkdir -p /var/log/supervisor
+
+# 复制 supervisor 配置文件
+COPY supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+
+# 复制启动脚本
+COPY start-es-supervised.sh /usr/local/bin/start-es-supervised.sh
+RUN chmod +x /usr/local/bin/start-es-supervised.sh
+
+# 创建数据目录并设置权限
+RUN mkdir -p /private/argus/log/elasticsearch && \
+    chown -R elasticsearch:elasticsearch /private/argus/log/elasticsearch
+
+# 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/start-es-supervised.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Elasticsearch under supervisor..."
+
+# 创建数据目录（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
+
+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 +1,27 @@
+[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
+
+[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/kibana/build/Dockerfile

@@ -0,0 +1,34 @@
+FROM docker.elastic.co/kibana/kibana:8.13.4
+
+# 切换到 root 用户进行系统级安装
+USER root
+
+# 安装 supervisor
+RUN apt-get update && \
+    apt-get install -y supervisor && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# 创建 supervisor 日志目录
+RUN mkdir -p /var/log/supervisor
+
+# 复制 supervisor 配置文件
+COPY supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+
+# 复制启动脚本
+COPY start-kibana-supervised.sh /usr/local/bin/start-kibana-supervised.sh
+COPY 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
+
+# 创建数据目录并设置权限
+RUN mkdir -p /private/argus/log/kibana && \
+    chown -R kibana:kibana /private/argus/log/kibana
+
+# 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/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/data_view" -H 'kbn-xsrf: true' 2>/dev/null | grep '"title":"train-\*"' | wc -l || echo "0")
+
+        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 "$KB_HOST/api/data_views/data_view" -H 'kbn-xsrf: true' 2>/dev/null | grep '"title":"infer-\*"' | wc -l || echo "0")
+
+        if [ "$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 +1,32 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "[INFO] Starting Kibana under supervisor..."
+
+# 创建数据目录（如果不存在）
+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"
+
+# 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 +1,27 @@
+[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
+
+[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/misc/01_bootstrap.sh

@@ -1,47 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-root="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
-
-# 创建private目录结构 (只需要ES和Kibana)
-echo "[INFO] Creating private directory structure..."
-mkdir -p "$root/private/es/data"
-mkdir -p "$root/private/es/scripts" 
-mkdir -p "$root/private/kibana/data"
-mkdir -p "$root/private/kibana/scripts"
-
-# 复制启动脚本
-echo "[INFO] Copying startup scripts..."
-cp "$root/scripts/start-es.sh" "$root/private/es/scripts/"
-cp "$root/scripts/start-kibana.sh" "$root/private/kibana/scripts/"
-
-# 设置执行权限
-chmod +x "$root/private/es/scripts/start-es.sh"
-chmod +x "$root/private/kibana/scripts/start-kibana.sh"
-
-# 设置数据目录权限（ES 和 Kibana 容器都使用 UID 1000）
-sudo chown -R 1000:1000 "$root/private/es/data" "$root/private/kibana/data" 2>/dev/null || true
-
-# 检查fluent-bit相关文件是否存在
-if [[ ! -f "$root/fluent-bit-bundle.tar.gz" ]]; then
-    echo "[INFO] Creating fluent-bit bundle..."
-    # 创建bundle（如果目录存在的话）
-    cd "$root"
-    if [[ -d "private/fluent-bit" ]]; then
-        cd private/fluent-bit && tar -czf ../../fluent-bit-bundle.tar.gz etc/ packages/ 2>/dev/null && cd ../..
-    elif [[ -d "fluent-bit" && -d "packages" ]]; then
-        # 临时创建目录结构来打包
-        mkdir -p temp-bundle/etc temp-bundle/packages
-        cp -r fluent-bit/* temp-bundle/etc/
-        cp -r packages/* temp-bundle/packages/
-        cd temp-bundle && tar -czf ../fluent-bit-bundle.tar.gz . && cd ..
-        rm -rf temp-bundle
-    else
-        echo "[WARN] 无法创建fluent-bit bundle，请确保fluent-bit配置和packages目录存在"
-    fi
-fi
-
-if [[ ! -f "$root/start-fluent-bit.sh" ]]; then
-    echo "[WARN] 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/{es,kibana}, fluent-bit-bundle.tar.gz, start-fluent-bit.sh"

src/log/misc/03_send_test.sh

@@ -1,17 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-echo "[INFO] 向两个fluent-bit容器发送测试日志..."
-
-# 发送日志到host01
-echo "[INFO] 发送日志到 host01..."
-./scripts/03_send_test_host01.sh
-
-echo ""
-
-# 发送日志到host02
-echo "[INFO] 发送日志到 host02..."
-./scripts/03_send_test_host02.sh
-
-echo ""
-echo "[OK] 已完成向两个主机发送测试日志"

src/log/misc/06_restart_fluentbit.sh

@@ -1,10 +0,0 @@
-#!/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 restart fluent-bit
-echo "[OK] 已重启 fluent-bit（该镜像不支持 SIGHUP 热重载）"

src/log/misc/07_create_data_views.sh

@@ -1,44 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-KB="${KB:-http://localhost:5601}"
-
-# 等待 Kibana 完全启动
-wait_for_kibana() {
-  echo "[i] 等待 Kibana 启动..."
-  local max_attempts=60
-  local attempt=1
-  
-  while [ $attempt -le $max_attempts ]; do
-    if curl -fs "$KB/api/status" >/dev/null 2>&1; then
-      local status=$(curl -s "$KB/api/status" | jq -r '.status.overall.level // "unknown"')
-      if [ "$status" = "available" ]; then
-        echo "[OK] Kibana 已启动 (status: $status)"
-        return 0
-      else
-        echo "    等待中... ($attempt/$max_attempts, status: $status)"
-      fi
-    else
-      echo "    等待中... ($attempt/$max_attempts, 连接失败)"
-    fi
-    sleep 5
-    ((attempt++))
-  done
-  
-  echo "[ERROR] Kibana 启动超时"
-  return 1
-}
-create_view() {
-  local name="$1" pattern="$2" timefield="${3:-@timestamp}"
-  echo "[i] 创建 Data View: $name ($pattern, time=$timefield)"
-  curl -fsS -X POST "$KB/api/data_views/data_view" \
-    -H 'kbn-xsrf: true' \
-    -H 'Content-Type: application/json' \
-    -d "{\"data_view\":{\"name\":\"$name\",\"title\":\"$pattern\",\"timeFieldName\":\"$timefield\"}}" \
-    >/dev/null && echo " -> OK" || { echo " -> 失败（可能是没有匹配索引）"; return 1; }
-}
-# 等待 Kibana 启动
-wait_for_kibana || exit 1
-
-create_view "train" "train-*" "@timestamp" || true
-create_view "infer" "infer-*" "@timestamp" || true
-echo "[DONE] 若提示失败，请先确保已产生 train-*/infer-* 索引，再重试本脚本。"

src/log/misc/08_fix_replicas.sh

@@ -1,8 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-ES="${ES:-http://localhost:9200}"
-for idx in train-* infer-*; do
-  echo "[i] 将 $idx 副本数设置为 0"
-  curl -fsS -X PUT "$ES/$idx/_settings" -H 'Content-Type: application/json' -d '{"index":{"number_of_replicas":0}}' || true
-done
-echo "[OK] 完成"

src/log/misc/99_clean.sh

@@ -1,33 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-cd "$(dirname "$0")/.."
-
-echo "清理选项："
-echo "1. 清理索引和缓冲区（保留持久化数据）"
-echo "2. 完全清理（包括持久化数据目录）"
-read -rp "请选择 (1/2): " choice
-
-case $choice in
-  1)
-    read -rp "危险操作：删除 ES 索引 train-*, infer-*（以及 logstash-*），并清空 buffers。确认? (yes/NO) " ans
-    if [[ "${ans:-NO}" != "yes" ]]; then echo "已取消"; exit 0; fi
-    ES="${ES:-http://localhost:9200}"
-    for idx in train-* infer-* logstash-*; do
-      echo "[i] 删除索引 $idx"
-      curl -fsS -X DELETE "$ES/$idx" || true
-    done
-    rm -rf ./private/fluent-bit/buffers/* || true
-    echo "[OK] 索引和缓冲区清理完成"
-    ;;
-  2)
-    read -rp "危险操作：删除所有数据包括持久化存储！确认? (yes/NO) " ans
-    if [[ "${ans:-NO}" != "yes" ]]; then echo "已取消"; exit 0; fi
-    rm -rf ./private/fluent-bit/buffers/* ./private/es/data/* ./private/kibana/data/* || true
-    rm -rf ./private/fluent-bit/logs/train/* ./private/fluent-bit/logs/infer/* || true
-    echo "[OK] 完全清理完成（包括持久化数据）"
-    ;;
-  *)
-    echo "已取消"
-    exit 0
-    ;;
-esac

src/log/misc/start-es.sh

@@ -1,19 +0,0 @@
-#!/bin/bash
-set -euo pipefail
-
-echo "[INFO] Starting Elasticsearch with private directory setup..."
-
-# 创建数据目录
-mkdir -p /private/es/data
-
-# 创建软链接到Elasticsearch预期的数据目录
-rm -rf /usr/share/elasticsearch/data 2>/dev/null || true
-ln -sf /private/es/data /usr/share/elasticsearch/data
-
-# 设置正确的权限 (Elasticsearch使用UID 1000)
-chown -R 1000:1000 /private/es/data
-
-echo "[INFO] Data directory linked: /usr/share/elasticsearch/data -> /private/es/data"
-
-# 启动原始的Elasticsearch entrypoint
-exec /usr/local/bin/docker-entrypoint.sh elasticsearch

src/log/misc/start-kibana.sh

@@ -1,19 +0,0 @@
-#!/bin/bash
-set -euo pipefail
-
-echo "[INFO] Starting Kibana with private directory setup..."
-
-# 创建数据目录
-mkdir -p /private/kibana/data
-
-# 创建软链接到Kibana预期的数据目录
-rm -rf /usr/share/kibana/data 2>/dev/null || true
-ln -sf /private/kibana/data /usr/share/kibana/data
-
-# 设置正确的权限 (Kibana使用UID 1000)
-chown -R 1000:1000 /private/kibana/data
-
-echo "[INFO] Data directory linked: /usr/share/kibana/data -> /private/kibana/data"
-
-# 启动原始的Kibana entrypoint
-exec /usr/local/bin/kibana-docker

src/log/scripts/build_images.sh

@@ -0,0 +1,20 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+root="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+cd "$root"
+
+echo "[INFO] Building custom Docker images with supervisor support..."
+
+# 构建 Elasticsearch 镜像
+echo "[INFO] Building Elasticsearch image..."
+docker build -t argus-elasticsearch:latest ./elasticsearch/build
+
+# 构建 Kibana 镜像
+echo "[INFO] Building Kibana image..."
+docker build -t argus-kibana:latest ./kibana/build
+
+echo "[OK] Custom images built successfully:"
+echo "  - argus-elasticsearch:latest"
+echo "  - argus-kibana:latest"
+

src/log/tests/docker-compose.yml

@@ -1,15 +1,17 @@
 version: "3.8"
 services:
   es:
-    image: docker.elastic.co/elasticsearch/elasticsearch:8.13.4
+    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:/private
+      - ./private/argus/log/elasticsearch:/private/argus/log/elasticsearch
     ports: ["9200:9200"]
-    command: /private/es/scripts/start-es.sh
     healthcheck:
       test: ["CMD-SHELL", "curl -fs http://localhost:9200 >/dev/null || exit 1"]
       interval: 10s
@@ -17,13 +19,15 @@ services:
       retries: 30
 
   kibana:
-    image: docker.elastic.co/kibana/kibana:8.13.4
+    build:
+      context: ../kibana/build
+      dockerfile: Dockerfile
+    image: argus-kibana:latest
     environment:
       - ELASTICSEARCH_HOSTS=http://es:9200
     volumes:
-      - ./private:/private
+      - ./private/argus/log/kibana:/private/argus/log/kibana
     ports: ["5601:5601"]
-    command: /private/kibana/scripts/start-kibana.sh
     depends_on:
       es:
         condition: service_healthy
@@ -37,8 +41,8 @@ services:
       - ES_HOST=es
       - ES_PORT=9200
     volumes:
-      - ./start-fluent-bit.sh:/private/start-fluent-bit.sh:ro
-      - ./fluent-bit-bundle.tar.gz:/private/fluent-bit-bundle.tar.gz:ro
+      - ../fluent-bit/start-fluent-bit.sh:/private/start-fluent-bit.sh:ro
+      - ../fluent-bit/fluent-bit-bundle.tar.gz:/private/fluent-bit-bundle.tar.gz:ro
     ports: ["2020:2020"]
     depends_on:
       es:
@@ -59,8 +63,8 @@ services:
       - ES_HOST=es
       - ES_PORT=9200
     volumes:
-      - ./start-fluent-bit.sh:/private/start-fluent-bit.sh:ro
-      - ./fluent-bit-bundle.tar.gz:/private/fluent-bit-bundle.tar.gz:ro
+      - ../fluent-bit/start-fluent-bit.sh:/private/start-fluent-bit.sh:ro
+      - ../fluent-bit/fluent-bit-bundle.tar.gz:/private/fluent-bit-bundle.tar.gz:ro
     ports: ["2021:2020"]
     depends_on:
       es:

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

@@ -0,0 +1,27 @@
+#!/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 "$root/private/argus/log/kibana"
+
+# 设置数据目录权限（ES 和 Kibana 容器都使用 UID 1000）
+echo "[INFO] Setting permissions for data directories..."
+sudo chown -R 1000:1000 "$root/private/argus/log/elasticsearch" 2>/dev/null || true
+sudo chown -R 1000:1000 "$root/private/argus/log/kibana" 2>/dev/null || true
+
+echo "[INFO] Supervisor-based containers will manage their own scripts and configurations"
+
+# 检查fluent-bit相关文件是否存在
+if [[ ! -f "$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/03_send_test_host01.sh

@@ -5,7 +5,7 @@ set -euo pipefail
 container_name="logging-mvp-fluent-bit-host01-1"
 
 # 检查容器是否存在并运行
-if ! docker ps --format "table {{.Names}}" | grep -q "$container_name"; then
+if ! docker ps | grep -q "$container_name"; then
     echo "[ERROR] Fluent Bit容器 $container_name 未运行"
     exit 1
 fi

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

@@ -5,7 +5,7 @@ set -euo pipefail
 container_name="logging-mvp-fluent-bit-host02-1"
 
 # 检查容器是否存在并运行
-if ! docker ps --format "table {{.Names}}" | grep -q "$container_name"; then
+if ! docker ps | grep -q "$container_name"; then
     echo "[ERROR] Fluent Bit容器 $container_name 未运行"
     exit 1
 fi

src/log/tests/scripts/e2e_test.sh

@@ -0,0 +1,168 @@
+#!/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 "   • All services started and stopped successfully"
+echo ""
+echo "✅ The ARGUS log system is working correctly!"
+echo ""