dev_1.0.0_yuyr_2：重新提交 PR，增加 master/agent 以及系统集成测试 (#17)

Reviewed-on: #17
Reviewed-by: sundapeng <sundp@mail.zgclab.edu.cn>
Reviewed-by: xuxt <xuxt@zgclab.edu.cn>

src/log/tests/docker-compose.yml

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

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

@@ -1,6 +1,10 @@
 #!/usr/bin/env bash
 set -euo pipefail
 root="$(cd "$(dirname "${BASH_SOURCE[0]}")/../" && pwd)"
+project_root="$(cd "$root/../../.." && pwd)"
+
+source "$project_root/scripts/common/build_user.sh"
+load_build_user
 
 # 创建新的private目录结构 (基于argus目录结构)
 echo "[INFO] Creating private directory structure for supervisor-based containers..."
@@ -11,9 +15,9 @@ mkdir -p "$root/private/argus/etc/"
 
 # 设置数据目录权限（ES 和 Kibana 容器都使用 UID 1000）
 echo "[INFO] Setting permissions for data directories..."
-sudo chown -R 2133:2015 "$root/private/argus/log/elasticsearch" 2>/dev/null || true
-sudo chown -R 2133:2015 "$root/private/argus/log/kibana" 2>/dev/null || true
-sudo chown -R 2133:2015 "$root/private/argus/etc" 2>/dev/null || true
+chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" "$root/private/argus/log/elasticsearch" 2>/dev/null || true
+chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" "$root/private/argus/log/kibana" 2>/dev/null || true
+chown -R "${ARGUS_BUILD_UID}:${ARGUS_BUILD_GID}" "$root/private/argus/etc" 2>/dev/null || true
 
 echo "[INFO] Supervisor-based containers will manage their own scripts and configurations"
 

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

@@ -4,8 +4,22 @@ set -euo pipefail
 # 获取fluent-bit-host01容器名称
 container_name="logging-mvp-fluent-bit-host01-1"
 
-# 检查容器是否存在并运行
-if ! docker ps | grep -q "$container_name"; then
+wait_for_container() {
+    local name="$1"
+    local attempts=30
+    local delay=5
+    local i
+    for ((i = 1; i <= attempts; i++)); do
+        if docker ps --format '{{.Names}}' | grep -Fx "$name" >/dev/null; then
+            return 0
+        fi
+        echo "[INFO] 等待容器 $name 启动中... ($i/$attempts)"
+        sleep "$delay"
+    done
+    return 1
+}
+
+if ! wait_for_container "$container_name"; then
     echo "[ERROR] Fluent Bit容器 $container_name 未运行"
     exit 1
 fi
@@ -28,4 +42,4 @@ 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"
+echo " - /logs/infer/infer-demo.log"

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

@@ -4,8 +4,22 @@ set -euo pipefail
 # 获取fluent-bit-host02容器名称
 container_name="logging-mvp-fluent-bit-host02-1"
 
-# 检查容器是否存在并运行
-if ! docker ps | grep -q "$container_name"; then
+wait_for_container() {
+    local name="$1"
+    local attempts=30
+    local delay=5
+    local i
+    for ((i = 1; i <= attempts; i++)); do
+        if docker ps --format '{{.Names}}' | grep -Fx "$name" >/dev/null; then
+            return 0
+        fi
+        echo "[INFO] 等待容器 $name 启动中... ($i/$attempts)"
+        sleep "$delay"
+    done
+    return 1
+}
+
+if ! wait_for_container "$container_name"; then
     echo "[ERROR] Fluent Bit容器 $container_name 未运行"
     exit 1
 fi
@@ -24,4 +38,4 @@ docker exec "$container_name" sh -c "printf '%s INFO [host02] inference complete
 
 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"
+echo " - /logs/infer/infer-demo.log"

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

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

src/log/tests/scripts/e2e_test.sh

@@ -19,7 +19,7 @@ get_log_count() {
 # 函数：等待服务就绪
 wait_for_services() {
     echo "[INFO] Waiting for all services to be ready..."
-    local max_attempts=60
+    local max_attempts=${SERVICE_WAIT_ATTEMPTS:-120}
     local attempt=1
 
     while [ $attempt -le $max_attempts ]; do