Merge remote-tracking branch 'origin/develop' into feat/registry-retry-stability

- 新增retry_cmd通用重试函数（3次重试+指数退避）
- 新增retry_docker_login：Registry登录失败自动重试
- 新增retry_docker_pull：镜像拉取失败自动重试
- 解决Registry网络抖动导致的部署偶发失败

scripts/ci_staging_deploy.sh

@@ -15,6 +15,40 @@
 
 set -eu
 
+# ---- 重试工具函数 ----
+retry_cmd() {
+  local max_attempts=$1
+  local backoff=$2
+  shift 2
+  local attempt=1
+  while [ $attempt -le $max_attempts ]; do
+    if "$@"; then
+      return 0
+    fi
+    echo "  attempt $attempt/$max_attempts failed, retrying in ${backoff}s..."
+    sleep $backoff
+    backoff=$((backoff * 2))
+    attempt=$((attempt + 1))
+  done
+  echo "  ERROR: failed after $max_attempts retries"
+  return 1
+}
+
+retry_docker_login() {
+  echo "Logging in to registry (up to 3 retries)"
+  if retry_cmd 3 5 sh -c "printf %s "$REGISTRY_TOKEN" | docker login "$REGISTRY_HOST" -u "$REGISTRY_USER" --password-stdin"; then
+    return 0
+  fi
+  echo "WARN: docker login failed after retries, will try pull anyway"
+  return 0
+}
+
+retry_docker_pull() {
+  local image=$1
+  echo "Pulling $image (up to 3 retries)"
+  retry_cmd 3 10 docker pull "$image"
+}
+
 IMAGE_TAG="${IMAGE_TAG:-}"
 REGISTRY="${REGISTRY:-xiaoxia-registry.cn-hangzhou.cr.aliyuncs.com/xiaoxiakeji}"
 REGISTRY_USER="${ACR_USERNAME:-${REGISTRY_USER:-nick0415343655}}"
@@ -41,11 +75,11 @@ echo "=========================================="
 
 # ---- 登录 Registry ----
 if [ -n "$REGISTRY_TOKEN" ]; then
-  echo "Logging in to registry: $REGISTRY"
+  echo "=========================================="
+  echo "  Login to Registry (with retries)"
+  echo "=========================================="
   REGISTRY_HOST=$(echo "$REGISTRY" | cut -d/ -f1)
-  printf %s "$REGISTRY_TOKEN" | docker login "$REGISTRY_HOST" -u "$REGISTRY_USER" --password-stdin 2>/dev/null || {
-    echo "WARN: docker login failed, will try to pull anyway"
-  }
+  retry_docker_login
 fi
 
 # ---- Pull 新版本镜像 ----
@@ -57,12 +91,12 @@ LOCAL_API="xiaoxia-saas-api:${IMAGE_TAG}"
 LOCAL_WORKER="xiaoxia-saas-worker:${IMAGE_TAG}"
 LOCAL_WEB="xiaoxia-saas-web:${IMAGE_TAG}"
 
-echo "Pulling API image..."
-docker pull "$REGISTRY_API"
-echo "Pulling Worker image..."
-docker pull "$REGISTRY_WORKER"
-echo "Pulling Web image..."
-docker pull "$REGISTRY_WEB"
+echo "=========================================="
+echo "  Pull images (with retries)"
+echo "=========================================="
+retry_docker_pull "$REGISTRY_API"
+retry_docker_pull "$REGISTRY_WORKER"
+retry_docker_pull "$REGISTRY_WEB"
 
 # Re-tag 成本地名
 docker tag "$REGISTRY_API" "$LOCAL_API"