fix: dos in annotation import (#29470)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

api/controllers/console/app/annotation.py

@@ -1,6 +1,6 @@
 from typing import Any, Literal
 
-from flask import request
+from flask import abort, request
 from flask_restx import Resource, fields, marshal, marshal_with
 from pydantic import BaseModel, Field, field_validator
 
@@ -8,6 +8,8 @@ from controllers.common.errors import NoFileUploadedError, TooManyFilesError
 from controllers.console import console_ns
 from controllers.console.wraps import (
     account_initialization_required,
+    annotation_import_concurrency_limit,
+    annotation_import_rate_limit,
     cloud_edition_billing_resource_check,
     edit_permission_required,
     setup_required,
@@ -314,18 +316,25 @@ class AnnotationUpdateDeleteApi(Resource):
 @console_ns.route("/apps/<uuid:app_id>/annotations/batch-import")
 class AnnotationBatchImportApi(Resource):
     @console_ns.doc("batch_import_annotations")
-    @console_ns.doc(description="Batch import annotations from CSV file")
+    @console_ns.doc(description="Batch import annotations from CSV file with rate limiting and security checks")
     @console_ns.doc(params={"app_id": "Application ID"})
     @console_ns.response(200, "Batch import started successfully")
     @console_ns.response(403, "Insufficient permissions")
     @console_ns.response(400, "No file uploaded or too many files")
+    @console_ns.response(413, "File too large")
+    @console_ns.response(429, "Too many requests or concurrent imports")
     @setup_required
     @login_required
     @account_initialization_required
     @cloud_edition_billing_resource_check("annotation")
+    @annotation_import_rate_limit
+    @annotation_import_concurrency_limit
     @edit_permission_required
     def post(self, app_id):
+        from configs import dify_config
+
         app_id = str(app_id)
+
         # check file
         if "file" not in request.files:
             raise NoFileUploadedError()
@@ -335,9 +344,27 @@ class AnnotationBatchImportApi(Resource):
 
         # get file from request
         file = request.files["file"]
+
         # check file type
         if not file.filename or not file.filename.lower().endswith(".csv"):
             raise ValueError("Invalid file type. Only CSV files are allowed")
+
+        # Check file size before processing
+        file.seek(0, 2)  # Seek to end of file
+        file_size = file.tell()
+        file.seek(0)  # Reset to beginning
+
+        max_size_bytes = dify_config.ANNOTATION_IMPORT_FILE_SIZE_LIMIT * 1024 * 1024
+        if file_size > max_size_bytes:
+            abort(
+                413,
+                f"File size exceeds maximum limit of {dify_config.ANNOTATION_IMPORT_FILE_SIZE_LIMIT}MB. "
+                f"Please reduce the file size and try again.",
+            )
+
+        if file_size == 0:
+            raise ValueError("The uploaded file is empty")
+
         return AppAnnotationService.batch_import_app_annotations(app_id, file)
 
 

api/controllers/console/wraps.py

@@ -331,3 +331,91 @@ def is_admin_or_owner_required(f: Callable[P, R]):
         return f(*args, **kwargs)
 
     return decorated_function
+
+
+def annotation_import_rate_limit(view: Callable[P, R]):
+    """
+    Rate limiting decorator for annotation import operations.
+
+    Implements sliding window rate limiting with two tiers:
+    - Short-term: Configurable requests per minute (default: 5)
+    - Long-term: Configurable requests per hour (default: 20)
+
+    Uses Redis ZSET for distributed rate limiting across multiple instances.
+    """
+
+    @wraps(view)
+    def decorated(*args: P.args, **kwargs: P.kwargs):
+        _, current_tenant_id = current_account_with_tenant()
+        current_time = int(time.time() * 1000)
+
+        # Check per-minute rate limit
+        minute_key = f"annotation_import_rate_limit:{current_tenant_id}:1min"
+        redis_client.zadd(minute_key, {current_time: current_time})
+        redis_client.zremrangebyscore(minute_key, 0, current_time - 60000)
+        minute_count = redis_client.zcard(minute_key)
+        redis_client.expire(minute_key, 120)  # 2 minutes TTL
+
+        if minute_count > dify_config.ANNOTATION_IMPORT_RATE_LIMIT_PER_MINUTE:
+            abort(
+                429,
+                f"Too many annotation import requests. Maximum {dify_config.ANNOTATION_IMPORT_RATE_LIMIT_PER_MINUTE} "
+                f"requests per minute allowed. Please try again later.",
+            )
+
+        # Check per-hour rate limit
+        hour_key = f"annotation_import_rate_limit:{current_tenant_id}:1hour"
+        redis_client.zadd(hour_key, {current_time: current_time})
+        redis_client.zremrangebyscore(hour_key, 0, current_time - 3600000)
+        hour_count = redis_client.zcard(hour_key)
+        redis_client.expire(hour_key, 7200)  # 2 hours TTL
+
+        if hour_count > dify_config.ANNOTATION_IMPORT_RATE_LIMIT_PER_HOUR:
+            abort(
+                429,
+                f"Too many annotation import requests. Maximum {dify_config.ANNOTATION_IMPORT_RATE_LIMIT_PER_HOUR} "
+                f"requests per hour allowed. Please try again later.",
+            )
+
+        return view(*args, **kwargs)
+
+    return decorated
+
+
+def annotation_import_concurrency_limit(view: Callable[P, R]):
+    """
+    Concurrency control decorator for annotation import operations.
+
+    Limits the number of concurrent import tasks per tenant to prevent
+    resource exhaustion and ensure fair resource allocation.
+
+    Uses Redis ZSET to track active import jobs with automatic cleanup
+    of stale entries (jobs older than 2 minutes).
+    """
+
+    @wraps(view)
+    def decorated(*args: P.args, **kwargs: P.kwargs):
+        _, current_tenant_id = current_account_with_tenant()
+        current_time = int(time.time() * 1000)
+
+        active_jobs_key = f"annotation_import_active:{current_tenant_id}"
+
+        # Clean up stale entries (jobs that should have completed or timed out)
+        stale_threshold = current_time - 120000  # 2 minutes ago
+        redis_client.zremrangebyscore(active_jobs_key, 0, stale_threshold)
+
+        # Check current active job count
+        active_count = redis_client.zcard(active_jobs_key)
+
+        if active_count >= dify_config.ANNOTATION_IMPORT_MAX_CONCURRENT:
+            abort(
+                429,
+                f"Too many concurrent import tasks. Maximum {dify_config.ANNOTATION_IMPORT_MAX_CONCURRENT} "
+                f"concurrent imports allowed per workspace. Please wait for existing imports to complete.",
+            )
+
+        # Allow the request to proceed
+        # The actual job registration will happen in the service layer
+        return view(*args, **kwargs)
+
+    return decorated