refactor: implement tenant self queue for rag tasks (#27559)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: -LAN- <laipz8200@outlook.com>
2025-12-19 17:27:16 -05:00 · 2025-11-06 21:25:50 +08:00
parent f4c82d0010
commit 37903722fe
24 changed files with 3433 additions and 94 deletions
--- a/api/.env.example
+++ b/api/.env.example
@@ -615,5 +615,8 @@ SWAGGER_UI_PATH=/swagger-ui.html
 # Set to false to export dataset IDs as plain text for easier cross-environment import
 DSL_EXPORT_ENCRYPT_DATASET_ID=true

+# Tenant isolated task queue configuration
+TENANT_ISOLATED_TASK_CONCURRENCY=1
+
 # Maximum number of segments for dataset segments API (0 for unlimited)
 DATASET_MAX_SEGMENTS_PER_REQUEST=0
--- a/api/configs/feature/init.py
+++ b/api/configs/feature/init.py
@@ -1142,6 +1142,13 @@ class SwaggerUIConfig(BaseSettings):
    )


+class TenantIsolatedTaskQueueConfig(BaseSettings):
+    TENANT_ISOLATED_TASK_CONCURRENCY: int = Field(
+        description="Number of tasks allowed to be delivered concurrently from isolated queue per tenant",
+        default=1,
+    )
+
+
 class FeatureConfig(
    # place the configs in alphabet order
    AppExecutionConfig,
@@ -1166,6 +1173,7 @@ class FeatureConfig(
    RagEtlConfig,
    RepositoryConfig,
    SecurityConfig,
+    TenantIsolatedTaskQueueConfig,
    ToolConfig,
    UpdateConfig,
    WorkflowConfig,
--- a/api/core/app/apps/pipeline/pipeline_generator.py
+++ b/api/core/app/apps/pipeline/pipeline_generator.py
@@ -40,20 +40,15 @@ from core.workflow.repositories.draft_variable_repository import DraftVariableSa
 from core.workflow.repositories.workflow_execution_repository import WorkflowExecutionRepository
 from core.workflow.repositories.workflow_node_execution_repository import WorkflowNodeExecutionRepository
 from core.workflow.variable_loader import DUMMY_VARIABLE_LOADER, VariableLoader
-from enums.cloud_plan import CloudPlan
 from extensions.ext_database import db
-from extensions.ext_redis import redis_client
 from libs.flask_utils import preserve_flask_contexts
 from models import Account, EndUser, Workflow, WorkflowNodeExecutionTriggeredFrom
 from models.dataset import Document, DocumentPipelineExecutionLog, Pipeline
 from models.enums import WorkflowRunTriggeredFrom
 from models.model import AppMode
 from services.datasource_provider_service import DatasourceProviderService
-from services.feature_service import FeatureService
-from services.file_service import FileService
+from services.rag_pipeline.rag_pipeline_task_proxy import RagPipelineTaskProxy
 from services.workflow_draft_variable_service import DraftVarLoader, WorkflowDraftVariableService
-from tasks.rag_pipeline.priority_rag_pipeline_run_task import priority_rag_pipeline_run_task
-from tasks.rag_pipeline.rag_pipeline_run_task import rag_pipeline_run_task

 logger = logging.getLogger(__name__)

@@ -249,34 +244,7 @@ class PipelineGenerator(BaseAppGenerator):
                )

        if rag_pipeline_invoke_entities:
-            # store the rag_pipeline_invoke_entities to object storage
-            text = [item.model_dump() for item in rag_pipeline_invoke_entities]
-            name = "rag_pipeline_invoke_entities.json"
-            # Convert list to proper JSON string
-            json_text = json.dumps(text)
-            upload_file = FileService(db.engine).upload_text(json_text, name, user.id, dataset.tenant_id)
-            features = FeatureService.get_features(dataset.tenant_id)
-            if features.billing.enabled and features.billing.subscription.plan == CloudPlan.SANDBOX:
-                tenant_pipeline_task_key = f"tenant_pipeline_task:{dataset.tenant_id}"
-                tenant_self_pipeline_task_queue = f"tenant_self_pipeline_task_queue:{dataset.tenant_id}"
-
-                if redis_client.get(tenant_pipeline_task_key):
-                    # Add to waiting queue using List operations (lpush)
-                    redis_client.lpush(tenant_self_pipeline_task_queue, upload_file.id)
-                else:
-                    # Set flag and execute task
-                    redis_client.set(tenant_pipeline_task_key, 1, ex=60 * 60)
-                    rag_pipeline_run_task.delay(  # type: ignore
-                        rag_pipeline_invoke_entities_file_id=upload_file.id,
-                        tenant_id=dataset.tenant_id,
-                    )
-
-            else:
-                priority_rag_pipeline_run_task.delay(  # type: ignore
-                    rag_pipeline_invoke_entities_file_id=upload_file.id,
-                    tenant_id=dataset.tenant_id,
-                )
-
+            RagPipelineTaskProxy(dataset.tenant_id, user.id, rag_pipeline_invoke_entities).delay()
        # return batch, dataset, documents
        return {
            "batch": batch,
--- a/api/core/entities/document_task.py
+++ b/api/core/entities/document_task.py
@@ -0,0 +1,15 @@
+from collections.abc import Sequence
+from dataclasses import dataclass
+
+
+@dataclass
+class DocumentTask:
+    """Document task entity for document indexing operations.
+
+    This class represents a document indexing task that can be queued
+    and processed by the document indexing system.
+    """
+
+    tenant_id: str
+    dataset_id: str
+    document_ids: Sequence[str]
--- a/api/core/rag/pipeline/init.py
+++ b/api/core/rag/pipeline/init.py
--- a/api/core/rag/pipeline/queue.py
+++ b/api/core/rag/pipeline/queue.py
@@ -0,0 +1,79 @@
+import json
+from collections.abc import Sequence
+from typing import Any
+
+from pydantic import BaseModel, ValidationError
+
+from extensions.ext_redis import redis_client
+
+_DEFAULT_TASK_TTL = 60 * 60  # 1 hour
+
+
+class TaskWrapper(BaseModel):
+    data: Any
+
+    def serialize(self) -> str:
+        return self.model_dump_json()
+
+    @classmethod
+    def deserialize(cls, serialized_data: str) -> "TaskWrapper":
+        return cls.model_validate_json(serialized_data)
+
+
+class TenantIsolatedTaskQueue:
+    """
+    Simple queue for tenant isolated tasks, used for rag related tenant tasks isolation.
+    It uses Redis list to store tasks, and Redis key to store task waiting flag.
+    Support tasks that can be serialized by json.
+    """
+
+    def __init__(self, tenant_id: str, unique_key: str):
+        self._tenant_id = tenant_id
+        self._unique_key = unique_key
+        self._queue = f"tenant_self_{unique_key}_task_queue:{tenant_id}"
+        self._task_key = f"tenant_{unique_key}_task:{tenant_id}"
+
+    def get_task_key(self):
+        return redis_client.get(self._task_key)
+
+    def set_task_waiting_time(self, ttl: int = _DEFAULT_TASK_TTL):
+        redis_client.setex(self._task_key, ttl, 1)
+
+    def delete_task_key(self):
+        redis_client.delete(self._task_key)
+
+    def push_tasks(self, tasks: Sequence[Any]):
+        serialized_tasks = []
+        for task in tasks:
+            # Store str list directly, maintaining full compatibility for pipeline scenarios
+            if isinstance(task, str):
+                serialized_tasks.append(task)
+            else:
+                # Use TaskWrapper to do JSON serialization for non-string tasks
+                wrapper = TaskWrapper(data=task)
+                serialized_data = wrapper.serialize()
+                serialized_tasks.append(serialized_data)
+
+        redis_client.lpush(self._queue, *serialized_tasks)
+
+    def pull_tasks(self, count: int = 1) -> Sequence[Any]:
+        if count <= 0:
+            return []
+
+        tasks = []
+        for _ in range(count):
+            serialized_task = redis_client.rpop(self._queue)
+            if not serialized_task:
+                break
+
+            if isinstance(serialized_task, bytes):
+                serialized_task = serialized_task.decode("utf-8")
+
+            try:
+                wrapper = TaskWrapper.deserialize(serialized_task)
+                tasks.append(wrapper.data)
+            except (json.JSONDecodeError, ValidationError, TypeError, ValueError):
+                # Fall back to raw string for legacy format or invalid JSON
+                tasks.append(serialized_task)
+
+        return tasks
--- a/api/docker/entrypoint.sh
+++ b/api/docker/entrypoint.sh
@@ -32,7 +32,7 @@ if [[ "${MODE}" == "worker" ]]; then

  exec celery -A celery_entrypoint.celery worker -P ${CELERY_WORKER_CLASS:-gevent} $CONCURRENCY_OPTION \
    --max-tasks-per-child ${MAX_TASKS_PER_CHILD:-50} --loglevel ${LOG_LEVEL:-INFO} \
-    -Q ${CELERY_QUEUES:-dataset,mail,ops_trace,app_deletion,plugin,workflow_storage,conversation,priority_pipeline,pipeline} \
+    -Q ${CELERY_QUEUES:-dataset,priority_dataset,mail,ops_trace,app_deletion,plugin,workflow_storage,conversation,priority_pipeline,pipeline} \
    --prefetch-multiplier=1

 elif [[ "${MODE}" == "beat" ]]; then
--- a/api/services/dataset_service.py
+++ b/api/services/dataset_service.py
@@ -50,6 +50,7 @@ from models.model import UploadFile
 from models.provider_ids import ModelProviderID
 from models.source import DataSourceOauthBinding
 from models.workflow import Workflow
+from services.document_indexing_task_proxy import DocumentIndexingTaskProxy
 from services.entities.knowledge_entities.knowledge_entities import (
    ChildChunkUpdateArgs,
    KnowledgeConfig,
@@ -79,7 +80,6 @@ from tasks.deal_dataset_vector_index_task import deal_dataset_vector_index_task
 from tasks.delete_segment_from_index_task import delete_segment_from_index_task
 from tasks.disable_segment_from_index_task import disable_segment_from_index_task
 from tasks.disable_segments_from_index_task import disable_segments_from_index_task
-from tasks.document_indexing_task import document_indexing_task
 from tasks.document_indexing_update_task import document_indexing_update_task
 from tasks.duplicate_document_indexing_task import duplicate_document_indexing_task
 from tasks.enable_segments_to_index_task import enable_segments_to_index_task
@@ -1694,7 +1694,7 @@ class DocumentService:

                # trigger async task
                if document_ids:
-                    document_indexing_task.delay(dataset.id, document_ids)
+                    DocumentIndexingTaskProxy(dataset.tenant_id, dataset.id, document_ids).delay()
                if duplicate_document_ids:
                    duplicate_document_indexing_task.delay(dataset.id, duplicate_document_ids)

--- a/api/services/document_indexing_task_proxy.py
+++ b/api/services/document_indexing_task_proxy.py
@@ -0,0 +1,83 @@
+import logging
+from collections.abc import Callable, Sequence
+from dataclasses import asdict
+from functools import cached_property
+
+from core.entities.document_task import DocumentTask
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+from enums.cloud_plan import CloudPlan
+from services.feature_service import FeatureService
+from tasks.document_indexing_task import normal_document_indexing_task, priority_document_indexing_task
+
+logger = logging.getLogger(__name__)
+
+
+class DocumentIndexingTaskProxy:
+    def __init__(self, tenant_id: str, dataset_id: str, document_ids: Sequence[str]):
+        self._tenant_id = tenant_id
+        self._dataset_id = dataset_id
+        self._document_ids = document_ids
+        self._tenant_isolated_task_queue = TenantIsolatedTaskQueue(tenant_id, "document_indexing")
+
+    @cached_property
+    def features(self):
+        return FeatureService.get_features(self._tenant_id)
+
+    def _send_to_direct_queue(self, task_func: Callable[[str, str, Sequence[str]], None]):
+        logger.info("send dataset %s to direct queue", self._dataset_id)
+        task_func.delay(  # type: ignore
+            tenant_id=self._tenant_id, dataset_id=self._dataset_id, document_ids=self._document_ids
+        )
+
+    def _send_to_tenant_queue(self, task_func: Callable[[str, str, Sequence[str]], None]):
+        logger.info("send dataset %s to tenant queue", self._dataset_id)
+        if self._tenant_isolated_task_queue.get_task_key():
+            # Add to waiting queue using List operations (lpush)
+            self._tenant_isolated_task_queue.push_tasks(
+                [
+                    asdict(
+                        DocumentTask(
+                            tenant_id=self._tenant_id, dataset_id=self._dataset_id, document_ids=self._document_ids
+                        )
+                    )
+                ]
+            )
+            logger.info("push tasks: %s - %s", self._dataset_id, self._document_ids)
+        else:
+            # Set flag and execute task
+            self._tenant_isolated_task_queue.set_task_waiting_time()
+            task_func.delay(  # type: ignore
+                tenant_id=self._tenant_id, dataset_id=self._dataset_id, document_ids=self._document_ids
+            )
+            logger.info("init tasks: %s - %s", self._dataset_id, self._document_ids)
+
+    def _send_to_default_tenant_queue(self):
+        self._send_to_tenant_queue(normal_document_indexing_task)
+
+    def _send_to_priority_tenant_queue(self):
+        self._send_to_tenant_queue(priority_document_indexing_task)
+
+    def _send_to_priority_direct_queue(self):
+        self._send_to_direct_queue(priority_document_indexing_task)
+
+    def _dispatch(self):
+        logger.info(
+            "dispatch args: %s - %s - %s",
+            self._tenant_id,
+            self.features.billing.enabled,
+            self.features.billing.subscription.plan,
+        )
+        # dispatch to different indexing queue with tenant isolation when billing enabled
+        if self.features.billing.enabled:
+            if self.features.billing.subscription.plan == CloudPlan.SANDBOX:
+                # dispatch to normal pipeline queue with tenant self sub queue for sandbox plan
+                self._send_to_default_tenant_queue()
+            else:
+                # dispatch to priority pipeline queue with tenant self sub queue for other plans
+                self._send_to_priority_tenant_queue()
+        else:
+            # dispatch to priority queue without tenant isolation for others, e.g.: self-hosted or enterprise
+            self._send_to_priority_direct_queue()
+
+    def delay(self):
+        self._dispatch()
--- a/api/services/rag_pipeline/rag_pipeline_task_proxy.py
+++ b/api/services/rag_pipeline/rag_pipeline_task_proxy.py
@@ -0,0 +1,106 @@
+import json
+import logging
+from collections.abc import Callable, Sequence
+from functools import cached_property
+
+from core.app.entities.rag_pipeline_invoke_entities import RagPipelineInvokeEntity
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+from enums.cloud_plan import CloudPlan
+from extensions.ext_database import db
+from services.feature_service import FeatureService
+from services.file_service import FileService
+from tasks.rag_pipeline.priority_rag_pipeline_run_task import priority_rag_pipeline_run_task
+from tasks.rag_pipeline.rag_pipeline_run_task import rag_pipeline_run_task
+
+logger = logging.getLogger(__name__)
+
+
+class RagPipelineTaskProxy:
+    # Default uploaded file name for rag pipeline invoke entities
+    _RAG_PIPELINE_INVOKE_ENTITIES_FILE_NAME = "rag_pipeline_invoke_entities.json"
+
+    def __init__(
+        self, dataset_tenant_id: str, user_id: str, rag_pipeline_invoke_entities: Sequence[RagPipelineInvokeEntity]
+    ):
+        self._dataset_tenant_id = dataset_tenant_id
+        self._user_id = user_id
+        self._rag_pipeline_invoke_entities = rag_pipeline_invoke_entities
+        self._tenant_isolated_task_queue = TenantIsolatedTaskQueue(dataset_tenant_id, "pipeline")
+
+    @cached_property
+    def features(self):
+        return FeatureService.get_features(self._dataset_tenant_id)
+
+    def _upload_invoke_entities(self) -> str:
+        text = [item.model_dump() for item in self._rag_pipeline_invoke_entities]
+        # Convert list to proper JSON string
+        json_text = json.dumps(text)
+        upload_file = FileService(db.engine).upload_text(
+            json_text, self._RAG_PIPELINE_INVOKE_ENTITIES_FILE_NAME, self._user_id, self._dataset_tenant_id
+        )
+        return upload_file.id
+
+    def _send_to_direct_queue(self, upload_file_id: str, task_func: Callable[[str, str], None]):
+        logger.info("send file %s to direct queue", upload_file_id)
+        task_func.delay(  # type: ignore
+            rag_pipeline_invoke_entities_file_id=upload_file_id,
+            tenant_id=self._dataset_tenant_id,
+        )
+
+    def _send_to_tenant_queue(self, upload_file_id: str, task_func: Callable[[str, str], None]):
+        logger.info("send file %s to tenant queue", upload_file_id)
+        if self._tenant_isolated_task_queue.get_task_key():
+            # Add to waiting queue using List operations (lpush)
+            self._tenant_isolated_task_queue.push_tasks([upload_file_id])
+            logger.info("push tasks: %s", upload_file_id)
+        else:
+            # Set flag and execute task
+            self._tenant_isolated_task_queue.set_task_waiting_time()
+            task_func.delay(  # type: ignore
+                rag_pipeline_invoke_entities_file_id=upload_file_id,
+                tenant_id=self._dataset_tenant_id,
+            )
+            logger.info("init tasks: %s", upload_file_id)
+
+    def _send_to_default_tenant_queue(self, upload_file_id: str):
+        self._send_to_tenant_queue(upload_file_id, rag_pipeline_run_task)
+
+    def _send_to_priority_tenant_queue(self, upload_file_id: str):
+        self._send_to_tenant_queue(upload_file_id, priority_rag_pipeline_run_task)
+
+    def _send_to_priority_direct_queue(self, upload_file_id: str):
+        self._send_to_direct_queue(upload_file_id, priority_rag_pipeline_run_task)
+
+    def _dispatch(self):
+        upload_file_id = self._upload_invoke_entities()
+        if not upload_file_id:
+            raise ValueError("upload_file_id is empty")
+
+        logger.info(
+            "dispatch args: %s - %s - %s",
+            self._dataset_tenant_id,
+            self.features.billing.enabled,
+            self.features.billing.subscription.plan,
+        )
+
+        # dispatch to different pipeline queue with tenant isolation when billing enabled
+        if self.features.billing.enabled:
+            if self.features.billing.subscription.plan == CloudPlan.SANDBOX:
+                # dispatch to normal pipeline queue with tenant isolation for sandbox plan
+                self._send_to_default_tenant_queue(upload_file_id)
+            else:
+                # dispatch to priority pipeline queue with tenant isolation for other plans
+                self._send_to_priority_tenant_queue(upload_file_id)
+        else:
+            # dispatch to priority pipeline queue without tenant isolation for others, e.g.: self-hosted or enterprise
+            self._send_to_priority_direct_queue(upload_file_id)
+
+    def delay(self):
+        if not self._rag_pipeline_invoke_entities:
+            logger.warning(
+                "Received empty rag pipeline invoke entities, no tasks delivered: %s %s",
+                self._dataset_tenant_id,
+                self._user_id,
+            )
+            return
+        self._dispatch()
--- a/api/tasks/document_indexing_task.py
+++ b/api/tasks/document_indexing_task.py
@@ -1,11 +1,14 @@
 import logging
 import time
+from collections.abc import Callable, Sequence

 import click
 from celery import shared_task

 from configs import dify_config
+from core.entities.document_task import DocumentTask
 from core.indexing_runner import DocumentIsPausedError, IndexingRunner
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
 from enums.cloud_plan import CloudPlan
 from extensions.ext_database import db
 from libs.datetime_utils import naive_utc_now
@@ -22,8 +25,24 @@ def document_indexing_task(dataset_id: str, document_ids: list):
    :param dataset_id:
    :param document_ids:

+    .. warning:: TO BE DEPRECATED
+        This function will be deprecated and removed in a future version.
+        Use normal_document_indexing_task or priority_document_indexing_task instead.
+
    Usage: document_indexing_task.delay(dataset_id, document_ids)
    """
+    logger.warning("document indexing legacy mode received: %s - %s", dataset_id, document_ids)
+    _document_indexing(dataset_id, document_ids)
+
+
+def _document_indexing(dataset_id: str, document_ids: Sequence[str]):
+    """
+    Process document for tasks
+    :param dataset_id:
+    :param document_ids:
+
+    Usage: _document_indexing(dataset_id, document_ids)
+    """
    documents = []
    start_at = time.perf_counter()

@@ -87,3 +106,63 @@ def document_indexing_task(dataset_id: str, document_ids: list):
        logger.exception("Document indexing task failed, dataset_id: %s", dataset_id)
    finally:
        db.session.close()
+
+
+def _document_indexing_with_tenant_queue(
+    tenant_id: str, dataset_id: str, document_ids: Sequence[str], task_func: Callable[[str, str, Sequence[str]], None]
+):
+    try:
+        _document_indexing(dataset_id, document_ids)
+    except Exception:
+        logger.exception("Error processing document indexing %s for tenant %s: %s", dataset_id, tenant_id)
+    finally:
+        tenant_isolated_task_queue = TenantIsolatedTaskQueue(tenant_id, "document_indexing")
+
+        # Check if there are waiting tasks in the queue
+        # Use rpop to get the next task from the queue (FIFO order)
+        next_tasks = tenant_isolated_task_queue.pull_tasks(count=dify_config.TENANT_ISOLATED_TASK_CONCURRENCY)
+
+        logger.info("document indexing tenant isolation queue next tasks: %s", next_tasks)
+
+        if next_tasks:
+            for next_task in next_tasks:
+                document_task = DocumentTask(**next_task)
+                # Process the next waiting task
+                # Keep the flag set to indicate a task is running
+                tenant_isolated_task_queue.set_task_waiting_time()
+                task_func.delay(  # type: ignore
+                    tenant_id=document_task.tenant_id,
+                    dataset_id=document_task.dataset_id,
+                    document_ids=document_task.document_ids,
+                )
+        else:
+            # No more waiting tasks, clear the flag
+            tenant_isolated_task_queue.delete_task_key()
+
+
+@shared_task(queue="dataset")
+def normal_document_indexing_task(tenant_id: str, dataset_id: str, document_ids: Sequence[str]):
+    """
+    Async process document
+    :param tenant_id:
+    :param dataset_id:
+    :param document_ids:
+
+    Usage: normal_document_indexing_task.delay(tenant_id, dataset_id, document_ids)
+    """
+    logger.info("normal document indexing task received: %s - %s - %s", tenant_id, dataset_id, document_ids)
+    _document_indexing_with_tenant_queue(tenant_id, dataset_id, document_ids, normal_document_indexing_task)
+
+
+@shared_task(queue="priority_dataset")
+def priority_document_indexing_task(tenant_id: str, dataset_id: str, document_ids: Sequence[str]):
+    """
+    Priority async process document
+    :param tenant_id:
+    :param dataset_id:
+    :param document_ids:
+
+    Usage: priority_document_indexing_task.delay(tenant_id, dataset_id, document_ids)
+    """
+    logger.info("priority document indexing task received: %s - %s - %s", tenant_id, dataset_id, document_ids)
+    _document_indexing_with_tenant_queue(tenant_id, dataset_id, document_ids, priority_document_indexing_task)
--- a/api/tasks/rag_pipeline/priority_rag_pipeline_run_task.py
+++ b/api/tasks/rag_pipeline/priority_rag_pipeline_run_task.py
@@ -12,8 +12,10 @@ from celery import shared_task  # type: ignore
 from flask import current_app, g
 from sqlalchemy.orm import Session, sessionmaker

+from configs import dify_config
 from core.app.entities.app_invoke_entities import InvokeFrom, RagPipelineGenerateEntity
 from core.app.entities.rag_pipeline_invoke_entities import RagPipelineInvokeEntity
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
 from core.repositories.factory import DifyCoreRepositoryFactory
 from extensions.ext_database import db
 from models import Account, Tenant
@@ -22,6 +24,8 @@ from models.enums import WorkflowRunTriggeredFrom
 from models.workflow import Workflow, WorkflowNodeExecutionTriggeredFrom
 from services.file_service import FileService

+logger = logging.getLogger(__name__)
+

@shared_task(queue="priority_pipeline")
 def priority_rag_pipeline_run_task(
@@ -69,6 +73,27 @@ def priority_rag_pipeline_run_task(
        logging.exception(click.style(f"Error running rag pipeline, tenant_id: {tenant_id}", fg="red"))
        raise
    finally:
+        tenant_isolated_task_queue = TenantIsolatedTaskQueue(tenant_id, "pipeline")
+
+        # Check if there are waiting tasks in the queue
+        # Use rpop to get the next task from the queue (FIFO order)
+        next_file_ids = tenant_isolated_task_queue.pull_tasks(count=dify_config.TENANT_ISOLATED_TASK_CONCURRENCY)
+        logger.info("priority rag pipeline tenant isolation queue next files: %s", next_file_ids)
+
+        if next_file_ids:
+            for next_file_id in next_file_ids:
+                # Process the next waiting task
+                # Keep the flag set to indicate a task is running
+                tenant_isolated_task_queue.set_task_waiting_time()
+                priority_rag_pipeline_run_task.delay(  # type: ignore
+                    rag_pipeline_invoke_entities_file_id=next_file_id.decode("utf-8")
+                    if isinstance(next_file_id, bytes)
+                    else next_file_id,
+                    tenant_id=tenant_id,
+                )
+        else:
+            # No more waiting tasks, clear the flag
+            tenant_isolated_task_queue.delete_task_key()
        file_service = FileService(db.engine)
        file_service.delete_file(rag_pipeline_invoke_entities_file_id)
        db.session.close()
--- a/api/tasks/rag_pipeline/rag_pipeline_run_task.py
+++ b/api/tasks/rag_pipeline/rag_pipeline_run_task.py
@@ -12,17 +12,20 @@ from celery import shared_task  # type: ignore
 from flask import current_app, g
 from sqlalchemy.orm import Session, sessionmaker

+from configs import dify_config
 from core.app.entities.app_invoke_entities import InvokeFrom, RagPipelineGenerateEntity
 from core.app.entities.rag_pipeline_invoke_entities import RagPipelineInvokeEntity
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
 from core.repositories.factory import DifyCoreRepositoryFactory
 from extensions.ext_database import db
-from extensions.ext_redis import redis_client
 from models import Account, Tenant
 from models.dataset import Pipeline
 from models.enums import WorkflowRunTriggeredFrom
 from models.workflow import Workflow, WorkflowNodeExecutionTriggeredFrom
 from services.file_service import FileService

+logger = logging.getLogger(__name__)
+

@shared_task(queue="pipeline")
 def rag_pipeline_run_task(
@@ -70,17 +73,18 @@ def rag_pipeline_run_task(
        logging.exception(click.style(f"Error running rag pipeline, tenant_id: {tenant_id}", fg="red"))
        raise
    finally:
-        tenant_self_pipeline_task_queue = f"tenant_self_pipeline_task_queue:{tenant_id}"
-        tenant_pipeline_task_key = f"tenant_pipeline_task:{tenant_id}"
+        tenant_isolated_task_queue = TenantIsolatedTaskQueue(tenant_id, "pipeline")

        # Check if there are waiting tasks in the queue
        # Use rpop to get the next task from the queue (FIFO order)
-        next_file_id = redis_client.rpop(tenant_self_pipeline_task_queue)
+        next_file_ids = tenant_isolated_task_queue.pull_tasks(count=dify_config.TENANT_ISOLATED_TASK_CONCURRENCY)
+        logger.info("rag pipeline tenant isolation queue next files: %s", next_file_ids)

-        if next_file_id:
+        if next_file_ids:
+            for next_file_id in next_file_ids:
                # Process the next waiting task
                # Keep the flag set to indicate a task is running
-            redis_client.setex(tenant_pipeline_task_key, 60 * 60, 1)
+                tenant_isolated_task_queue.set_task_waiting_time()
                rag_pipeline_run_task.delay(  # type: ignore
                    rag_pipeline_invoke_entities_file_id=next_file_id.decode("utf-8")
                    if isinstance(next_file_id, bytes)
@@ -89,7 +93,7 @@ def rag_pipeline_run_task(
                )
        else:
            # No more waiting tasks, clear the flag
-            redis_client.delete(tenant_pipeline_task_key)
+            tenant_isolated_task_queue.delete_task_key()
        file_service = FileService(db.engine)
        file_service.delete_file(rag_pipeline_invoke_entities_file_id)
        db.session.close()
--- a/api/tests/test_containers_integration_tests/core/rag/init.py
+++ b/api/tests/test_containers_integration_tests/core/rag/init.py
@@ -0,0 +1 @@
+
--- a/api/tests/test_containers_integration_tests/core/rag/pipeline/init.py
+++ b/api/tests/test_containers_integration_tests/core/rag/pipeline/init.py
--- a/api/tests/test_containers_integration_tests/core/rag/pipeline/test_queue_integration.py
+++ b/api/tests/test_containers_integration_tests/core/rag/pipeline/test_queue_integration.py
@@ -0,0 +1,595 @@
+"""
+Integration tests for TenantIsolatedTaskQueue using testcontainers.
+
+These tests verify the Redis-based task queue functionality with real Redis instances,
+testing tenant isolation, task serialization, and queue operations in a realistic environment.
+Includes compatibility tests for migrating from legacy string-only queues.
+
+All tests use generic naming to avoid coupling to specific business implementations.
+"""
+
+import time
+from dataclasses import dataclass
+from typing import Any
+from uuid import uuid4
+
+import pytest
+from faker import Faker
+
+from core.rag.pipeline.queue import TaskWrapper, TenantIsolatedTaskQueue
+from extensions.ext_redis import redis_client
+from models import Account, Tenant, TenantAccountJoin, TenantAccountRole
+
+
+@dataclass
+class TestTask:
+    """Test task data structure for testing complex object serialization."""
+
+    task_id: str
+    tenant_id: str
+    data: dict[str, Any]
+    metadata: dict[str, Any]
+
+
+class TestTenantIsolatedTaskQueueIntegration:
+    """Integration tests for TenantIsolatedTaskQueue using testcontainers."""
+
+    @pytest.fixture
+    def fake(self):
+        """Faker instance for generating test data."""
+        return Faker()
+
+    @pytest.fixture
+    def test_tenant_and_account(self, db_session_with_containers, fake):
+        """Create test tenant and account for testing."""
+        # Create account
+        account = Account(
+            email=fake.email(),
+            name=fake.name(),
+            interface_language="en-US",
+            status="active",
+        )
+        db_session_with_containers.add(account)
+        db_session_with_containers.commit()
+
+        # Create tenant
+        tenant = Tenant(
+            name=fake.company(),
+            status="normal",
+        )
+        db_session_with_containers.add(tenant)
+        db_session_with_containers.commit()
+
+        # Create tenant-account join
+        join = TenantAccountJoin(
+            tenant_id=tenant.id,
+            account_id=account.id,
+            role=TenantAccountRole.OWNER,
+            current=True,
+        )
+        db_session_with_containers.add(join)
+        db_session_with_containers.commit()
+
+        return tenant, account
+
+    @pytest.fixture
+    def test_queue(self, test_tenant_and_account):
+        """Create a generic test queue for testing."""
+        tenant, _ = test_tenant_and_account
+        return TenantIsolatedTaskQueue(tenant.id, "test_queue")
+
+    @pytest.fixture
+    def secondary_queue(self, test_tenant_and_account):
+        """Create a secondary test queue for testing isolation."""
+        tenant, _ = test_tenant_and_account
+        return TenantIsolatedTaskQueue(tenant.id, "secondary_queue")
+
+    def test_queue_initialization(self, test_tenant_and_account):
+        """Test queue initialization with correct key generation."""
+        tenant, _ = test_tenant_and_account
+        queue = TenantIsolatedTaskQueue(tenant.id, "test-key")
+
+        assert queue._tenant_id == tenant.id
+        assert queue._unique_key == "test-key"
+        assert queue._queue == f"tenant_self_test-key_task_queue:{tenant.id}"
+        assert queue._task_key == f"tenant_test-key_task:{tenant.id}"
+
+    def test_tenant_isolation(self, test_tenant_and_account, db_session_with_containers, fake):
+        """Test that different tenants have isolated queues."""
+        tenant1, _ = test_tenant_and_account
+
+        # Create second tenant
+        tenant2 = Tenant(
+            name=fake.company(),
+            status="normal",
+        )
+        db_session_with_containers.add(tenant2)
+        db_session_with_containers.commit()
+
+        queue1 = TenantIsolatedTaskQueue(tenant1.id, "same-key")
+        queue2 = TenantIsolatedTaskQueue(tenant2.id, "same-key")
+
+        assert queue1._queue != queue2._queue
+        assert queue1._task_key != queue2._task_key
+        assert queue1._queue == f"tenant_self_same-key_task_queue:{tenant1.id}"
+        assert queue2._queue == f"tenant_self_same-key_task_queue:{tenant2.id}"
+
+    def test_key_isolation(self, test_tenant_and_account):
+        """Test that different keys have isolated queues."""
+        tenant, _ = test_tenant_and_account
+        queue1 = TenantIsolatedTaskQueue(tenant.id, "key1")
+        queue2 = TenantIsolatedTaskQueue(tenant.id, "key2")
+
+        assert queue1._queue != queue2._queue
+        assert queue1._task_key != queue2._task_key
+        assert queue1._queue == f"tenant_self_key1_task_queue:{tenant.id}"
+        assert queue2._queue == f"tenant_self_key2_task_queue:{tenant.id}"
+
+    def test_task_key_operations(self, test_queue):
+        """Test task key operations (get, set, delete)."""
+        # Initially no task key should exist
+        assert test_queue.get_task_key() is None
+
+        # Set task waiting time with default TTL
+        test_queue.set_task_waiting_time()
+        task_key = test_queue.get_task_key()
+        # Redis returns bytes, convert to string for comparison
+        assert task_key in (b"1", "1")
+
+        # Set task waiting time with custom TTL
+        custom_ttl = 30
+        test_queue.set_task_waiting_time(custom_ttl)
+        task_key = test_queue.get_task_key()
+        assert task_key in (b"1", "1")
+
+        # Delete task key
+        test_queue.delete_task_key()
+        assert test_queue.get_task_key() is None
+
+    def test_push_and_pull_string_tasks(self, test_queue):
+        """Test pushing and pulling string tasks."""
+        tasks = ["task1", "task2", "task3"]
+
+        # Push tasks
+        test_queue.push_tasks(tasks)
+
+        # Pull tasks (FIFO order)
+        pulled_tasks = test_queue.pull_tasks(3)
+
+        # Should get tasks in FIFO order (lpush + rpop = FIFO)
+        assert pulled_tasks == ["task1", "task2", "task3"]
+
+    def test_push_and_pull_multiple_tasks(self, test_queue):
+        """Test pushing and pulling multiple tasks at once."""
+        tasks = ["task1", "task2", "task3", "task4", "task5"]
+
+        # Push tasks
+        test_queue.push_tasks(tasks)
+
+        # Pull multiple tasks
+        pulled_tasks = test_queue.pull_tasks(3)
+        assert len(pulled_tasks) == 3
+        assert pulled_tasks == ["task1", "task2", "task3"]
+
+        # Pull remaining tasks
+        remaining_tasks = test_queue.pull_tasks(5)
+        assert len(remaining_tasks) == 2
+        assert remaining_tasks == ["task4", "task5"]
+
+    def test_push_and_pull_complex_objects(self, test_queue, fake):
+        """Test pushing and pulling complex object tasks."""
+        # Create complex task objects as dictionaries (not dataclass instances)
+        tasks = [
+            {
+                "task_id": str(uuid4()),
+                "tenant_id": test_queue._tenant_id,
+                "data": {
+                    "file_id": str(uuid4()),
+                    "content": fake.text(),
+                    "metadata": {"size": fake.random_int(1000, 10000)},
+                },
+                "metadata": {"created_at": fake.iso8601(), "tags": fake.words(3)},
+            },
+            {
+                "task_id": str(uuid4()),
+                "tenant_id": test_queue._tenant_id,
+                "data": {
+                    "file_id": str(uuid4()),
+                    "content": "测试中文内容",
+                    "metadata": {"size": fake.random_int(1000, 10000)},
+                },
+                "metadata": {"created_at": fake.iso8601(), "tags": ["中文", "测试", "emoji🚀"]},
+            },
+        ]
+
+        # Push complex tasks
+        test_queue.push_tasks(tasks)
+
+        # Pull tasks
+        pulled_tasks = test_queue.pull_tasks(2)
+        assert len(pulled_tasks) == 2
+
+        # Verify deserialized tasks match original (FIFO order)
+        for i, pulled_task in enumerate(pulled_tasks):
+            original_task = tasks[i]  # FIFO order
+            assert isinstance(pulled_task, dict)
+            assert pulled_task["task_id"] == original_task["task_id"]
+            assert pulled_task["tenant_id"] == original_task["tenant_id"]
+            assert pulled_task["data"] == original_task["data"]
+            assert pulled_task["metadata"] == original_task["metadata"]
+
+    def test_mixed_task_types(self, test_queue, fake):
+        """Test pushing and pulling mixed string and object tasks."""
+        string_task = "simple_string_task"
+        object_task = {
+            "task_id": str(uuid4()),
+            "dataset_id": str(uuid4()),
+            "document_ids": [str(uuid4()) for _ in range(3)],
+        }
+
+        tasks = [string_task, object_task, "another_string"]
+
+        # Push mixed tasks
+        test_queue.push_tasks(tasks)
+
+        # Pull all tasks
+        pulled_tasks = test_queue.pull_tasks(3)
+        assert len(pulled_tasks) == 3
+
+        # Verify types and content
+        assert pulled_tasks[0] == string_task
+        assert isinstance(pulled_tasks[1], dict)
+        assert pulled_tasks[1] == object_task
+        assert pulled_tasks[2] == "another_string"
+
+    def test_empty_queue_operations(self, test_queue):
+        """Test operations on empty queue."""
+        # Pull from empty queue
+        tasks = test_queue.pull_tasks(5)
+        assert tasks == []
+
+        # Pull zero or negative count
+        assert test_queue.pull_tasks(0) == []
+        assert test_queue.pull_tasks(-1) == []
+
+    def test_task_ttl_expiration(self, test_queue):
+        """Test task key TTL expiration."""
+        # Set task with short TTL
+        short_ttl = 2
+        test_queue.set_task_waiting_time(short_ttl)
+
+        # Verify task key exists
+        assert test_queue.get_task_key() == b"1" or test_queue.get_task_key() == "1"
+
+        # Wait for TTL to expire
+        time.sleep(short_ttl + 1)
+
+        # Verify task key has expired
+        assert test_queue.get_task_key() is None
+
+    def test_large_task_batch(self, test_queue, fake):
+        """Test handling large batches of tasks."""
+        # Create large batch of tasks
+        large_batch = []
+        for i in range(100):
+            task = {
+                "task_id": str(uuid4()),
+                "index": i,
+                "data": fake.text(max_nb_chars=100),
+                "metadata": {"batch_id": str(uuid4())},
+            }
+            large_batch.append(task)
+
+        # Push large batch
+        test_queue.push_tasks(large_batch)
+
+        # Pull all tasks
+        pulled_tasks = test_queue.pull_tasks(100)
+        assert len(pulled_tasks) == 100
+
+        # Verify all tasks were retrieved correctly (FIFO order)
+        for i, task in enumerate(pulled_tasks):
+            assert isinstance(task, dict)
+            assert task["index"] == i  # FIFO order
+
+    def test_queue_operations_isolation(self, test_tenant_and_account, fake):
+        """Test concurrent operations on different queues."""
+        tenant, _ = test_tenant_and_account
+
+        # Create multiple queues for the same tenant
+        queue1 = TenantIsolatedTaskQueue(tenant.id, "queue1")
+        queue2 = TenantIsolatedTaskQueue(tenant.id, "queue2")
+
+        # Push tasks to different queues
+        queue1.push_tasks(["task1_queue1", "task2_queue1"])
+        queue2.push_tasks(["task1_queue2", "task2_queue2"])
+
+        # Verify queues are isolated
+        tasks1 = queue1.pull_tasks(2)
+        tasks2 = queue2.pull_tasks(2)
+
+        assert tasks1 == ["task1_queue1", "task2_queue1"]
+        assert tasks2 == ["task1_queue2", "task2_queue2"]
+        assert tasks1 != tasks2
+
+    def test_task_wrapper_serialization_roundtrip(self, test_queue, fake):
+        """Test TaskWrapper serialization and deserialization roundtrip."""
+        # Create complex nested data
+        complex_data = {
+            "id": str(uuid4()),
+            "nested": {"deep": {"value": "test", "numbers": [1, 2, 3, 4, 5], "unicode": "测试中文", "emoji": "🚀"}},
+            "metadata": {"created_at": fake.iso8601(), "tags": ["tag1", "tag2", "tag3"]},
+        }
+
+        # Create wrapper and serialize
+        wrapper = TaskWrapper(data=complex_data)
+        serialized = wrapper.serialize()
+
+        # Verify serialization
+        assert isinstance(serialized, str)
+        assert "测试中文" in serialized
+        assert "🚀" in serialized
+
+        # Deserialize and verify
+        deserialized_wrapper = TaskWrapper.deserialize(serialized)
+        assert deserialized_wrapper.data == complex_data
+
+    def test_error_handling_invalid_json(self, test_queue):
+        """Test error handling for invalid JSON in wrapped tasks."""
+        # Manually create invalid JSON task (not a valid TaskWrapper JSON)
+        invalid_json_task = "invalid json data"
+
+        # Push invalid task directly to Redis
+        redis_client.lpush(test_queue._queue, invalid_json_task)
+
+        # Pull task - should fall back to string since it's not valid JSON
+        task = test_queue.pull_tasks(1)
+        assert task[0] == invalid_json_task
+
+    def test_real_world_batch_processing_scenario(self, test_queue, fake):
+        """Test realistic batch processing scenario."""
+        # Simulate batch processing tasks
+        batch_tasks = []
+        for i in range(3):
+            task = {
+                "file_id": str(uuid4()),
+                "tenant_id": test_queue._tenant_id,
+                "user_id": str(uuid4()),
+                "processing_config": {
+                    "model": fake.random_element(["model_a", "model_b", "model_c"]),
+                    "temperature": fake.random.uniform(0.1, 1.0),
+                    "max_tokens": fake.random_int(1000, 4000),
+                },
+                "metadata": {
+                    "source": fake.random_element(["upload", "api", "webhook"]),
+                    "priority": fake.random_element(["low", "normal", "high"]),
+                },
+            }
+            batch_tasks.append(task)
+
+        # Push tasks
+        test_queue.push_tasks(batch_tasks)
+
+        # Process tasks in batches
+        batch_size = 2
+        processed_tasks = []
+
+        while True:
+            batch = test_queue.pull_tasks(batch_size)
+            if not batch:
+                break
+
+            processed_tasks.extend(batch)
+
+        # Verify all tasks were processed
+        assert len(processed_tasks) == 3
+
+        # Verify task structure
+        for task in processed_tasks:
+            assert isinstance(task, dict)
+            assert "file_id" in task
+            assert "tenant_id" in task
+            assert "processing_config" in task
+            assert "metadata" in task
+            assert task["tenant_id"] == test_queue._tenant_id
+
+
+class TestTenantIsolatedTaskQueueCompatibility:
+    """Compatibility tests for migrating from legacy string-only queues."""
+
+    @pytest.fixture
+    def fake(self):
+        """Faker instance for generating test data."""
+        return Faker()
+
+    @pytest.fixture
+    def test_tenant_and_account(self, db_session_with_containers, fake):
+        """Create test tenant and account for testing."""
+        # Create account
+        account = Account(
+            email=fake.email(),
+            name=fake.name(),
+            interface_language="en-US",
+            status="active",
+        )
+        db_session_with_containers.add(account)
+        db_session_with_containers.commit()
+
+        # Create tenant
+        tenant = Tenant(
+            name=fake.company(),
+            status="normal",
+        )
+        db_session_with_containers.add(tenant)
+        db_session_with_containers.commit()
+
+        # Create tenant-account join
+        join = TenantAccountJoin(
+            tenant_id=tenant.id,
+            account_id=account.id,
+            role=TenantAccountRole.OWNER,
+            current=True,
+        )
+        db_session_with_containers.add(join)
+        db_session_with_containers.commit()
+
+        return tenant, account
+
+    def test_legacy_string_queue_compatibility(self, test_tenant_and_account, fake):
+        """
+        Test compatibility with legacy queues containing only string data.
+
+        This simulates the scenario where Redis queues already contain string data
+        from the old architecture, and we need to ensure the new code can read them.
+        """
+        tenant, _ = test_tenant_and_account
+        queue = TenantIsolatedTaskQueue(tenant.id, "legacy_queue")
+
+        # Simulate legacy string data in Redis queue (using old format)
+        legacy_strings = ["legacy_task_1", "legacy_task_2", "legacy_task_3", "legacy_task_4", "legacy_task_5"]
+
+        # Manually push legacy strings directly to Redis (simulating old system)
+        for legacy_string in legacy_strings:
+            redis_client.lpush(queue._queue, legacy_string)
+
+        # Verify new code can read legacy string data
+        pulled_tasks = queue.pull_tasks(5)
+        assert len(pulled_tasks) == 5
+
+        # Verify all tasks are strings (not wrapped)
+        for task in pulled_tasks:
+            assert isinstance(task, str)
+            assert task.startswith("legacy_task_")
+
+        # Verify order (FIFO from Redis list)
+        expected_order = ["legacy_task_1", "legacy_task_2", "legacy_task_3", "legacy_task_4", "legacy_task_5"]
+        assert pulled_tasks == expected_order
+
+    def test_legacy_queue_migration_scenario(self, test_tenant_and_account, fake):
+        """
+        Test complete migration scenario from legacy to new system.
+
+        This simulates the real-world scenario where:
+        1. Legacy system has string data in Redis
+        2. New system starts processing the same queue
+        3. Both legacy and new tasks coexist during migration
+        4. New system can handle both formats seamlessly
+        """
+        tenant, _ = test_tenant_and_account
+        queue = TenantIsolatedTaskQueue(tenant.id, "migration_queue")
+
+        # Phase 1: Legacy system has data
+        legacy_tasks = [f"legacy_resource_{i}" for i in range(1, 6)]
+        redis_client.lpush(queue._queue, *legacy_tasks)
+
+        # Phase 2: New system starts processing legacy data
+        processed_legacy = []
+        while True:
+            tasks = queue.pull_tasks(1)
+            if not tasks:
+                break
+            processed_legacy.extend(tasks)
+
+        # Verify legacy data was processed correctly
+        assert len(processed_legacy) == 5
+        for task in processed_legacy:
+            assert isinstance(task, str)
+            assert task.startswith("legacy_resource_")
+
+        # Phase 3: New system adds new tasks (mixed types)
+        new_string_tasks = ["new_resource_1", "new_resource_2"]
+        new_object_tasks = [
+            {
+                "resource_id": str(uuid4()),
+                "tenant_id": tenant.id,
+                "processing_type": "new_system",
+                "metadata": {"version": "2.0", "features": ["ai", "ml"]},
+            },
+            {
+                "resource_id": str(uuid4()),
+                "tenant_id": tenant.id,
+                "processing_type": "new_system",
+                "metadata": {"version": "2.0", "features": ["ai", "ml"]},
+            },
+        ]
+
+        # Push new tasks using new system
+        queue.push_tasks(new_string_tasks)
+        queue.push_tasks(new_object_tasks)
+
+        # Phase 4: Process all new tasks
+        processed_new = []
+        while True:
+            tasks = queue.pull_tasks(1)
+            if not tasks:
+                break
+            processed_new.extend(tasks)
+
+        # Verify new tasks were processed correctly
+        assert len(processed_new) == 4
+
+        string_tasks = [task for task in processed_new if isinstance(task, str)]
+        object_tasks = [task for task in processed_new if isinstance(task, dict)]
+
+        assert len(string_tasks) == 2
+        assert len(object_tasks) == 2
+
+        # Verify string tasks
+        for task in string_tasks:
+            assert task.startswith("new_resource_")
+
+        # Verify object tasks
+        for task in object_tasks:
+            assert isinstance(task, dict)
+            assert "resource_id" in task
+            assert "tenant_id" in task
+            assert task["tenant_id"] == tenant.id
+            assert task["processing_type"] == "new_system"
+
+    def test_legacy_queue_error_recovery(self, test_tenant_and_account, fake):
+        """
+        Test error recovery when legacy queue contains malformed data.
+
+        This ensures the new system can gracefully handle corrupted or
+        malformed legacy data without crashing.
+        """
+        tenant, _ = test_tenant_and_account
+        queue = TenantIsolatedTaskQueue(tenant.id, "error_recovery_queue")
+
+        # Create mix of valid and malformed legacy data
+        mixed_legacy_data = [
+            "valid_legacy_task_1",
+            "valid_legacy_task_2",
+            "malformed_data_string",  # This should be treated as string
+            "valid_legacy_task_3",
+            "invalid_json_not_taskwrapper_format",  # This should fall back to string (not valid TaskWrapper JSON)
+            "valid_legacy_task_4",
+        ]
+
+        # Manually push mixed data directly to Redis
+        redis_client.lpush(queue._queue, *mixed_legacy_data)
+
+        # Process all tasks
+        processed_tasks = []
+        while True:
+            tasks = queue.pull_tasks(1)
+            if not tasks:
+                break
+            processed_tasks.extend(tasks)
+
+        # Verify all tasks were processed (no crashes)
+        assert len(processed_tasks) == 6
+
+        # Verify all tasks are strings (malformed data falls back to string)
+        for task in processed_tasks:
+            assert isinstance(task, str)
+
+        # Verify valid tasks are preserved
+        valid_tasks = [task for task in processed_tasks if task.startswith("valid_legacy_task_")]
+        assert len(valid_tasks) == 4
+
+        # Verify malformed data is handled gracefully
+        malformed_tasks = [task for task in processed_tasks if not task.startswith("valid_legacy_task_")]
+        assert len(malformed_tasks) == 2
+        assert "malformed_data_string" in malformed_tasks
+        assert "invalid_json_not_taskwrapper_format" in malformed_tasks
--- a/api/tests/test_containers_integration_tests/tasks/test_document_indexing_task.py
+++ b/api/tests/test_containers_integration_tests/tasks/test_document_indexing_task.py
@@ -1,17 +1,33 @@
+from dataclasses import asdict
 from unittest.mock import MagicMock, patch

 import pytest
 from faker import Faker

+from core.entities.document_task import DocumentTask
 from enums.cloud_plan import CloudPlan
 from extensions.ext_database import db
 from models import Account, Tenant, TenantAccountJoin, TenantAccountRole
 from models.dataset import Dataset, Document
-from tasks.document_indexing_task import document_indexing_task
+from tasks.document_indexing_task import (
+    _document_indexing,  # Core function
+    _document_indexing_with_tenant_queue,  # Tenant queue wrapper function
+    document_indexing_task,  # Deprecated old interface
+    normal_document_indexing_task,  # New normal task
+    priority_document_indexing_task,  # New priority task
+)


-class TestDocumentIndexingTask:
-    """Integration tests for document_indexing_task using testcontainers."""
+class TestDocumentIndexingTasks:
+    """Integration tests for document indexing tasks using testcontainers.
+
+    This test class covers:
+    - Core _document_indexing function
+    - Deprecated document_indexing_task function
+    - New normal_document_indexing_task function
+    - New priority_document_indexing_task function
+    - Tenant queue wrapper _document_indexing_with_tenant_queue function
+    """

    @pytest.fixture
    def mock_external_service_dependencies(self):
@@ -224,7 +240,7 @@ class TestDocumentIndexingTask:
        document_ids = [doc.id for doc in documents]

        # Act: Execute the task
-        document_indexing_task(dataset.id, document_ids)
+        _document_indexing(dataset.id, document_ids)

        # Assert: Verify the expected outcomes
        # Verify indexing runner was called correctly
@@ -232,10 +248,11 @@ class TestDocumentIndexingTask:
        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()

        # Verify documents were updated to parsing status
-        for document in documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "parsing"
-            assert document.processing_started_at is not None
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None

        # Verify the run method was called with correct documents
        call_args = mock_external_service_dependencies["indexing_runner_instance"].run.call_args
@@ -261,7 +278,7 @@ class TestDocumentIndexingTask:
        document_ids = [fake.uuid4() for _ in range(3)]

        # Act: Execute the task with non-existent dataset
-        document_indexing_task(non_existent_dataset_id, document_ids)
+        _document_indexing(non_existent_dataset_id, document_ids)

        # Assert: Verify no processing occurred
        mock_external_service_dependencies["indexing_runner"].assert_not_called()
@@ -291,17 +308,18 @@ class TestDocumentIndexingTask:
        all_document_ids = existing_document_ids + non_existent_document_ids

        # Act: Execute the task with mixed document IDs
-        document_indexing_task(dataset.id, all_document_ids)
+        _document_indexing(dataset.id, all_document_ids)

        # Assert: Verify only existing documents were processed
        mock_external_service_dependencies["indexing_runner"].assert_called_once()
        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()

        # Verify only existing documents were updated
-        for document in documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "parsing"
-            assert document.processing_started_at is not None
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in existing_document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None

        # Verify the run method was called with only existing documents
        call_args = mock_external_service_dependencies["indexing_runner_instance"].run.call_args
@@ -333,7 +351,7 @@ class TestDocumentIndexingTask:
        )

        # Act: Execute the task
-        document_indexing_task(dataset.id, document_ids)
+        _document_indexing(dataset.id, document_ids)

        # Assert: Verify exception was handled gracefully
        # The task should complete without raising exceptions
@@ -341,10 +359,11 @@ class TestDocumentIndexingTask:
        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()

        # Verify documents were still updated to parsing status before the exception
-        for document in documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "parsing"
-            assert document.processing_started_at is not None
+        # Re-query documents from database since _document_indexing close the session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None

    def test_document_indexing_task_mixed_document_states(
        self, db_session_with_containers, mock_external_service_dependencies
@@ -407,17 +426,18 @@ class TestDocumentIndexingTask:
        document_ids = [doc.id for doc in all_documents]

        # Act: Execute the task with mixed document states
-        document_indexing_task(dataset.id, document_ids)
+        _document_indexing(dataset.id, document_ids)

        # Assert: Verify processing
        mock_external_service_dependencies["indexing_runner"].assert_called_once()
        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()

        # Verify all documents were updated to parsing status
-        for document in all_documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "parsing"
-            assert document.processing_started_at is not None
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None

        # Verify the run method was called with all documents
        call_args = mock_external_service_dependencies["indexing_runner_instance"].run.call_args
@@ -470,15 +490,16 @@ class TestDocumentIndexingTask:
        document_ids = [doc.id for doc in all_documents]

        # Act: Execute the task with too many documents for sandbox plan
-        document_indexing_task(dataset.id, document_ids)
+        _document_indexing(dataset.id, document_ids)

        # Assert: Verify error handling
-        for document in all_documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "error"
-            assert document.error is not None
-            assert "batch upload" in document.error
-            assert document.stopped_at is not None
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "error"
+            assert updated_document.error is not None
+            assert "batch upload" in updated_document.error
+            assert updated_document.stopped_at is not None

        # Verify no indexing runner was called
        mock_external_service_dependencies["indexing_runner"].assert_not_called()
@@ -503,17 +524,18 @@ class TestDocumentIndexingTask:
        document_ids = [doc.id for doc in documents]

        # Act: Execute the task with billing disabled
-        document_indexing_task(dataset.id, document_ids)
+        _document_indexing(dataset.id, document_ids)

        # Assert: Verify successful processing
        mock_external_service_dependencies["indexing_runner"].assert_called_once()
        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()

        # Verify documents were updated to parsing status
-        for document in documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "parsing"
-            assert document.processing_started_at is not None
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None

    def test_document_indexing_task_document_is_paused_error(
        self, db_session_with_containers, mock_external_service_dependencies
@@ -541,7 +563,7 @@ class TestDocumentIndexingTask:
        )

        # Act: Execute the task
-        document_indexing_task(dataset.id, document_ids)
+        _document_indexing(dataset.id, document_ids)

        # Assert: Verify exception was handled gracefully
        # The task should complete without raising exceptions
@@ -549,7 +571,317 @@ class TestDocumentIndexingTask:
        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()

        # Verify documents were still updated to parsing status before the exception
-        for document in documents:
-            db.session.refresh(document)
-            assert document.indexing_status == "parsing"
-            assert document.processing_started_at is not None
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None
+
+    # ==================== NEW TESTS FOR REFACTORED FUNCTIONS ====================
+    def test_old_document_indexing_task_success(self, db_session_with_containers, mock_external_service_dependencies):
+        """
+        Test document_indexing_task basic functionality.
+
+        This test verifies:
+        - Task function calls the wrapper correctly
+        - Basic parameter passing works
+        """
+        # Arrange: Create test data
+        dataset, documents = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+        document_ids = [doc.id for doc in documents]
+
+        # Act: Execute the deprecated task (it only takes 2 parameters)
+        document_indexing_task(dataset.id, document_ids)
+
+        # Assert: Verify processing occurred (core logic is tested in _document_indexing tests)
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+    def test_normal_document_indexing_task_success(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test normal_document_indexing_task basic functionality.
+
+        This test verifies:
+        - Task function calls the wrapper correctly
+        - Basic parameter passing works
+        """
+        # Arrange: Create test data
+        dataset, documents = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+        document_ids = [doc.id for doc in documents]
+        tenant_id = dataset.tenant_id
+
+        # Act: Execute the new normal task
+        normal_document_indexing_task(tenant_id, dataset.id, document_ids)
+
+        # Assert: Verify processing occurred (core logic is tested in _document_indexing tests)
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+    def test_priority_document_indexing_task_success(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test priority_document_indexing_task basic functionality.
+
+        This test verifies:
+        - Task function calls the wrapper correctly
+        - Basic parameter passing works
+        """
+        # Arrange: Create test data
+        dataset, documents = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+        document_ids = [doc.id for doc in documents]
+        tenant_id = dataset.tenant_id
+
+        # Act: Execute the new priority task
+        priority_document_indexing_task(tenant_id, dataset.id, document_ids)
+
+        # Assert: Verify processing occurred (core logic is tested in _document_indexing tests)
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+    def test_document_indexing_with_tenant_queue_success(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test _document_indexing_with_tenant_queue function with no waiting tasks.
+
+        This test verifies:
+        - Core indexing logic execution (same as _document_indexing)
+        - Tenant queue cleanup when no waiting tasks
+        - Task function parameter passing
+        - Queue management after processing
+        """
+        # Arrange: Create test data
+        dataset, documents = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=2
+        )
+        document_ids = [doc.id for doc in documents]
+        tenant_id = dataset.tenant_id
+
+        # Mock the task function
+        from unittest.mock import MagicMock
+
+        mock_task_func = MagicMock()
+
+        # Act: Execute the wrapper function
+        _document_indexing_with_tenant_queue(tenant_id, dataset.id, document_ids, mock_task_func)
+
+        # Assert: Verify core processing occurred (same as _document_indexing)
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+        # Verify documents were updated (same as _document_indexing)
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None
+
+        # Verify the run method was called with correct documents
+        call_args = mock_external_service_dependencies["indexing_runner_instance"].run.call_args
+        assert call_args is not None
+        processed_documents = call_args[0][0]
+        assert len(processed_documents) == 2
+
+        # Verify task function was not called (no waiting tasks)
+        mock_task_func.delay.assert_not_called()
+
+    def test_document_indexing_with_tenant_queue_with_waiting_tasks(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test _document_indexing_with_tenant_queue function with waiting tasks in queue using real Redis.
+
+        This test verifies:
+        - Core indexing logic execution
+        - Real Redis-based tenant queue processing of waiting tasks
+        - Task function calls for waiting tasks
+        - Queue management with multiple tasks using actual Redis operations
+        """
+        # Arrange: Create test data
+        dataset, documents = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+        document_ids = [doc.id for doc in documents]
+        tenant_id = dataset.tenant_id
+        dataset_id = dataset.id
+
+        # Mock the task function
+        from unittest.mock import MagicMock
+
+        mock_task_func = MagicMock()
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+
+        # Create real queue instance
+        queue = TenantIsolatedTaskQueue(tenant_id, "document_indexing")
+
+        # Add waiting tasks to the real Redis queue
+        waiting_tasks = [
+            DocumentTask(tenant_id=tenant_id, dataset_id=dataset.id, document_ids=["waiting-doc-1"]),
+            DocumentTask(tenant_id=tenant_id, dataset_id=dataset.id, document_ids=["waiting-doc-2"]),
+        ]
+        # Convert DocumentTask objects to dictionaries for serialization
+        waiting_task_dicts = [asdict(task) for task in waiting_tasks]
+        queue.push_tasks(waiting_task_dicts)
+
+        # Act: Execute the wrapper function
+        _document_indexing_with_tenant_queue(tenant_id, dataset.id, document_ids, mock_task_func)
+
+        # Assert: Verify core processing occurred
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+        # Verify task function was called for each waiting task
+        assert mock_task_func.delay.call_count == 1
+
+        # Verify correct parameters for each call
+        calls = mock_task_func.delay.call_args_list
+        assert calls[0][1] == {"tenant_id": tenant_id, "dataset_id": dataset_id, "document_ids": ["waiting-doc-1"]}
+
+        # Verify queue is empty after processing (tasks were pulled)
+        remaining_tasks = queue.pull_tasks(count=10)  # Pull more than we added
+        assert len(remaining_tasks) == 1
+
+    def test_document_indexing_with_tenant_queue_error_handling(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test error handling in _document_indexing_with_tenant_queue using real Redis.
+
+        This test verifies:
+        - Exception handling during core processing
+        - Tenant queue cleanup even on errors using real Redis
+        - Proper error logging
+        - Function completes without raising exceptions
+        - Queue management continues despite core processing errors
+        """
+        # Arrange: Create test data
+        dataset, documents = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+        document_ids = [doc.id for doc in documents]
+        tenant_id = dataset.tenant_id
+        dataset_id = dataset.id
+
+        # Mock IndexingRunner to raise an exception
+        mock_external_service_dependencies["indexing_runner_instance"].run.side_effect = Exception("Test error")
+
+        # Mock the task function
+        from unittest.mock import MagicMock
+
+        mock_task_func = MagicMock()
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+
+        # Create real queue instance
+        queue = TenantIsolatedTaskQueue(tenant_id, "document_indexing")
+
+        # Add waiting task to the real Redis queue
+        waiting_task = DocumentTask(tenant_id=tenant_id, dataset_id=dataset.id, document_ids=["waiting-doc-1"])
+        queue.push_tasks([asdict(waiting_task)])
+
+        # Act: Execute the wrapper function
+        _document_indexing_with_tenant_queue(tenant_id, dataset.id, document_ids, mock_task_func)
+
+        # Assert: Verify error was handled gracefully
+        # The function should not raise exceptions
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+        # Verify documents were still updated to parsing status before the exception
+        # Re-query documents from database since _document_indexing uses a different session
+        for doc_id in document_ids:
+            updated_document = db.session.query(Document).where(Document.id == doc_id).first()
+            assert updated_document.indexing_status == "parsing"
+            assert updated_document.processing_started_at is not None
+
+        # Verify waiting task was still processed despite core processing error
+        mock_task_func.delay.assert_called_once()
+
+        # Verify correct parameters for the call
+        call = mock_task_func.delay.call_args
+        assert call[1] == {"tenant_id": tenant_id, "dataset_id": dataset_id, "document_ids": ["waiting-doc-1"]}
+
+        # Verify queue is empty after processing (task was pulled)
+        remaining_tasks = queue.pull_tasks(count=10)
+        assert len(remaining_tasks) == 0
+
+    def test_document_indexing_with_tenant_queue_tenant_isolation(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test tenant isolation in _document_indexing_with_tenant_queue using real Redis.
+
+        This test verifies:
+        - Different tenants have isolated queues
+        - Tasks from one tenant don't affect another tenant's queue
+        - Queue operations are properly scoped to tenant
+        """
+        # Arrange: Create test data for two different tenants
+        dataset1, documents1 = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+        dataset2, documents2 = self._create_test_dataset_and_documents(
+            db_session_with_containers, mock_external_service_dependencies, document_count=1
+        )
+
+        tenant1_id = dataset1.tenant_id
+        tenant2_id = dataset2.tenant_id
+        dataset1_id = dataset1.id
+        dataset2_id = dataset2.id
+        document_ids1 = [doc.id for doc in documents1]
+        document_ids2 = [doc.id for doc in documents2]
+
+        # Mock the task function
+        from unittest.mock import MagicMock
+
+        mock_task_func = MagicMock()
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+
+        # Create queue instances for both tenants
+        queue1 = TenantIsolatedTaskQueue(tenant1_id, "document_indexing")
+        queue2 = TenantIsolatedTaskQueue(tenant2_id, "document_indexing")
+
+        # Add waiting tasks to both queues
+        waiting_task1 = DocumentTask(tenant_id=tenant1_id, dataset_id=dataset1.id, document_ids=["tenant1-doc-1"])
+        waiting_task2 = DocumentTask(tenant_id=tenant2_id, dataset_id=dataset2.id, document_ids=["tenant2-doc-1"])
+
+        queue1.push_tasks([asdict(waiting_task1)])
+        queue2.push_tasks([asdict(waiting_task2)])
+
+        # Act: Execute the wrapper function for tenant1 only
+        _document_indexing_with_tenant_queue(tenant1_id, dataset1.id, document_ids1, mock_task_func)
+
+        # Assert: Verify core processing occurred for tenant1
+        mock_external_service_dependencies["indexing_runner"].assert_called_once()
+        mock_external_service_dependencies["indexing_runner_instance"].run.assert_called_once()
+
+        # Verify only tenant1's waiting task was processed
+        mock_task_func.delay.assert_called_once()
+        call = mock_task_func.delay.call_args
+        assert call[1] == {"tenant_id": tenant1_id, "dataset_id": dataset1_id, "document_ids": ["tenant1-doc-1"]}
+
+        # Verify tenant1's queue is empty
+        remaining_tasks1 = queue1.pull_tasks(count=10)
+        assert len(remaining_tasks1) == 0
+
+        # Verify tenant2's queue still has its task (isolation)
+        remaining_tasks2 = queue2.pull_tasks(count=10)
+        assert len(remaining_tasks2) == 1
+
+        # Verify queue keys are different
+        assert queue1._queue != queue2._queue
+        assert queue1._task_key != queue2._task_key
--- a/api/tests/test_containers_integration_tests/tasks/test_rag_pipeline_run_tasks.py
+++ b/api/tests/test_containers_integration_tests/tasks/test_rag_pipeline_run_tasks.py
@@ -0,0 +1,936 @@
+import json
+import uuid
+from unittest.mock import patch
+
+import pytest
+from faker import Faker
+
+from core.app.entities.app_invoke_entities import InvokeFrom, RagPipelineGenerateEntity
+from core.app.entities.rag_pipeline_invoke_entities import RagPipelineInvokeEntity
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+from extensions.ext_database import db
+from models import Account, Tenant, TenantAccountJoin, TenantAccountRole
+from models.dataset import Pipeline
+from models.workflow import Workflow
+from tasks.rag_pipeline.priority_rag_pipeline_run_task import (
+    priority_rag_pipeline_run_task,
+    run_single_rag_pipeline_task,
+)
+from tasks.rag_pipeline.rag_pipeline_run_task import rag_pipeline_run_task
+
+
+class TestRagPipelineRunTasks:
+    """Integration tests for RAG pipeline run tasks using testcontainers.
+
+    This test class covers:
+    - priority_rag_pipeline_run_task function
+    - rag_pipeline_run_task function
+    - run_single_rag_pipeline_task function
+    - Real Redis-based TenantIsolatedTaskQueue operations
+    - PipelineGenerator._generate method mocking and parameter validation
+    - File operations and cleanup
+    - Error handling and queue management
+    """
+
+    @pytest.fixture
+    def mock_pipeline_generator(self):
+        """Mock PipelineGenerator._generate method."""
+        with patch("core.app.apps.pipeline.pipeline_generator.PipelineGenerator._generate") as mock_generate:
+            # Mock the _generate method to return a simple response
+            mock_generate.return_value = {"answer": "Test response", "metadata": {"test": "data"}}
+            yield mock_generate
+
+    @pytest.fixture
+    def mock_file_service(self):
+        """Mock FileService for file operations."""
+        with (
+            patch("services.file_service.FileService.get_file_content") as mock_get_content,
+            patch("services.file_service.FileService.delete_file") as mock_delete_file,
+        ):
+            yield {
+                "get_content": mock_get_content,
+                "delete_file": mock_delete_file,
+            }
+
+    def _create_test_pipeline_and_workflow(self, db_session_with_containers):
+        """
+        Helper method to create test pipeline and workflow for testing.
+
+        Args:
+            db_session_with_containers: Database session from testcontainers infrastructure
+
+        Returns:
+            tuple: (account, tenant, pipeline, workflow) - Created entities
+        """
+        fake = Faker()
+
+        # Create account and tenant
+        account = Account(
+            email=fake.email(),
+            name=fake.name(),
+            interface_language="en-US",
+            status="active",
+        )
+        db.session.add(account)
+        db.session.commit()
+
+        tenant = Tenant(
+            name=fake.company(),
+            status="normal",
+        )
+        db.session.add(tenant)
+        db.session.commit()
+
+        # Create tenant-account join
+        join = TenantAccountJoin(
+            tenant_id=tenant.id,
+            account_id=account.id,
+            role=TenantAccountRole.OWNER,
+            current=True,
+        )
+        db.session.add(join)
+        db.session.commit()
+
+        # Create workflow
+        workflow = Workflow(
+            id=str(uuid.uuid4()),
+            tenant_id=tenant.id,
+            app_id=str(uuid.uuid4()),
+            type="workflow",
+            version="draft",
+            graph="{}",
+            features="{}",
+            marked_name=fake.company(),
+            marked_comment=fake.text(max_nb_chars=100),
+            created_by=account.id,
+            environment_variables=[],
+            conversation_variables=[],
+            rag_pipeline_variables=[],
+        )
+        db.session.add(workflow)
+        db.session.commit()
+
+        # Create pipeline
+        pipeline = Pipeline(
+            id=str(uuid.uuid4()),
+            tenant_id=tenant.id,
+            workflow_id=workflow.id,
+            name=fake.company(),
+            description=fake.text(max_nb_chars=100),
+            created_by=account.id,
+        )
+        db.session.add(pipeline)
+        db.session.commit()
+
+        # Refresh entities to ensure they're properly loaded
+        db.session.refresh(account)
+        db.session.refresh(tenant)
+        db.session.refresh(workflow)
+        db.session.refresh(pipeline)
+
+        return account, tenant, pipeline, workflow
+
+    def _create_rag_pipeline_invoke_entities(self, account, tenant, pipeline, workflow, count=2):
+        """
+        Helper method to create RAG pipeline invoke entities for testing.
+
+        Args:
+            account: Account instance
+            tenant: Tenant instance
+            pipeline: Pipeline instance
+            workflow: Workflow instance
+            count: Number of entities to create
+
+        Returns:
+            list: List of RagPipelineInvokeEntity instances
+        """
+        fake = Faker()
+        entities = []
+
+        for i in range(count):
+            # Create application generate entity
+            app_config = {
+                "app_id": str(uuid.uuid4()),
+                "app_name": fake.company(),
+                "mode": "workflow",
+                "workflow_id": workflow.id,
+                "tenant_id": tenant.id,
+                "app_mode": "workflow",
+            }
+
+            application_generate_entity = {
+                "task_id": str(uuid.uuid4()),
+                "app_config": app_config,
+                "inputs": {"query": f"Test query {i}"},
+                "files": [],
+                "user_id": account.id,
+                "stream": False,
+                "invoke_from": "published",
+                "workflow_execution_id": str(uuid.uuid4()),
+                "pipeline_config": {
+                    "app_id": str(uuid.uuid4()),
+                    "app_name": fake.company(),
+                    "mode": "workflow",
+                    "workflow_id": workflow.id,
+                    "tenant_id": tenant.id,
+                    "app_mode": "workflow",
+                },
+                "datasource_type": "upload_file",
+                "datasource_info": {},
+                "dataset_id": str(uuid.uuid4()),
+                "batch": "test_batch",
+            }
+
+            entity = RagPipelineInvokeEntity(
+                pipeline_id=pipeline.id,
+                application_generate_entity=application_generate_entity,
+                user_id=account.id,
+                tenant_id=tenant.id,
+                workflow_id=workflow.id,
+                streaming=False,
+                workflow_execution_id=str(uuid.uuid4()),
+                workflow_thread_pool_id=str(uuid.uuid4()),
+            )
+            entities.append(entity)
+
+        return entities
+
+    def _create_file_content_for_entities(self, entities):
+        """
+        Helper method to create file content for RAG pipeline invoke entities.
+
+        Args:
+            entities: List of RagPipelineInvokeEntity instances
+
+        Returns:
+            str: JSON string containing serialized entities
+        """
+        entities_data = [entity.model_dump() for entity in entities]
+        return json.dumps(entities_data)
+
+    def test_priority_rag_pipeline_run_task_success(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test successful priority RAG pipeline run task execution.
+
+        This test verifies:
+        - Task execution with multiple RAG pipeline invoke entities
+        - File content retrieval and parsing
+        - PipelineGenerator._generate method calls with correct parameters
+        - Thread pool execution
+        - File cleanup after execution
+        - Queue management with no waiting tasks
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=2)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Act: Execute the priority task
+        priority_rag_pipeline_run_task(file_id, tenant.id)
+
+        # Assert: Verify expected outcomes
+        # Verify file operations
+        mock_file_service["get_content"].assert_called_once_with(file_id)
+        mock_file_service["delete_file"].assert_called_once_with(file_id)
+
+        # Verify PipelineGenerator._generate was called for each entity
+        assert mock_pipeline_generator.call_count == 2
+
+        # Verify call parameters for each entity
+        calls = mock_pipeline_generator.call_args_list
+        for call in calls:
+            call_kwargs = call[1]  # Get keyword arguments
+            assert call_kwargs["pipeline"].id == pipeline.id
+            assert call_kwargs["workflow_id"] == workflow.id
+            assert call_kwargs["user"].id == account.id
+            assert call_kwargs["invoke_from"] == InvokeFrom.PUBLISHED
+            assert call_kwargs["streaming"] == False
+            assert isinstance(call_kwargs["application_generate_entity"], RagPipelineGenerateEntity)
+
+    def test_rag_pipeline_run_task_success(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test successful regular RAG pipeline run task execution.
+
+        This test verifies:
+        - Task execution with multiple RAG pipeline invoke entities
+        - File content retrieval and parsing
+        - PipelineGenerator._generate method calls with correct parameters
+        - Thread pool execution
+        - File cleanup after execution
+        - Queue management with no waiting tasks
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=3)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Act: Execute the regular task
+        rag_pipeline_run_task(file_id, tenant.id)
+
+        # Assert: Verify expected outcomes
+        # Verify file operations
+        mock_file_service["get_content"].assert_called_once_with(file_id)
+        mock_file_service["delete_file"].assert_called_once_with(file_id)
+
+        # Verify PipelineGenerator._generate was called for each entity
+        assert mock_pipeline_generator.call_count == 3
+
+        # Verify call parameters for each entity
+        calls = mock_pipeline_generator.call_args_list
+        for call in calls:
+            call_kwargs = call[1]  # Get keyword arguments
+            assert call_kwargs["pipeline"].id == pipeline.id
+            assert call_kwargs["workflow_id"] == workflow.id
+            assert call_kwargs["user"].id == account.id
+            assert call_kwargs["invoke_from"] == InvokeFrom.PUBLISHED
+            assert call_kwargs["streaming"] == False
+            assert isinstance(call_kwargs["application_generate_entity"], RagPipelineGenerateEntity)
+
+    def test_priority_rag_pipeline_run_task_with_waiting_tasks(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test priority RAG pipeline run task with waiting tasks in queue using real Redis.
+
+        This test verifies:
+        - Core task execution
+        - Real Redis-based tenant queue processing of waiting tasks
+        - Task function calls for waiting tasks
+        - Queue management with multiple tasks using actual Redis operations
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=1)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+        # Add waiting tasks to the real Redis queue
+        waiting_file_ids = [str(uuid.uuid4()) for _ in range(2)]
+        queue.push_tasks(waiting_file_ids)
+
+        # Mock the task function calls
+        with patch(
+            "tasks.rag_pipeline.priority_rag_pipeline_run_task.priority_rag_pipeline_run_task.delay"
+        ) as mock_delay:
+            # Act: Execute the priority task
+            priority_rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify core processing occurred
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_file_service["delete_file"].assert_called_once_with(file_id)
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify waiting tasks were processed, pull 1 task a time by default
+            assert mock_delay.call_count == 1
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_ids[0]
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify queue still has remaining tasks (only 1 was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 1  # 2 original - 1 pulled = 1 remaining
+
+    def test_rag_pipeline_run_task_legacy_compatibility(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test regular RAG pipeline run task with legacy Redis queue format for backward compatibility.
+
+        This test simulates the scenario where:
+        - Old code writes file IDs directly to Redis list using lpush
+        - New worker processes these legacy queue entries
+        - Ensures backward compatibility during deployment transition
+
+        Legacy format: redis_client.lpush(tenant_self_pipeline_task_queue, upload_file.id)
+        New format: TenantIsolatedTaskQueue.push_tasks([file_id])
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=1)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Simulate legacy Redis queue format - direct file IDs in Redis list
+        from extensions.ext_redis import redis_client
+
+        # Legacy queue key format (old code)
+        legacy_queue_key = f"tenant_self_pipeline_task_queue:{tenant.id}"
+        legacy_task_key = f"tenant_pipeline_task:{tenant.id}"
+
+        # Add legacy format data to Redis (simulating old code behavior)
+        legacy_file_ids = [str(uuid.uuid4()) for _ in range(3)]
+        for file_id_legacy in legacy_file_ids:
+            redis_client.lpush(legacy_queue_key, file_id_legacy)
+
+        # Set the task key to indicate there are waiting tasks (legacy behavior)
+        redis_client.set(legacy_task_key, 1, ex=60 * 60)
+
+        # Mock the task function calls
+        with patch("tasks.rag_pipeline.rag_pipeline_run_task.rag_pipeline_run_task.delay") as mock_delay:
+            # Act: Execute the priority task with new code but legacy queue data
+            rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify core processing occurred
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_file_service["delete_file"].assert_called_once_with(file_id)
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify waiting tasks were processed, pull 1 task a time by default
+            assert mock_delay.call_count == 1
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == legacy_file_ids[0]
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify that new code can process legacy queue entries
+            # The new TenantIsolatedTaskQueue should be able to read from the legacy format
+            queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+            # Verify queue still has remaining tasks (only 1 was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 2  # 3 original - 1 pulled = 2 remaining
+
+        # Cleanup: Remove legacy test data
+        redis_client.delete(legacy_queue_key)
+        redis_client.delete(legacy_task_key)
+
+    def test_rag_pipeline_run_task_with_waiting_tasks(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test regular RAG pipeline run task with waiting tasks in queue using real Redis.
+
+        This test verifies:
+        - Core task execution
+        - Real Redis-based tenant queue processing of waiting tasks
+        - Task function calls for waiting tasks
+        - Queue management with multiple tasks using actual Redis operations
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=1)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+        # Add waiting tasks to the real Redis queue
+        waiting_file_ids = [str(uuid.uuid4()) for _ in range(3)]
+        queue.push_tasks(waiting_file_ids)
+
+        # Mock the task function calls
+        with patch("tasks.rag_pipeline.rag_pipeline_run_task.rag_pipeline_run_task.delay") as mock_delay:
+            # Act: Execute the regular task
+            rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify core processing occurred
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_file_service["delete_file"].assert_called_once_with(file_id)
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify waiting tasks were processed, pull 1 task a time by default
+            assert mock_delay.call_count == 1
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_ids[0]
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify queue still has remaining tasks (only 1 was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 2  # 3 original - 1 pulled = 2 remaining
+
+    def test_priority_rag_pipeline_run_task_error_handling(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test error handling in priority RAG pipeline run task using real Redis.
+
+        This test verifies:
+        - Exception handling during core processing
+        - Tenant queue cleanup even on errors using real Redis
+        - Proper error logging
+        - Function completes without raising exceptions
+        - Queue management continues despite core processing errors
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=1)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Mock PipelineGenerator to raise an exception
+        mock_pipeline_generator.side_effect = Exception("Pipeline generation failed")
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+        # Add waiting task to the real Redis queue
+        waiting_file_id = str(uuid.uuid4())
+        queue.push_tasks([waiting_file_id])
+
+        # Mock the task function calls
+        with patch(
+            "tasks.rag_pipeline.priority_rag_pipeline_run_task.priority_rag_pipeline_run_task.delay"
+        ) as mock_delay:
+            # Act: Execute the priority task (should not raise exception)
+            priority_rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify error was handled gracefully
+            # The function should not raise exceptions
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_file_service["delete_file"].assert_called_once_with(file_id)
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify waiting task was still processed despite core processing error
+            mock_delay.assert_called_once()
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_id
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify queue is empty after processing (task was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 0
+
+    def test_rag_pipeline_run_task_error_handling(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test error handling in regular RAG pipeline run task using real Redis.
+
+        This test verifies:
+        - Exception handling during core processing
+        - Tenant queue cleanup even on errors using real Redis
+        - Proper error logging
+        - Function completes without raising exceptions
+        - Queue management continues despite core processing errors
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=1)
+        file_content = self._create_file_content_for_entities(entities)
+
+        # Mock file service
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].return_value = file_content
+
+        # Mock PipelineGenerator to raise an exception
+        mock_pipeline_generator.side_effect = Exception("Pipeline generation failed")
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+        # Add waiting task to the real Redis queue
+        waiting_file_id = str(uuid.uuid4())
+        queue.push_tasks([waiting_file_id])
+
+        # Mock the task function calls
+        with patch("tasks.rag_pipeline.rag_pipeline_run_task.rag_pipeline_run_task.delay") as mock_delay:
+            # Act: Execute the regular task (should not raise exception)
+            rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify error was handled gracefully
+            # The function should not raise exceptions
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_file_service["delete_file"].assert_called_once_with(file_id)
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify waiting task was still processed despite core processing error
+            mock_delay.assert_called_once()
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_id
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify queue is empty after processing (task was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 0
+
+    def test_priority_rag_pipeline_run_task_tenant_isolation(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test tenant isolation in priority RAG pipeline run task using real Redis.
+
+        This test verifies:
+        - Different tenants have isolated queues
+        - Tasks from one tenant don't affect another tenant's queue
+        - Queue operations are properly scoped to tenant
+        """
+        # Arrange: Create test data for two different tenants
+        account1, tenant1, pipeline1, workflow1 = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        account2, tenant2, pipeline2, workflow2 = self._create_test_pipeline_and_workflow(db_session_with_containers)
+
+        entities1 = self._create_rag_pipeline_invoke_entities(account1, tenant1, pipeline1, workflow1, count=1)
+        entities2 = self._create_rag_pipeline_invoke_entities(account2, tenant2, pipeline2, workflow2, count=1)
+
+        file_content1 = self._create_file_content_for_entities(entities1)
+        file_content2 = self._create_file_content_for_entities(entities2)
+
+        # Mock file service
+        file_id1 = str(uuid.uuid4())
+        file_id2 = str(uuid.uuid4())
+        mock_file_service["get_content"].side_effect = [file_content1, file_content2]
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue1 = TenantIsolatedTaskQueue(tenant1.id, "pipeline")
+        queue2 = TenantIsolatedTaskQueue(tenant2.id, "pipeline")
+
+        # Add waiting tasks to both queues
+        waiting_file_id1 = str(uuid.uuid4())
+        waiting_file_id2 = str(uuid.uuid4())
+
+        queue1.push_tasks([waiting_file_id1])
+        queue2.push_tasks([waiting_file_id2])
+
+        # Mock the task function calls
+        with patch(
+            "tasks.rag_pipeline.priority_rag_pipeline_run_task.priority_rag_pipeline_run_task.delay"
+        ) as mock_delay:
+            # Act: Execute the priority task for tenant1 only
+            priority_rag_pipeline_run_task(file_id1, tenant1.id)
+
+            # Assert: Verify core processing occurred for tenant1
+            assert mock_file_service["get_content"].call_count == 1
+            assert mock_file_service["delete_file"].call_count == 1
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify only tenant1's waiting task was processed
+            mock_delay.assert_called_once()
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_id1
+            assert call_kwargs.get("tenant_id") == tenant1.id
+
+            # Verify tenant1's queue is empty
+            remaining_tasks1 = queue1.pull_tasks(count=10)
+            assert len(remaining_tasks1) == 0
+
+            # Verify tenant2's queue still has its task (isolation)
+            remaining_tasks2 = queue2.pull_tasks(count=10)
+            assert len(remaining_tasks2) == 1
+
+            # Verify queue keys are different
+            assert queue1._queue != queue2._queue
+            assert queue1._task_key != queue2._task_key
+
+    def test_rag_pipeline_run_task_tenant_isolation(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test tenant isolation in regular RAG pipeline run task using real Redis.
+
+        This test verifies:
+        - Different tenants have isolated queues
+        - Tasks from one tenant don't affect another tenant's queue
+        - Queue operations are properly scoped to tenant
+        """
+        # Arrange: Create test data for two different tenants
+        account1, tenant1, pipeline1, workflow1 = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        account2, tenant2, pipeline2, workflow2 = self._create_test_pipeline_and_workflow(db_session_with_containers)
+
+        entities1 = self._create_rag_pipeline_invoke_entities(account1, tenant1, pipeline1, workflow1, count=1)
+        entities2 = self._create_rag_pipeline_invoke_entities(account2, tenant2, pipeline2, workflow2, count=1)
+
+        file_content1 = self._create_file_content_for_entities(entities1)
+        file_content2 = self._create_file_content_for_entities(entities2)
+
+        # Mock file service
+        file_id1 = str(uuid.uuid4())
+        file_id2 = str(uuid.uuid4())
+        mock_file_service["get_content"].side_effect = [file_content1, file_content2]
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue1 = TenantIsolatedTaskQueue(tenant1.id, "pipeline")
+        queue2 = TenantIsolatedTaskQueue(tenant2.id, "pipeline")
+
+        # Add waiting tasks to both queues
+        waiting_file_id1 = str(uuid.uuid4())
+        waiting_file_id2 = str(uuid.uuid4())
+
+        queue1.push_tasks([waiting_file_id1])
+        queue2.push_tasks([waiting_file_id2])
+
+        # Mock the task function calls
+        with patch("tasks.rag_pipeline.rag_pipeline_run_task.rag_pipeline_run_task.delay") as mock_delay:
+            # Act: Execute the regular task for tenant1 only
+            rag_pipeline_run_task(file_id1, tenant1.id)
+
+            # Assert: Verify core processing occurred for tenant1
+            assert mock_file_service["get_content"].call_count == 1
+            assert mock_file_service["delete_file"].call_count == 1
+            assert mock_pipeline_generator.call_count == 1
+
+            # Verify only tenant1's waiting task was processed
+            mock_delay.assert_called_once()
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_id1
+            assert call_kwargs.get("tenant_id") == tenant1.id
+
+            # Verify tenant1's queue is empty
+            remaining_tasks1 = queue1.pull_tasks(count=10)
+            assert len(remaining_tasks1) == 0
+
+            # Verify tenant2's queue still has its task (isolation)
+            remaining_tasks2 = queue2.pull_tasks(count=10)
+            assert len(remaining_tasks2) == 1
+
+            # Verify queue keys are different
+            assert queue1._queue != queue2._queue
+            assert queue1._task_key != queue2._task_key
+
+    def test_run_single_rag_pipeline_task_success(
+        self, db_session_with_containers, mock_pipeline_generator, flask_app_with_containers
+    ):
+        """
+        Test successful run_single_rag_pipeline_task execution.
+
+        This test verifies:
+        - Single RAG pipeline task execution within Flask app context
+        - Entity validation and database queries
+        - PipelineGenerator._generate method call with correct parameters
+        - Proper Flask context handling
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+        entities = self._create_rag_pipeline_invoke_entities(account, tenant, pipeline, workflow, count=1)
+        entity_data = entities[0].model_dump()
+
+        # Act: Execute the single task
+        with flask_app_with_containers.app_context():
+            run_single_rag_pipeline_task(entity_data, flask_app_with_containers)
+
+        # Assert: Verify expected outcomes
+        # Verify PipelineGenerator._generate was called
+        assert mock_pipeline_generator.call_count == 1
+
+        # Verify call parameters
+        call = mock_pipeline_generator.call_args
+        call_kwargs = call[1]  # Get keyword arguments
+        assert call_kwargs["pipeline"].id == pipeline.id
+        assert call_kwargs["workflow_id"] == workflow.id
+        assert call_kwargs["user"].id == account.id
+        assert call_kwargs["invoke_from"] == InvokeFrom.PUBLISHED
+        assert call_kwargs["streaming"] == False
+        assert isinstance(call_kwargs["application_generate_entity"], RagPipelineGenerateEntity)
+
+    def test_run_single_rag_pipeline_task_entity_validation_error(
+        self, db_session_with_containers, mock_pipeline_generator, flask_app_with_containers
+    ):
+        """
+        Test run_single_rag_pipeline_task with invalid entity data.
+
+        This test verifies:
+        - Proper error handling for invalid entity data
+        - Exception logging
+        - Function raises ValueError for missing entities
+        """
+        # Arrange: Create entity data with valid UUIDs but non-existent entities
+        fake = Faker()
+        invalid_entity_data = {
+            "pipeline_id": str(uuid.uuid4()),
+            "application_generate_entity": {
+                "app_config": {
+                    "app_id": str(uuid.uuid4()),
+                    "app_name": "Test App",
+                    "mode": "workflow",
+                    "workflow_id": str(uuid.uuid4()),
+                },
+                "inputs": {"query": "Test query"},
+                "query": "Test query",
+                "response_mode": "blocking",
+                "user": str(uuid.uuid4()),
+                "files": [],
+                "conversation_id": str(uuid.uuid4()),
+            },
+            "user_id": str(uuid.uuid4()),
+            "tenant_id": str(uuid.uuid4()),
+            "workflow_id": str(uuid.uuid4()),
+            "streaming": False,
+            "workflow_execution_id": str(uuid.uuid4()),
+            "workflow_thread_pool_id": str(uuid.uuid4()),
+        }
+
+        # Act & Assert: Execute the single task with non-existent entities (should raise ValueError)
+        with flask_app_with_containers.app_context():
+            with pytest.raises(ValueError, match="Account .* not found"):
+                run_single_rag_pipeline_task(invalid_entity_data, flask_app_with_containers)
+
+        # Assert: Pipeline generator should not be called
+        mock_pipeline_generator.assert_not_called()
+
+    def test_run_single_rag_pipeline_task_database_entity_not_found(
+        self, db_session_with_containers, mock_pipeline_generator, flask_app_with_containers
+    ):
+        """
+        Test run_single_rag_pipeline_task with non-existent database entities.
+
+        This test verifies:
+        - Proper error handling for missing database entities
+        - Exception logging
+        - Function raises ValueError for missing entities
+        """
+        # Arrange: Create test data with non-existent IDs
+        fake = Faker()
+        entity_data = {
+            "pipeline_id": str(uuid.uuid4()),
+            "application_generate_entity": {
+                "app_config": {
+                    "app_id": str(uuid.uuid4()),
+                    "app_name": "Test App",
+                    "mode": "workflow",
+                    "workflow_id": str(uuid.uuid4()),
+                },
+                "inputs": {"query": "Test query"},
+                "query": "Test query",
+                "response_mode": "blocking",
+                "user": str(uuid.uuid4()),
+                "files": [],
+                "conversation_id": str(uuid.uuid4()),
+            },
+            "user_id": str(uuid.uuid4()),
+            "tenant_id": str(uuid.uuid4()),
+            "workflow_id": str(uuid.uuid4()),
+            "streaming": False,
+            "workflow_execution_id": str(uuid.uuid4()),
+            "workflow_thread_pool_id": str(uuid.uuid4()),
+        }
+
+        # Act & Assert: Execute the single task with non-existent entities (should raise ValueError)
+        with flask_app_with_containers.app_context():
+            with pytest.raises(ValueError, match="Account .* not found"):
+                run_single_rag_pipeline_task(entity_data, flask_app_with_containers)
+
+        # Assert: Pipeline generator should not be called
+        mock_pipeline_generator.assert_not_called()
+
+    def test_priority_rag_pipeline_run_task_file_not_found(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test priority RAG pipeline run task with non-existent file.
+
+        This test verifies:
+        - Proper error handling for missing files
+        - Exception logging
+        - Function raises Exception for file errors
+        - Queue management continues despite file errors
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+
+        # Mock file service to raise exception
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].side_effect = Exception("File not found")
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+        # Add waiting task to the real Redis queue
+        waiting_file_id = str(uuid.uuid4())
+        queue.push_tasks([waiting_file_id])
+
+        # Mock the task function calls
+        with patch(
+            "tasks.rag_pipeline.priority_rag_pipeline_run_task.priority_rag_pipeline_run_task.delay"
+        ) as mock_delay:
+            # Act & Assert: Execute the priority task (should raise Exception)
+            with pytest.raises(Exception, match="File not found"):
+                priority_rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify error was handled gracefully
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_pipeline_generator.assert_not_called()
+
+            # Verify waiting task was still processed despite file error
+            mock_delay.assert_called_once()
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_id
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify queue is empty after processing (task was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 0
+
+    def test_rag_pipeline_run_task_file_not_found(
+        self, db_session_with_containers, mock_pipeline_generator, mock_file_service
+    ):
+        """
+        Test regular RAG pipeline run task with non-existent file.
+
+        This test verifies:
+        - Proper error handling for missing files
+        - Exception logging
+        - Function raises Exception for file errors
+        - Queue management continues despite file errors
+        """
+        # Arrange: Create test data
+        account, tenant, pipeline, workflow = self._create_test_pipeline_and_workflow(db_session_with_containers)
+
+        # Mock file service to raise exception
+        file_id = str(uuid.uuid4())
+        mock_file_service["get_content"].side_effect = Exception("File not found")
+
+        # Use real Redis for TenantIsolatedTaskQueue
+        queue = TenantIsolatedTaskQueue(tenant.id, "pipeline")
+
+        # Add waiting task to the real Redis queue
+        waiting_file_id = str(uuid.uuid4())
+        queue.push_tasks([waiting_file_id])
+
+        # Mock the task function calls
+        with patch("tasks.rag_pipeline.rag_pipeline_run_task.rag_pipeline_run_task.delay") as mock_delay:
+            # Act & Assert: Execute the regular task (should raise Exception)
+            with pytest.raises(Exception, match="File not found"):
+                rag_pipeline_run_task(file_id, tenant.id)
+
+            # Assert: Verify error was handled gracefully
+            mock_file_service["get_content"].assert_called_once_with(file_id)
+            mock_pipeline_generator.assert_not_called()
+
+            # Verify waiting task was still processed despite file error
+            mock_delay.assert_called_once()
+
+            # Verify correct parameters for the call
+            call_kwargs = mock_delay.call_args[1] if mock_delay.call_args else {}
+            assert call_kwargs.get("rag_pipeline_invoke_entities_file_id") == waiting_file_id
+            assert call_kwargs.get("tenant_id") == tenant.id
+
+            # Verify queue is empty after processing (task was pulled)
+            remaining_tasks = queue.pull_tasks(count=10)
+            assert len(remaining_tasks) == 0
--- a/api/tests/unit_tests/core/rag/pipeline/test_queue.py
+++ b/api/tests/unit_tests/core/rag/pipeline/test_queue.py
@@ -0,0 +1,301 @@
+"""
+Unit tests for TenantIsolatedTaskQueue.
+
+These tests verify the Redis-based task queue functionality for tenant-specific
+task management with proper serialization and deserialization.
+"""
+
+import json
+from unittest.mock import MagicMock, patch
+from uuid import uuid4
+
+import pytest
+from pydantic import ValidationError
+
+from core.rag.pipeline.queue import TaskWrapper, TenantIsolatedTaskQueue
+
+
+class TestTaskWrapper:
+    """Test cases for TaskWrapper serialization/deserialization."""
+
+    def test_serialize_simple_data(self):
+        """Test serialization of simple data types."""
+        data = {"key": "value", "number": 42, "list": [1, 2, 3]}
+        wrapper = TaskWrapper(data=data)
+
+        serialized = wrapper.serialize()
+        assert isinstance(serialized, str)
+
+        # Verify it's valid JSON
+        parsed = json.loads(serialized)
+        assert parsed["data"] == data
+
+    def test_serialize_complex_data(self):
+        """Test serialization of complex nested data."""
+        data = {
+            "nested": {"deep": {"value": "test", "numbers": [1, 2, 3, 4, 5]}},
+            "unicode": "测试中文",
+            "special_chars": "!@#$%^&*()",
+        }
+        wrapper = TaskWrapper(data=data)
+
+        serialized = wrapper.serialize()
+        parsed = json.loads(serialized)
+        assert parsed["data"] == data
+
+    def test_deserialize_valid_data(self):
+        """Test deserialization of valid JSON data."""
+        original_data = {"key": "value", "number": 42}
+        # Serialize using TaskWrapper to get the correct format
+        wrapper = TaskWrapper(data=original_data)
+        serialized = wrapper.serialize()
+
+        wrapper = TaskWrapper.deserialize(serialized)
+        assert wrapper.data == original_data
+
+    def test_deserialize_invalid_json(self):
+        """Test deserialization handles invalid JSON gracefully."""
+        invalid_json = "{invalid json}"
+
+        # Pydantic will raise ValidationError for invalid JSON
+        with pytest.raises(ValidationError):
+            TaskWrapper.deserialize(invalid_json)
+
+    def test_serialize_ensure_ascii_false(self):
+        """Test that serialization preserves Unicode characters."""
+        data = {"chinese": "中文测试", "emoji": "🚀"}
+        wrapper = TaskWrapper(data=data)
+
+        serialized = wrapper.serialize()
+        assert "中文测试" in serialized
+        assert "🚀" in serialized
+
+
+class TestTenantIsolatedTaskQueue:
+    """Test cases for TenantIsolatedTaskQueue functionality."""
+
+    @pytest.fixture
+    def mock_redis_client(self):
+        """Mock Redis client for testing."""
+        mock_redis = MagicMock()
+        return mock_redis
+
+    @pytest.fixture
+    def sample_queue(self, mock_redis_client):
+        """Create a sample TenantIsolatedTaskQueue instance."""
+        return TenantIsolatedTaskQueue("tenant-123", "test-key")
+
+    def test_initialization(self, sample_queue):
+        """Test queue initialization with correct key generation."""
+        assert sample_queue._tenant_id == "tenant-123"
+        assert sample_queue._unique_key == "test-key"
+        assert sample_queue._queue == "tenant_self_test-key_task_queue:tenant-123"
+        assert sample_queue._task_key == "tenant_test-key_task:tenant-123"
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_get_task_key_exists(self, mock_redis, sample_queue):
+        """Test getting task key when it exists."""
+        mock_redis.get.return_value = "1"
+
+        result = sample_queue.get_task_key()
+
+        assert result == "1"
+        mock_redis.get.assert_called_once_with("tenant_test-key_task:tenant-123")
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_get_task_key_not_exists(self, mock_redis, sample_queue):
+        """Test getting task key when it doesn't exist."""
+        mock_redis.get.return_value = None
+
+        result = sample_queue.get_task_key()
+
+        assert result is None
+        mock_redis.get.assert_called_once_with("tenant_test-key_task:tenant-123")
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_set_task_waiting_time_default_ttl(self, mock_redis, sample_queue):
+        """Test setting task waiting flag with default TTL."""
+        sample_queue.set_task_waiting_time()
+
+        mock_redis.setex.assert_called_once_with(
+            "tenant_test-key_task:tenant-123",
+            3600,  # DEFAULT_TASK_TTL
+            1,
+        )
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_set_task_waiting_time_custom_ttl(self, mock_redis, sample_queue):
+        """Test setting task waiting flag with custom TTL."""
+        custom_ttl = 1800
+        sample_queue.set_task_waiting_time(custom_ttl)
+
+        mock_redis.setex.assert_called_once_with("tenant_test-key_task:tenant-123", custom_ttl, 1)
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_delete_task_key(self, mock_redis, sample_queue):
+        """Test deleting task key."""
+        sample_queue.delete_task_key()
+
+        mock_redis.delete.assert_called_once_with("tenant_test-key_task:tenant-123")
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_push_tasks_string_list(self, mock_redis, sample_queue):
+        """Test pushing string tasks directly."""
+        tasks = ["task1", "task2", "task3"]
+
+        sample_queue.push_tasks(tasks)
+
+        mock_redis.lpush.assert_called_once_with(
+            "tenant_self_test-key_task_queue:tenant-123", "task1", "task2", "task3"
+        )
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_push_tasks_mixed_types(self, mock_redis, sample_queue):
+        """Test pushing mixed string and object tasks."""
+        tasks = ["string_task", {"object_task": "data", "id": 123}, "another_string"]
+
+        sample_queue.push_tasks(tasks)
+
+        # Verify lpush was called
+        mock_redis.lpush.assert_called_once()
+        call_args = mock_redis.lpush.call_args
+
+        # Check queue name
+        assert call_args[0][0] == "tenant_self_test-key_task_queue:tenant-123"
+
+        # Check serialized tasks
+        serialized_tasks = call_args[0][1:]
+        assert len(serialized_tasks) == 3
+        assert serialized_tasks[0] == "string_task"
+        assert serialized_tasks[2] == "another_string"
+
+        # Check object task is serialized as TaskWrapper JSON (without prefix)
+        # It should be a valid JSON string that can be deserialized by TaskWrapper
+        wrapper = TaskWrapper.deserialize(serialized_tasks[1])
+        assert wrapper.data == {"object_task": "data", "id": 123}
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_push_tasks_empty_list(self, mock_redis, sample_queue):
+        """Test pushing empty task list."""
+        sample_queue.push_tasks([])
+
+        mock_redis.lpush.assert_called_once_with("tenant_self_test-key_task_queue:tenant-123")
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_default_count(self, mock_redis, sample_queue):
+        """Test pulling tasks with default count (1)."""
+        mock_redis.rpop.side_effect = ["task1", None]
+
+        result = sample_queue.pull_tasks()
+
+        assert result == ["task1"]
+        assert mock_redis.rpop.call_count == 1
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_custom_count(self, mock_redis, sample_queue):
+        """Test pulling tasks with custom count."""
+        # First test: pull 3 tasks
+        mock_redis.rpop.side_effect = ["task1", "task2", "task3", None]
+
+        result = sample_queue.pull_tasks(3)
+
+        assert result == ["task1", "task2", "task3"]
+        assert mock_redis.rpop.call_count == 3
+
+        # Reset mock for second test
+        mock_redis.reset_mock()
+        mock_redis.rpop.side_effect = ["task1", "task2", None]
+
+        result = sample_queue.pull_tasks(3)
+
+        assert result == ["task1", "task2"]
+        assert mock_redis.rpop.call_count == 3
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_zero_count(self, mock_redis, sample_queue):
+        """Test pulling tasks with zero count returns empty list."""
+        result = sample_queue.pull_tasks(0)
+
+        assert result == []
+        mock_redis.rpop.assert_not_called()
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_negative_count(self, mock_redis, sample_queue):
+        """Test pulling tasks with negative count returns empty list."""
+        result = sample_queue.pull_tasks(-1)
+
+        assert result == []
+        mock_redis.rpop.assert_not_called()
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_with_wrapped_objects(self, mock_redis, sample_queue):
+        """Test pulling tasks that include wrapped objects."""
+        # Create a wrapped task
+        task_data = {"task_id": 123, "data": "test"}
+        wrapper = TaskWrapper(data=task_data)
+        wrapped_task = wrapper.serialize()
+
+        mock_redis.rpop.side_effect = [
+            "string_task",
+            wrapped_task.encode("utf-8"),  # Simulate bytes from Redis
+            None,
+        ]
+
+        result = sample_queue.pull_tasks(2)
+
+        assert len(result) == 2
+        assert result[0] == "string_task"
+        assert result[1] == {"task_id": 123, "data": "test"}
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_with_invalid_wrapped_data(self, mock_redis, sample_queue):
+        """Test pulling tasks with invalid JSON falls back to string."""
+        # Invalid JSON string that cannot be deserialized
+        invalid_json = "invalid json data"
+        mock_redis.rpop.side_effect = [invalid_json, None]
+
+        result = sample_queue.pull_tasks(1)
+
+        assert result == [invalid_json]
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_pull_tasks_bytes_decoding(self, mock_redis, sample_queue):
+        """Test pulling tasks handles bytes from Redis correctly."""
+        mock_redis.rpop.side_effect = [
+            b"task1",  # bytes
+            "task2",  # string
+            None,
+        ]
+
+        result = sample_queue.pull_tasks(2)
+
+        assert result == ["task1", "task2"]
+
+    @patch("core.rag.pipeline.queue.redis_client")
+    def test_complex_object_serialization_roundtrip(self, mock_redis, sample_queue):
+        """Test complex object serialization and deserialization roundtrip."""
+        complex_task = {
+            "id": uuid4().hex,
+            "data": {"nested": {"deep": [1, 2, 3], "unicode": "测试中文", "special": "!@#$%^&*()"}},
+            "metadata": {"created_at": "2024-01-01T00:00:00Z", "tags": ["tag1", "tag2", "tag3"]},
+        }
+
+        # Push the complex task
+        sample_queue.push_tasks([complex_task])
+
+        # Verify it was serialized as TaskWrapper JSON
+        call_args = mock_redis.lpush.call_args
+        wrapped_task = call_args[0][1]
+        # Verify it's a valid TaskWrapper JSON (starts with {"data":)
+        assert wrapped_task.startswith('{"data":')
+
+        # Verify it can be deserialized
+        wrapper = TaskWrapper.deserialize(wrapped_task)
+        assert wrapper.data == complex_task
+
+        # Simulate pulling it back
+        mock_redis.rpop.return_value = wrapped_task
+        result = sample_queue.pull_tasks(1)
+
+        assert len(result) == 1
+        assert result[0] == complex_task
--- a/api/tests/unit_tests/services/test_document_indexing_task_proxy.py
+++ b/api/tests/unit_tests/services/test_document_indexing_task_proxy.py
@@ -0,0 +1,317 @@
+from unittest.mock import Mock, patch
+
+from core.entities.document_task import DocumentTask
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+from enums.cloud_plan import CloudPlan
+from services.document_indexing_task_proxy import DocumentIndexingTaskProxy
+
+
+class DocumentIndexingTaskProxyTestDataFactory:
+    """Factory class for creating test data and mock objects for DocumentIndexingTaskProxy tests."""
+
+    @staticmethod
+    def create_mock_features(billing_enabled: bool = False, plan: CloudPlan = CloudPlan.SANDBOX) -> Mock:
+        """Create mock features with billing configuration."""
+        features = Mock()
+        features.billing = Mock()
+        features.billing.enabled = billing_enabled
+        features.billing.subscription = Mock()
+        features.billing.subscription.plan = plan
+        return features
+
+    @staticmethod
+    def create_mock_tenant_queue(has_task_key: bool = False) -> Mock:
+        """Create mock TenantIsolatedTaskQueue."""
+        queue = Mock(spec=TenantIsolatedTaskQueue)
+        queue.get_task_key.return_value = "task_key" if has_task_key else None
+        queue.push_tasks = Mock()
+        queue.set_task_waiting_time = Mock()
+        return queue
+
+    @staticmethod
+    def create_document_task_proxy(
+        tenant_id: str = "tenant-123", dataset_id: str = "dataset-456", document_ids: list[str] | None = None
+    ) -> DocumentIndexingTaskProxy:
+        """Create DocumentIndexingTaskProxy instance for testing."""
+        if document_ids is None:
+            document_ids = ["doc-1", "doc-2", "doc-3"]
+        return DocumentIndexingTaskProxy(tenant_id, dataset_id, document_ids)
+
+
+class TestDocumentIndexingTaskProxy:
+    """Test cases for DocumentIndexingTaskProxy class."""
+
+    def test_initialization(self):
+        """Test DocumentIndexingTaskProxy initialization."""
+        # Arrange
+        tenant_id = "tenant-123"
+        dataset_id = "dataset-456"
+        document_ids = ["doc-1", "doc-2", "doc-3"]
+
+        # Act
+        proxy = DocumentIndexingTaskProxy(tenant_id, dataset_id, document_ids)
+
+        # Assert
+        assert proxy._tenant_id == tenant_id
+        assert proxy._dataset_id == dataset_id
+        assert proxy._document_ids == document_ids
+        assert isinstance(proxy._tenant_isolated_task_queue, TenantIsolatedTaskQueue)
+        assert proxy._tenant_isolated_task_queue._tenant_id == tenant_id
+        assert proxy._tenant_isolated_task_queue._unique_key == "document_indexing"
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_features_property(self, mock_feature_service):
+        """Test cached_property features."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features()
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+
+        # Act
+        features1 = proxy.features
+        features2 = proxy.features  # Second call should use cached property
+
+        # Assert
+        assert features1 == mock_features
+        assert features2 == mock_features
+        assert features1 is features2  # Should be the same instance due to caching
+        mock_feature_service.get_features.assert_called_once_with("tenant-123")
+
+    @patch("services.document_indexing_task_proxy.normal_document_indexing_task")
+    def test_send_to_direct_queue(self, mock_task):
+        """Test _send_to_direct_queue method."""
+        # Arrange
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        mock_task.delay = Mock()
+
+        # Act
+        proxy._send_to_direct_queue(mock_task)
+
+        # Assert
+        mock_task.delay.assert_called_once_with(
+            tenant_id="tenant-123", dataset_id="dataset-456", document_ids=["doc-1", "doc-2", "doc-3"]
+        )
+
+    @patch("services.document_indexing_task_proxy.normal_document_indexing_task")
+    def test_send_to_tenant_queue_with_existing_task_key(self, mock_task):
+        """Test _send_to_tenant_queue when task key exists."""
+        # Arrange
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._tenant_isolated_task_queue = DocumentIndexingTaskProxyTestDataFactory.create_mock_tenant_queue(
+            has_task_key=True
+        )
+        mock_task.delay = Mock()
+
+        # Act
+        proxy._send_to_tenant_queue(mock_task)
+
+        # Assert
+        proxy._tenant_isolated_task_queue.push_tasks.assert_called_once()
+        pushed_tasks = proxy._tenant_isolated_task_queue.push_tasks.call_args[0][0]
+        assert len(pushed_tasks) == 1
+        assert isinstance(DocumentTask(**pushed_tasks[0]), DocumentTask)
+        assert pushed_tasks[0]["tenant_id"] == "tenant-123"
+        assert pushed_tasks[0]["dataset_id"] == "dataset-456"
+        assert pushed_tasks[0]["document_ids"] == ["doc-1", "doc-2", "doc-3"]
+        mock_task.delay.assert_not_called()
+
+    @patch("services.document_indexing_task_proxy.normal_document_indexing_task")
+    def test_send_to_tenant_queue_without_task_key(self, mock_task):
+        """Test _send_to_tenant_queue when no task key exists."""
+        # Arrange
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._tenant_isolated_task_queue = DocumentIndexingTaskProxyTestDataFactory.create_mock_tenant_queue(
+            has_task_key=False
+        )
+        mock_task.delay = Mock()
+
+        # Act
+        proxy._send_to_tenant_queue(mock_task)
+
+        # Assert
+        proxy._tenant_isolated_task_queue.set_task_waiting_time.assert_called_once()
+        mock_task.delay.assert_called_once_with(
+            tenant_id="tenant-123", dataset_id="dataset-456", document_ids=["doc-1", "doc-2", "doc-3"]
+        )
+        proxy._tenant_isolated_task_queue.push_tasks.assert_not_called()
+
+    @patch("services.document_indexing_task_proxy.normal_document_indexing_task")
+    def test_send_to_default_tenant_queue(self, mock_task):
+        """Test _send_to_default_tenant_queue method."""
+        # Arrange
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_tenant_queue = Mock()
+
+        # Act
+        proxy._send_to_default_tenant_queue()
+
+        # Assert
+        proxy._send_to_tenant_queue.assert_called_once_with(mock_task)
+
+    @patch("services.document_indexing_task_proxy.priority_document_indexing_task")
+    def test_send_to_priority_tenant_queue(self, mock_task):
+        """Test _send_to_priority_tenant_queue method."""
+        # Arrange
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_tenant_queue = Mock()
+
+        # Act
+        proxy._send_to_priority_tenant_queue()
+
+        # Assert
+        proxy._send_to_tenant_queue.assert_called_once_with(mock_task)
+
+    @patch("services.document_indexing_task_proxy.priority_document_indexing_task")
+    def test_send_to_priority_direct_queue(self, mock_task):
+        """Test _send_to_priority_direct_queue method."""
+        # Arrange
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_direct_queue = Mock()
+
+        # Act
+        proxy._send_to_priority_direct_queue()
+
+        # Assert
+        proxy._send_to_direct_queue.assert_called_once_with(mock_task)
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_dispatch_with_billing_enabled_sandbox_plan(self, mock_feature_service):
+        """Test _dispatch method when billing is enabled with sandbox plan."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features(
+            billing_enabled=True, plan=CloudPlan.SANDBOX
+        )
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_default_tenant_queue = Mock()
+
+        # Act
+        proxy._dispatch()
+
+        # Assert
+        proxy._send_to_default_tenant_queue.assert_called_once()
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_dispatch_with_billing_enabled_non_sandbox_plan(self, mock_feature_service):
+        """Test _dispatch method when billing is enabled with non-sandbox plan."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features(
+            billing_enabled=True, plan=CloudPlan.TEAM
+        )
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_priority_tenant_queue = Mock()
+
+        # Act
+        proxy._dispatch()
+
+        # If billing enabled with non sandbox plan, should send to priority tenant queue
+        proxy._send_to_priority_tenant_queue.assert_called_once()
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_dispatch_with_billing_disabled(self, mock_feature_service):
+        """Test _dispatch method when billing is disabled."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features(billing_enabled=False)
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_priority_direct_queue = Mock()
+
+        # Act
+        proxy._dispatch()
+
+        # If billing disabled, for example: self-hosted or enterprise, should send to priority direct queue
+        proxy._send_to_priority_direct_queue.assert_called_once()
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_delay_method(self, mock_feature_service):
+        """Test delay method integration."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features(
+            billing_enabled=True, plan=CloudPlan.SANDBOX
+        )
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_default_tenant_queue = Mock()
+
+        # Act
+        proxy.delay()
+
+        # Assert
+        # If billing enabled with sandbox plan, should send to default tenant queue
+        proxy._send_to_default_tenant_queue.assert_called_once()
+
+    def test_document_task_dataclass(self):
+        """Test DocumentTask dataclass."""
+        # Arrange
+        tenant_id = "tenant-123"
+        dataset_id = "dataset-456"
+        document_ids = ["doc-1", "doc-2"]
+
+        # Act
+        task = DocumentTask(tenant_id=tenant_id, dataset_id=dataset_id, document_ids=document_ids)
+
+        # Assert
+        assert task.tenant_id == tenant_id
+        assert task.dataset_id == dataset_id
+        assert task.document_ids == document_ids
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_dispatch_edge_case_empty_plan(self, mock_feature_service):
+        """Test _dispatch method with empty plan string."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features(billing_enabled=True, plan="")
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_priority_tenant_queue = Mock()
+
+        # Act
+        proxy._dispatch()
+
+        # Assert
+        proxy._send_to_priority_tenant_queue.assert_called_once()
+
+    @patch("services.document_indexing_task_proxy.FeatureService")
+    def test_dispatch_edge_case_none_plan(self, mock_feature_service):
+        """Test _dispatch method with None plan."""
+        # Arrange
+        mock_features = DocumentIndexingTaskProxyTestDataFactory.create_mock_features(billing_enabled=True, plan=None)
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = DocumentIndexingTaskProxyTestDataFactory.create_document_task_proxy()
+        proxy._send_to_priority_tenant_queue = Mock()
+
+        # Act
+        proxy._dispatch()
+
+        # Assert
+        proxy._send_to_priority_tenant_queue.assert_called_once()
+
+    def test_initialization_with_empty_document_ids(self):
+        """Test initialization with empty document_ids list."""
+        # Arrange
+        tenant_id = "tenant-123"
+        dataset_id = "dataset-456"
+        document_ids = []
+
+        # Act
+        proxy = DocumentIndexingTaskProxy(tenant_id, dataset_id, document_ids)
+
+        # Assert
+        assert proxy._tenant_id == tenant_id
+        assert proxy._dataset_id == dataset_id
+        assert proxy._document_ids == document_ids
+
+    def test_initialization_with_single_document_id(self):
+        """Test initialization with single document_id."""
+        # Arrange
+        tenant_id = "tenant-123"
+        dataset_id = "dataset-456"
+        document_ids = ["doc-1"]
+
+        # Act
+        proxy = DocumentIndexingTaskProxy(tenant_id, dataset_id, document_ids)
+
+        # Assert
+        assert proxy._tenant_id == tenant_id
+        assert proxy._dataset_id == dataset_id
+        assert proxy._document_ids == document_ids
--- a/api/tests/unit_tests/services/test_rag_pipeline_task_proxy.py
+++ b/api/tests/unit_tests/services/test_rag_pipeline_task_proxy.py
@@ -0,0 +1,483 @@
+import json
+from unittest.mock import Mock, patch
+
+import pytest
+
+from core.app.entities.rag_pipeline_invoke_entities import RagPipelineInvokeEntity
+from core.rag.pipeline.queue import TenantIsolatedTaskQueue
+from enums.cloud_plan import CloudPlan
+from services.rag_pipeline.rag_pipeline_task_proxy import RagPipelineTaskProxy
+
+
+class RagPipelineTaskProxyTestDataFactory:
+    """Factory class for creating test data and mock objects for RagPipelineTaskProxy tests."""
+
+    @staticmethod
+    def create_mock_features(billing_enabled: bool = False, plan: CloudPlan = CloudPlan.SANDBOX) -> Mock:
+        """Create mock features with billing configuration."""
+        features = Mock()
+        features.billing = Mock()
+        features.billing.enabled = billing_enabled
+        features.billing.subscription = Mock()
+        features.billing.subscription.plan = plan
+        return features
+
+    @staticmethod
+    def create_mock_tenant_queue(has_task_key: bool = False) -> Mock:
+        """Create mock TenantIsolatedTaskQueue."""
+        queue = Mock(spec=TenantIsolatedTaskQueue)
+        queue.get_task_key.return_value = "task_key" if has_task_key else None
+        queue.push_tasks = Mock()
+        queue.set_task_waiting_time = Mock()
+        return queue
+
+    @staticmethod
+    def create_rag_pipeline_invoke_entity(
+        pipeline_id: str = "pipeline-123",
+        user_id: str = "user-456",
+        tenant_id: str = "tenant-789",
+        workflow_id: str = "workflow-101",
+        streaming: bool = True,
+        workflow_execution_id: str | None = None,
+        workflow_thread_pool_id: str | None = None,
+    ) -> RagPipelineInvokeEntity:
+        """Create RagPipelineInvokeEntity instance for testing."""
+        return RagPipelineInvokeEntity(
+            pipeline_id=pipeline_id,
+            application_generate_entity={"key": "value"},
+            user_id=user_id,
+            tenant_id=tenant_id,
+            workflow_id=workflow_id,
+            streaming=streaming,
+            workflow_execution_id=workflow_execution_id,
+            workflow_thread_pool_id=workflow_thread_pool_id,
+        )
+
+    @staticmethod
+    def create_rag_pipeline_task_proxy(
+        dataset_tenant_id: str = "tenant-123",
+        user_id: str = "user-456",
+        rag_pipeline_invoke_entities: list[RagPipelineInvokeEntity] | None = None,
+    ) -> RagPipelineTaskProxy:
+        """Create RagPipelineTaskProxy instance for testing."""
+        if rag_pipeline_invoke_entities is None:
+            rag_pipeline_invoke_entities = [RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity()]
+        return RagPipelineTaskProxy(dataset_tenant_id, user_id, rag_pipeline_invoke_entities)
+
+    @staticmethod
+    def create_mock_upload_file(file_id: str = "file-123") -> Mock:
+        """Create mock upload file."""
+        upload_file = Mock()
+        upload_file.id = file_id
+        return upload_file
+
+
+class TestRagPipelineTaskProxy:
+    """Test cases for RagPipelineTaskProxy class."""
+
+    def test_initialization(self):
+        """Test RagPipelineTaskProxy initialization."""
+        # Arrange
+        dataset_tenant_id = "tenant-123"
+        user_id = "user-456"
+        rag_pipeline_invoke_entities = [RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity()]
+
+        # Act
+        proxy = RagPipelineTaskProxy(dataset_tenant_id, user_id, rag_pipeline_invoke_entities)
+
+        # Assert
+        assert proxy._dataset_tenant_id == dataset_tenant_id
+        assert proxy._user_id == user_id
+        assert proxy._rag_pipeline_invoke_entities == rag_pipeline_invoke_entities
+        assert isinstance(proxy._tenant_isolated_task_queue, TenantIsolatedTaskQueue)
+        assert proxy._tenant_isolated_task_queue._tenant_id == dataset_tenant_id
+        assert proxy._tenant_isolated_task_queue._unique_key == "pipeline"
+
+    def test_initialization_with_empty_entities(self):
+        """Test initialization with empty rag_pipeline_invoke_entities."""
+        # Arrange
+        dataset_tenant_id = "tenant-123"
+        user_id = "user-456"
+        rag_pipeline_invoke_entities = []
+
+        # Act
+        proxy = RagPipelineTaskProxy(dataset_tenant_id, user_id, rag_pipeline_invoke_entities)
+
+        # Assert
+        assert proxy._dataset_tenant_id == dataset_tenant_id
+        assert proxy._user_id == user_id
+        assert proxy._rag_pipeline_invoke_entities == []
+
+    def test_initialization_with_multiple_entities(self):
+        """Test initialization with multiple rag_pipeline_invoke_entities."""
+        # Arrange
+        dataset_tenant_id = "tenant-123"
+        user_id = "user-456"
+        rag_pipeline_invoke_entities = [
+            RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity(pipeline_id="pipeline-1"),
+            RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity(pipeline_id="pipeline-2"),
+            RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity(pipeline_id="pipeline-3"),
+        ]
+
+        # Act
+        proxy = RagPipelineTaskProxy(dataset_tenant_id, user_id, rag_pipeline_invoke_entities)
+
+        # Assert
+        assert len(proxy._rag_pipeline_invoke_entities) == 3
+        assert proxy._rag_pipeline_invoke_entities[0].pipeline_id == "pipeline-1"
+        assert proxy._rag_pipeline_invoke_entities[1].pipeline_id == "pipeline-2"
+        assert proxy._rag_pipeline_invoke_entities[2].pipeline_id == "pipeline-3"
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    def test_features_property(self, mock_feature_service):
+        """Test cached_property features."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features()
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+
+        # Act
+        features1 = proxy.features
+        features2 = proxy.features  # Second call should use cached property
+
+        # Assert
+        assert features1 == mock_features
+        assert features2 == mock_features
+        assert features1 is features2  # Should be the same instance due to caching
+        mock_feature_service.get_features.assert_called_once_with("tenant-123")
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_upload_invoke_entities(self, mock_db, mock_file_service_class):
+        """Test _upload_invoke_entities method."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        result = proxy._upload_invoke_entities()
+
+        # Assert
+        assert result == "file-123"
+        mock_file_service_class.assert_called_once_with(mock_db.engine)
+
+        # Verify upload_text was called with correct parameters
+        mock_file_service.upload_text.assert_called_once()
+        call_args = mock_file_service.upload_text.call_args
+        json_text, name, user_id, tenant_id = call_args[0]
+
+        assert name == "rag_pipeline_invoke_entities.json"
+        assert user_id == "user-456"
+        assert tenant_id == "tenant-123"
+
+        # Verify JSON content
+        parsed_json = json.loads(json_text)
+        assert len(parsed_json) == 1
+        assert parsed_json[0]["pipeline_id"] == "pipeline-123"
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_upload_invoke_entities_with_multiple_entities(self, mock_db, mock_file_service_class):
+        """Test _upload_invoke_entities method with multiple entities."""
+        # Arrange
+        entities = [
+            RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity(pipeline_id="pipeline-1"),
+            RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_invoke_entity(pipeline_id="pipeline-2"),
+        ]
+        proxy = RagPipelineTaskProxy("tenant-123", "user-456", entities)
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-456")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        result = proxy._upload_invoke_entities()
+
+        # Assert
+        assert result == "file-456"
+
+        # Verify JSON content contains both entities
+        call_args = mock_file_service.upload_text.call_args
+        json_text = call_args[0][0]
+        parsed_json = json.loads(json_text)
+        assert len(parsed_json) == 2
+        assert parsed_json[0]["pipeline_id"] == "pipeline-1"
+        assert parsed_json[1]["pipeline_id"] == "pipeline-2"
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.rag_pipeline_run_task")
+    def test_send_to_direct_queue(self, mock_task):
+        """Test _send_to_direct_queue method."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._tenant_isolated_task_queue = RagPipelineTaskProxyTestDataFactory.create_mock_tenant_queue()
+        upload_file_id = "file-123"
+        mock_task.delay = Mock()
+
+        # Act
+        proxy._send_to_direct_queue(upload_file_id, mock_task)
+
+        # If sent to direct queue, tenant_isolated_task_queue should not be called
+        proxy._tenant_isolated_task_queue.push_tasks.assert_not_called()
+
+        # Celery should be called directly
+        mock_task.delay.assert_called_once_with(
+            rag_pipeline_invoke_entities_file_id=upload_file_id, tenant_id="tenant-123"
+        )
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.rag_pipeline_run_task")
+    def test_send_to_tenant_queue_with_existing_task_key(self, mock_task):
+        """Test _send_to_tenant_queue when task key exists."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._tenant_isolated_task_queue = RagPipelineTaskProxyTestDataFactory.create_mock_tenant_queue(
+            has_task_key=True
+        )
+        upload_file_id = "file-123"
+        mock_task.delay = Mock()
+
+        # Act
+        proxy._send_to_tenant_queue(upload_file_id, mock_task)
+
+        # If task key exists, should push tasks to the queue
+        proxy._tenant_isolated_task_queue.push_tasks.assert_called_once_with([upload_file_id])
+        # Celery should not be called directly
+        mock_task.delay.assert_not_called()
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.rag_pipeline_run_task")
+    def test_send_to_tenant_queue_without_task_key(self, mock_task):
+        """Test _send_to_tenant_queue when no task key exists."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._tenant_isolated_task_queue = RagPipelineTaskProxyTestDataFactory.create_mock_tenant_queue(
+            has_task_key=False
+        )
+        upload_file_id = "file-123"
+        mock_task.delay = Mock()
+
+        # Act
+        proxy._send_to_tenant_queue(upload_file_id, mock_task)
+
+        # If no task key, should set task waiting time key first
+        proxy._tenant_isolated_task_queue.set_task_waiting_time.assert_called_once()
+        mock_task.delay.assert_called_once_with(
+            rag_pipeline_invoke_entities_file_id=upload_file_id, tenant_id="tenant-123"
+        )
+
+        # The first task should be sent to celery directly, so push tasks should not be called
+        proxy._tenant_isolated_task_queue.push_tasks.assert_not_called()
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.rag_pipeline_run_task")
+    def test_send_to_default_tenant_queue(self, mock_task):
+        """Test _send_to_default_tenant_queue method."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_tenant_queue = Mock()
+        upload_file_id = "file-123"
+
+        # Act
+        proxy._send_to_default_tenant_queue(upload_file_id)
+
+        # Assert
+        proxy._send_to_tenant_queue.assert_called_once_with(upload_file_id, mock_task)
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.priority_rag_pipeline_run_task")
+    def test_send_to_priority_tenant_queue(self, mock_task):
+        """Test _send_to_priority_tenant_queue method."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_tenant_queue = Mock()
+        upload_file_id = "file-123"
+
+        # Act
+        proxy._send_to_priority_tenant_queue(upload_file_id)
+
+        # Assert
+        proxy._send_to_tenant_queue.assert_called_once_with(upload_file_id, mock_task)
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.priority_rag_pipeline_run_task")
+    def test_send_to_priority_direct_queue(self, mock_task):
+        """Test _send_to_priority_direct_queue method."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_direct_queue = Mock()
+        upload_file_id = "file-123"
+
+        # Act
+        proxy._send_to_priority_direct_queue(upload_file_id)
+
+        # Assert
+        proxy._send_to_direct_queue.assert_called_once_with(upload_file_id, mock_task)
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_dispatch_with_billing_enabled_sandbox_plan(self, mock_db, mock_file_service_class, mock_feature_service):
+        """Test _dispatch method when billing is enabled with sandbox plan."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features(
+            billing_enabled=True, plan=CloudPlan.SANDBOX
+        )
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_default_tenant_queue = Mock()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        proxy._dispatch()
+
+        # If billing is enabled with sandbox plan, should send to default tenant queue
+        proxy._send_to_default_tenant_queue.assert_called_once_with("file-123")
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_dispatch_with_billing_enabled_non_sandbox_plan(
+        self, mock_db, mock_file_service_class, mock_feature_service
+    ):
+        """Test _dispatch method when billing is enabled with non-sandbox plan."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features(
+            billing_enabled=True, plan=CloudPlan.TEAM
+        )
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_priority_tenant_queue = Mock()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        proxy._dispatch()
+
+        # If billing is enabled with non-sandbox plan, should send to priority tenant queue
+        proxy._send_to_priority_tenant_queue.assert_called_once_with("file-123")
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_dispatch_with_billing_disabled(self, mock_db, mock_file_service_class, mock_feature_service):
+        """Test _dispatch method when billing is disabled."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features(billing_enabled=False)
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_priority_direct_queue = Mock()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        proxy._dispatch()
+
+        # If billing is disabled, for example: self-hosted or enterprise, should send to priority direct queue
+        proxy._send_to_priority_direct_queue.assert_called_once_with("file-123")
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_dispatch_with_empty_upload_file_id(self, mock_db, mock_file_service_class):
+        """Test _dispatch method when upload_file_id is empty."""
+        # Arrange
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = Mock()
+        mock_upload_file.id = ""  # Empty file ID
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act & Assert
+        with pytest.raises(ValueError, match="upload_file_id is empty"):
+            proxy._dispatch()
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_dispatch_edge_case_empty_plan(self, mock_db, mock_file_service_class, mock_feature_service):
+        """Test _dispatch method with empty plan string."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features(billing_enabled=True, plan="")
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_priority_tenant_queue = Mock()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        proxy._dispatch()
+
+        # Assert
+        proxy._send_to_priority_tenant_queue.assert_called_once_with("file-123")
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_dispatch_edge_case_none_plan(self, mock_db, mock_file_service_class, mock_feature_service):
+        """Test _dispatch method with None plan."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features(billing_enabled=True, plan=None)
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._send_to_priority_tenant_queue = Mock()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        proxy._dispatch()
+
+        # Assert
+        proxy._send_to_priority_tenant_queue.assert_called_once_with("file-123")
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FeatureService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.FileService")
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.db")
+    def test_delay_method(self, mock_db, mock_file_service_class, mock_feature_service):
+        """Test delay method integration."""
+        # Arrange
+        mock_features = RagPipelineTaskProxyTestDataFactory.create_mock_features(
+            billing_enabled=True, plan=CloudPlan.SANDBOX
+        )
+        mock_feature_service.get_features.return_value = mock_features
+        proxy = RagPipelineTaskProxyTestDataFactory.create_rag_pipeline_task_proxy()
+        proxy._dispatch = Mock()
+
+        mock_file_service = Mock()
+        mock_file_service_class.return_value = mock_file_service
+        mock_upload_file = RagPipelineTaskProxyTestDataFactory.create_mock_upload_file("file-123")
+        mock_file_service.upload_text.return_value = mock_upload_file
+
+        # Act
+        proxy.delay()
+
+        # Assert
+        proxy._dispatch.assert_called_once()
+
+    @patch("services.rag_pipeline.rag_pipeline_task_proxy.logger")
+    def test_delay_method_with_empty_entities(self, mock_logger):
+        """Test delay method with empty rag_pipeline_invoke_entities."""
+        # Arrange
+        proxy = RagPipelineTaskProxy("tenant-123", "user-456", [])
+
+        # Act
+        proxy.delay()
+
+        # Assert
+        mock_logger.warning.assert_called_once_with(
+            "Received empty rag pipeline invoke entities, no tasks delivered: %s %s", "tenant-123", "user-456"
+        )
--- a/dev/start-worker
+++ b/dev/start-worker
@@ -7,4 +7,4 @@ cd "$SCRIPT_DIR/.."

 uv --directory api run \
 	celery -A app.celery worker \
-	-P gevent -c 1 --loglevel INFO -Q dataset,mail,ops_trace,app_deletion,plugin,workflow_storage,conversation,priority_pipeline,pipeline
+	-P gevent -c 1 --loglevel INFO -Q dataset,priority_dataset,generation,mail,ops_trace,app_deletion,plugin,workflow_storage,conversation,priority_pipeline,pipeline
--- a/docker/.env.example
+++ b/docker/.env.example
@@ -1370,3 +1370,6 @@ ENABLE_CLEAN_MESSAGES=false
 ENABLE_MAIL_CLEAN_DOCUMENT_NOTIFY_TASK=false
 ENABLE_DATASETS_QUEUE_MONITOR=false
 ENABLE_CHECK_UPGRADABLE_PLUGIN_TASK=true
+
+# Tenant isolated task queue configuration
+TENANT_ISOLATED_TASK_CONCURRENCY=1
--- a/docker/docker-compose.yaml
+++ b/docker/docker-compose.yaml
@@ -614,6 +614,7 @@ x-shared-env: &shared-api-worker-env
  ENABLE_MAIL_CLEAN_DOCUMENT_NOTIFY_TASK: ${ENABLE_MAIL_CLEAN_DOCUMENT_NOTIFY_TASK:-false}
  ENABLE_DATASETS_QUEUE_MONITOR: ${ENABLE_DATASETS_QUEUE_MONITOR:-false}
  ENABLE_CHECK_UPGRADABLE_PLUGIN_TASK: ${ENABLE_CHECK_UPGRADABLE_PLUGIN_TASK:-true}
+  TENANT_ISOLATED_TASK_CONCURRENCY: ${TENANT_ISOLATED_TASK_CONCURRENCY:-1}

 services:
  # API service