[Bug]: React Worker calls sync _process_actions inside async _arun_step_stream

[rc19]

Bug Description

The React Worker calls sync _process_actions inside async _arun_step_stream

llama_index/llama-index-core/llama_index/core/agent/react/step.py

Line 701 in cc77487

reasoning_steps, is_done = self._process_actions(

even when there is the async counterpart _aprocess_actions. This causes errors downstream as the tools are called sync instead of being async.

Version

0.10.28

Steps to Reproduce

llama_index/llama-index-core/llama_index/core/agent/react/step.py

Line 701 in cc77487

reasoning_steps, is_done = self._process_actions(

Relevant Logs/Tracbacks

No response

[dosubot]

To address the issue of calling a synchronous function within an asynchronous context in the React Worker, you should switch from using _process_actions to its asynchronous counterpart _aprocess_actions. This change is necessary to maintain the asynchronous execution flow and avoid errors related to mixing synchronous and asynchronous calls.

Here's the specific change you need to make in your code:

Replace:

reasoning_steps, is_done = self._process_actions(
    task, tools=tools, output=full_response, is_streaming=True
)

With:

reasoning_steps, is_done = await self._aprocess_actions(
    task, tools=tools, output=full_response, is_streaming=True
)

By making this change, you ensure that the execution remains within the asynchronous context, leveraging the await keyword for the _aprocess_actions method. This method is designed for asynchronous operation, aligning with the asynchronous workflow requirements and preventing the issues you've encountered.

If you have any more questions or need further assistance, feel free to ask.

Sources

• llama-index-core/llama_index/core/agent/react/step.py
• llama-index-legacy/llama_index/legacy/agent/legacy/react/base.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.