Python: Bug: Python: Infinite Loop in OpenAI Assistants Agent Run

[corytomlinson]

It is possible for an agent to get stuck in an infinite loop while waiting for run_agent to complete using OpenAI Assistants.

When this happens the output loops like this indefinitely.

INFO:httpx:HTTP Request: GET https://.../openai/threads/thread_qHmca1g1DbVJ9pptM8k3a0hW/runs/run_p8Bv3O4ffO68aGGgDA23QzmN/steps?api-version=2024-09-01-preview "HTTP/1.1 200 OK" INFO:semantic_kernel.agents.open_ai.open_ai_assistant_base:Polling run status: run_p8Bv3O4ffO68aGGgDA23QzmN, threadId: thread_qHmca1g1DbVJ9pptM8k3a0hW INFO:httpx:HTTP Request: GET https://.../openai/threads/thread_qHmca1g1DbVJ9pptM8k3a0hW/runs/run_p8Bv3O4ffO68aGGgDA23QzmN?api-version=2024-09-01-preview "HTTP/1.1 200 OK" INFO:semantic_kernel.agents.open_ai.open_ai_assistant_base:Polled run status: incomplete, run_p8Bv3O4ffO68aGGgDA23QzmN, threadId: thread_qHmca1g1DbVJ9pptM8k3a0hW INFO:httpx:HTTP Request: GET https://.../openai/threads/thread_qHmca1g1DbVJ9pptM8k3a0hW/runs/run_p8Bv3O4ffO68aGGgDA23QzmN/steps?api-version=2024-09-01-preview "HTTP/1.1 200 OK" INFO:semantic_kernel.agents.open_ai.open_ai_assistant_base:Polling run status: run_p8Bv3O4ffO68aGGgDA23QzmN, threadId: thread_qHmca1g1DbVJ9pptM8k3a0hW INFO:httpx:HTTP Request: GET https://.../openai/threads/thread_qHmca1g1DbVJ9pptM8k3a0hW/runs/run_p8Bv3O4ffO68aGGgDA23QzmN?api-version=2024-09-01-preview "HTTP/1.1 200 OK" INFO:semantic_kernel.agents.open_ai.open_ai_assistant_base:Polled run status: incomplete, run_p8Bv3O4ffO68aGGgDA23QzmN, threadId: thread_qHmca1g1DbVJ9pptM8k3a0hW

Some traceback that may be helpful...

Traceback (most recent call last): File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.12/asyncio/base_events.py", line 687, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File ".../app.py", line 672, in main results = await run_agent( ^^^^^^^^^^^^^^^^ File ".../app.py", line 544, in run_agent async for content in chat.invoke(): File ".../semantic_kernel/agents/group_chat/agent_group_chat.py", line 144, in invoke async for message in super().invoke_agent(selected_agent): File ".../semantic_kernel/agents/group_chat/agent_chat.py", line 144, in invoke_agent async for is_visible, message in channel.invoke(agent): File ".../semantic_kernel/agents/channels/open_ai_assistant_channel.py", line 64, in invoke async for is_visible, message in agent._invoke_internal(thread_id=self.thread_id): File ".../semantic_kernel/agents/open_ai/open_ai_assistant_base.py", line 767, in _invoke_internal steps_response = await self.client.beta.threads.runs.steps.list(run_id=run.id, thread_id=thread_id)

Expected behavior
An incomplete run should fail after a period of time, not continue to run indefinitely.

Platform

• OS: Ubuntu in WSL2 on Windows 11
• IDE: VS Code
• Language: Python
• Source: pip package version 1.11.0, main branch of repository

Additional context
Add any other context about the problem here.