o3-deep-research expires without reason

[agdhruv]

• Package Name: "azure-ai-projects>=1.0.0b12",
• Package Version: "azure-ai-projects>=1.0.0b12",
• Operating System: Windows 11
• Python Version: 3.12

Describe the bug
The deep research agent unexpectedly fails sometimes with run.status as expired. There's no way to extend the expiration period. run.last_error, run.incomplete_details, run.expires_at, run.cancelled_at, run.failed_at are all blank, so there's no information about the problem so that I can fix it.

To Reproduce
Not reproducible; seems to happen if the deep research runs too long. Using the started code give here.

Expected behavior
Deep research completed successfully, provides information about why it expired, and ideally, provide a way to avoid the expiration when a long deep research run is expected.

Screenshots
N/A

Additional context
N/A

[github-actions]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @dargilco @glharper @howieleung @jhakulin @nick863 @trangevi.

[buhlj]

We have the exact same issue, so are quite interested in knowing what are the plans to fix this? As it prohibit us to use it as it always expires which happens after approximately 30 min.

Also no information is provided towards why it expired

[alvaroborras]

The expiration happens because Deep Research hits a requires_action state waiting for tool approval (SubmitToolApprovalAction). The execution loop just keeps polling until the run times out. This can solved by auto‑approving these requested tool calls.