Eval run remains queued and never dispatches — 0b1a1ff1-7aec-43b0-a6c7-5043f03939eb
A
Atif M G
Hi Vapi Support,
We are performing a controlled text-layer Eval as part of a provider integration test. The Eval run was created successfully but has remained queued and has not started execution.
Eval Run ID: 0b1a1ff1-7aec-43b0-a6c7-5043f03939eb
Current observations:
- Run state: queued
- Created: 2026-09-18T17:40:58.247Z
- Last provider update: 2026-09-18T17:40:58.430Z
- Falcon/custom LLM endpoint received no request
- No Eval results were generated
- No Vapi charge was incurred
- Account credit balance remained unchanged
- No second Eval has been submitted
- The original run has not been cancelled or modified
- The API reports inconsistent startedAt/endedAt timestamps while authoritative status remains queued; endedAt predates creation
Our Falcon custom LLM endpoint has already been independently validated and is reachable through the deployed TEST route. We are intentionally preserving this Eval run unchanged so the original provider-side state remains available for investigation.
Could you please confirm:
- Why this Eval run has remained queued?
- Whether there is an account, Eval scheduler, transient configuration, or execution prerequisite preventing it from starting.
- Whether you can safely resume/start this existing run without requiring us to create a replacement.
- If the run cannot be resumed, whether you recommend cancelling it before submitting a replacement Eval.
- Whether there is any additional run-level diagnostic information available for this Eval ID.
We would prefer not to submit another Eval until the state of this existing run is understood.
Thank you,
Atif
FalconAgents
Autopilot
Merged in a post:
Eval run remains queued and never dispatches — 0b1a1ff1-7aec-43b0-a6c7-5043f03939eb
A
Atif M G
Hi Vapi Support,
We are performing a controlled text-layer Eval as part of a provider integration test. The Eval run was created successfully but has remained queued and has not started execution.
Eval Run ID: 0b1a1ff1-7aec-43b0-a6c7-5043f03939eb
Current observations:
- Run state: queued
- Created: 2026-09-18T17:40:58.247Z
- Last provider update: 2026-09-18T17:40:58.430Z
- Falcon/custom LLM endpoint received no request
- No Eval results were generated
- No Vapi charge was incurred
- Account credit balance remained unchanged
- No second Eval has been submitted
- The original run has not been cancelled or modified
- The API reports inconsistent startedAt/endedAt timestamps while authoritative status remains queued; endedAt predates creation
Our Falcon custom LLM endpoint has already been independently validated and is reachable through the deployed TEST route. We are intentionally preserving this Eval run unchanged so the original provider-side state remains available for investigation.
Could you please confirm:
- Why this Eval run has remained queued?
- Whether there is an account, Eval scheduler, transient configuration, or execution prerequisite preventing it from starting.
- Whether you can safely resume/start this existing run without requiring us to create a replacement.
- If the run cannot be resumed, whether you recommend cancelling it before submitting a replacement Eval.
- Whether there is any additional run-level diagnostic information available for this Eval ID.
We would prefer not to submit another Eval until the state of this existing run is understood.
Thank you,
Atif
FalconAgents