Has anyone else noticed recently that requests with `background: true` in the Responses API are getting stuck in ‘queued’ status indefinitely?
The model simply does not provide an answer, even the logs are empty.
And sometimes it gives an error:
The request timed out. You can retry your request, or contact us through our help center at help.openai.com if the error persists.
Im seeing the exact same thing!
Yes, either stuck in queued or the requests time out
Same problem here since this morning
It stays in_queue for longer than usual then transitions to in_progress then failed.
It looks like a time_out issue:
Response_ids:
- resp_04e5c27ebae29b8200690391bbfce4819086d2591fe760a350
- resp_0572f4a507cb3ec70069038fc3361c81959010908289aa0dd7
- resp_0339b82e14cd672100690389a92818819784a3204c067e4732
1 Like
Seeing the same thing on a large scale project, across the board.
I’ve been getting the same today. It happened before a few months ago as well. Random failures for a few hours then gets cleared up.
@OpenAI_Support any idea what’s going on here!!
@OpenAI_Support same here, Reponse API in background mode is broken. Mosly jobs are stuck in queue, in_progress (many minutes) or just fail. It has caused a mess in out system today and OpenAI Status is green in the API. Something is wrong there, now seeing many reports in this thread.
2 Likes
Same issues as OP and others - even after 17 hours. How do we even contact OpenAI about a critical issue like this? @OpenAI_Support
Seeing the same issues since yesterday as well.
@OpenAI_Support Perhaps you could update the status page at least?
Unless anybody here has an Enterprise plan / some other way of escalating it?
Seeing exactly the same issue. Have tried tweaking all sorts of params because I’m time pressed, but no luck. Incredibly frustrating. Hope they sort it out ASAP.
1 Like
same problem. (Besides many other problems) I wonder, wether anyone uses openAI for production
1 Like
Same issue. Finally addressed in the status page saying the issue has been mitigated. I am still receiving failed requests with tooling attached to the the Responses API request (MCP server, Code Interpreter, …)
1 Like
Indeed there an issue in monitoring but I still getting errors as before.
1 Like
We continue to face long queues and timeouts. This issue is severely impacting our operations… 
2 Likes
Certainly a major feature in our app is down for 24 hours already.
@OpenAI_Support havent seen any changes yet status page is saying “The traffic has stabilised and the incident is mitigated. We keep monitoring.” what is going on??
2 Likes
Status page is reporting the issue was resolved, but still receiving failed requests with tooling attached to the the Responses API request (MCP server, Code Interpreter, …)
1 Like