The session kept spinning and thinking non‑stop for three hours with no valid output. Massive compute quota was wasted on idle looping. Could this be a backend bug? My‑weekly usage allowance was exhausted as a result. #207790
Unanswered
lei123666
asked this question in
Other Feature Feedback, Questions, & Ideas
Replies: 1 comment
|
💬 Your Product Feedback Has Been Submitted 🎉 Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users. Here's what you can expect moving forward ⏩
Where to look to see what's shipping 👀
What you can do in the meantime 💻
As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities. Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐ |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
🏷️ Discussion Type
Product Feedback
💬 Feature/Topic Area
Other
Body
Hello, here is the requested log evidence. I redacted the full user prompt, internal developer instructions, encrypted reasoning blobs, account details, and unrelated workspace configuration. Timestamps, IDs, event types, models, token counts, abort reason, and durations are unchanged.
Source session file:
rollout-2026-09-13T09-08-45-01a0984f-3da1-7b92-9b8b-2928a5be4e79.jsonl
Local source line numbers are included. Timestamps below are UTC (add eight hours for UTC+8).
1111 {"timestamp":"2026-09-13T04:32:39.693Z","type":"event_msg","payload":{"type":"item_completed"}}
1112 {"timestamp":"2026-09-13T04:32:55.339Z","type":"event_msg","payload":{"type":"item_completed"}}
1113 {"timestamp":"2026-09-13T04:32:55.349Z","type":"response_item","payload":{"type":"reasoning","encrypted_content":"[REDACTED]"}}
1114 {"timestamp":"2026-09-13T04:33:01.315Z","type":"event_msg","payload":{"type":"item_completed"}}
1115 {"timestamp":"2026-09-13T04:33:01.326Z","type":"response_item","payload":{"type":"reasoning","encrypted_content":"[REDACTED]"}}
1116 {"timestamp":"2026-09-13T04:33:34.237Z","type":"event_msg","payload":{"type":"item_completed"}}
1117 {"timestamp":"2026-09-13T04:33:34.246Z","type":"response_item","payload":{"type":"message","role":"assistant","content":"[PREVIOUS TURN RESPONSE REDACTED]"}}
1118 {"timestamp":"2026-09-13T04:33:34.477Z","type":"token_usage_record","payload":{"thread_id":"01a0984f-3da1-7b92-9b8b-2928a5be4e79"}}
1119 {"timestamp":"2026-09-13T04:33:34.484Z","type":"event_msg","payload":{"type":"token_count"}}
1120 {"timestamp":"2026-09-13T04:33:34.506Z","type":"event_msg","payload":{"type":"task_complete"}}
1121 {"timestamp":"2026-09-13T05:46:28.810Z","type":"event_msg","payload":{"type":"thread_settings_applied","thread_id":"01a0984f-3da1-7b92-9b8b-2928a5be4e79","thread_settings":{"model":"gpt-5.6-sol","service_tier":"priority","reasoning_effort":"high"}}}
1122 {"timestamp":"2026-09-13T05:46:28.881Z","type":"event_msg","payload":{"type":"task_started","turn_id":"01a0994d-8236-7ad2-bf6f-07577a714a18","started_at":1789278388,"model_context_window":258400}}
1123 {"timestamp":"2026-09-13T05:46:33.904Z","type":"turn_context","payload":{"turn_id":"01a0994d-8236-7ad2-bf6f-07577a714a18","model":"gpt-5.6-sol","effort":"high","timezone":"Asia/Shanghai"}}
1124 {"timestamp":"2026-09-13T05:46:33.932Z","type":"response_item","payload":{"type":"message","role":"user","content":"[USER MESSAGE REDACTED]"}}
1125 {"timestamp":"2026-09-13T05:46:33.940Z","type":"event_msg","payload":{"type":"item_completed","thread_id":"01a0984f-3da1-7b92-9b8b-2928a5be4e79","turn_id":"01a0994d-8236-7ad2-bf6f-07577a714a18","item":{"type":"UserMessage","content":"[USER MESSAGE REDACTED]"}}}
1126 {"timestamp":"2026-09-13T09:19:53.210Z","type":"response_item","payload":{"type":"message","role":"developer","content":"<turn_aborted> The previous turn was interrupted on purpose. Any running unified exec processes may still be running in the background. If any tools/commands were aborted, they may have partially executed. </turn_aborted>","internal_chat_message_metadata_passthrough":{"turn_id":"01a0994d-8236-7ad2-bf6f-07577a714a18","create_time":1789291193.2109382,"content_item_kinds":["generic.turn_aborted"]}}}
1127 {"timestamp":"2026-09-13T09:19:53.221Z","type":"event_msg","payload":{"type":"turn_aborted","turn_id":"01a0994d-8236-7ad2-bf6f-07577a714a18","reason":"interrupted","started_at":1789278388,"completed_at":1789291193,"duration_ms":12804365}}
1128 {"timestamp":"2026-09-13T09:20:07.739Z","type":"event_msg","payload":{"type":"thread_settings_applied","thread_id":"01a0984f-3da1-7b92-9b8b-2928a5be4e79"}}
1129 {"timestamp":"2026-09-13T09:20:07.772Z","type":"event_msg","payload":{"type":"task_started"}}
1130 {"timestamp":"2026-09-13T09:20:12.794Z","type":"turn_context","payload":{"model":"gpt-5.6-sol","effort":"high"}}
1131 {"timestamp":"2026-09-13T09:20:12.818Z","type":"response_item","payload":{"type":"message","role":"user","content":"[FOLLOW-UP MESSAGE REDACTED]"}}
1132 {"timestamp":"2026-09-13T09:20:12.829Z","type":"event_msg","payload":{"type":"item_completed"}}
1133 {"timestamp":"2026-09-13T09:25:43.281Z","type":"event_msg","payload":{"type":"thread_settings_applied","thread_id":"01a0984f-3da1-7b92-9b8b-2928a5be4e79"}}
1134 {"timestamp":"2026-09-13T09:25:43.288Z","type":"event_msg","payload":{"type":"task_started"}}
1135 {"timestamp":"2026-09-13T09:25:58.293Z","type":"turn_context","payload":{"model":"gpt-5.6-sol","effort":"high"}}
The affected turn ran from 2026-09-13 13:46:28 to 17:19:53 UTC+8 (12,804,365 ms, approximately 3 hours 33 minutes). The recorded reason is "interrupted". There is no timeout, disconnect, server_error, request_error, or response_error record in this range. There is also no token_usage_record for the aborted turn between its task_started and turn_aborted records.
Source session file:
rollout-2026-09-13T10-47-27-01a098a9-9a1d-7570-bf44-5fec18936b8a.jsonl
1609 {"timestamp":"2026-09-13T20:00:07.875Z","type":"event_msg","payload":{"type":"task_started","turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406","started_at":1789329607,"model_context_window":121600}}
1610 {"timestamp":"2026-09-13T20:00:07.889Z","type":"turn_context","payload":{"turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406","model":"gpt-5.3-codex-spark","effort":"high"}}
1613 {"timestamp":"2026-09-13T20:00:10.455Z","type":"event_msg","payload":{"type":"item_completed","turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406","item":{"type":"Reasoning","summary_text":[],"raw_content":[]}}}
1615 {"timestamp":"2026-09-13T20:00:10.626Z","type":"event_msg","payload":{"type":"item_completed","turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406","item":{"type":"AgentMessage","content":"[SHORT UNRELATED RESPONSE]"}}}
1616 {"timestamp":"2026-09-13T20:00:10.638Z","type":"response_item","payload":{"type":"message","role":"assistant","content":"[SHORT UNRELATED RESPONSE]","internal_chat_message_metadata_passthrough":{"turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406"}}}
1617 {"timestamp":"2026-09-13T20:00:10.646Z","type":"token_usage_record","payload":{"thread_id":"01a098a9-9a1d-7570-bf44-5fec18936b8a","turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406","usage":{"input_tokens":27820,"cached_input_tokens":11776,"output_tokens":473,"reasoning_output_tokens":416,"total_tokens":28293}}}
1619 {"timestamp":"2026-09-13T20:00:10.669Z","type":"event_msg","payload":{"type":"task_complete","turn_id":"01a09c5b-0cb4-76a2-9430-27f751e06406","started_at":1789329607,"completed_at":1789329610,"duration_ms":2793}}
Correction to my earlier description: this turn was not an empty string in the local log. It returned a short, unrelated response that did not address the delegated task. The UI appeared to make no task progress. There is no timeout, disconnect, server_error, request_error, or response_error record in this turn range.
Source session file:
rollout-2026-09-13T10-47-12-01a098a9-6077-7661-8340-22a67e5fe69f.jsonl
2188 {"timestamp":"2026-09-13T20:48:17.248Z","type":"event_msg","payload":{"type":"task_started","turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034"}}
2189 {"timestamp":"2026-09-13T20:48:17.267Z","type":"turn_context","payload":{"turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034","model":"gpt-5.6-sol","effort":"high"}}
2192 {"timestamp":"2026-09-13T20:48:21.391Z","type":"event_msg","payload":{"type":"item_completed","turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034","item":{"type":"Reasoning","summary_text":[],"raw_content":[]}}}
2194 {"timestamp":"2026-09-13T20:48:21.550Z","type":"event_msg","payload":{"type":"item_completed","turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034","item":{"type":"AgentMessage","content":[{"type":"Text","text":""}],"phase":"final_answer"}}}
2195 {"timestamp":"2026-09-13T20:48:21.556Z","type":"response_item","payload":{"type":"message","role":"assistant","content":[{"type":"output_text","text":""}],"phase":"final_answer","internal_chat_message_metadata_passthrough":{"turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034"}}}
2196 {"timestamp":"2026-09-13T20:48:21.645Z","type":"token_usage_record","payload":{"thread_id":"01a098a9-6077-7661-8340-22a67e5fe69f","turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034","usage":{"input_tokens":173471,"cached_input_tokens":172416,"output_tokens":121,"reasoning_output_tokens":115,"total_tokens":173592}}}
2198 {"timestamp":"2026-09-13T20:48:21.666Z","type":"event_msg","payload":{"type":"task_complete","turn_id":"01a09c87-2331-7bb1-a4e3-6266d3ce8034","last_agent_message":null,"started_at":1789332497,"completed_at":1789332501,"duration_ms":4438}}
This turn has an explicitly empty AgentMessage and response_item, while recording 173,592 total tokens, including 172,416 cached input tokens. There is no timeout, disconnect, server_error, request_error, or response_error record in this turn range.
The current Codex CLI header/footer displays:
gpt-5.6-sol low fast - E:\太空大亨
The local config contains:
service_tier = "priority"
I attempted
/fast statusin Codex CLI v0.154.0-alpha.6.2. In the TERM=dumb terminal capture it was submitted as an ordinary model prompt instead of producing a standalone slash-command result, so I interrupted it. The visible CLI session indicator nevertheless explicitly says "fast". The earlier affected long-turn record also shows service_tier "priority".Please correlate these records with server-side metering and investigate whether the 3-hour aborted turn and the final empty-message turn were counted incorrectly.
Diagnostic note: during collection, the ordinary weekly meter changed from 79% used to 81% used. This interval included the present support-analysis turn and the interrupted CLI attempt described above, so I cannot attribute the two-point change to one operation.
All reactions