Abnormal Codex Usage #42492

Activity

1. hellojazlynwang-lgtm commented on Sep 3, 2026

   

   hellojazlynwang-lgtm

   on Sep 3, 2026

   Author

   More actions

   gpt version 26.831.21537
2. github-actions

   added

   bugSomething isn't workingSomething isn't working

   rate-limitsIssues related to rate limits, quotas, and token usage reportingIssues related to rate limits, quotas, and token usage reporting

   on Sep 3, 2026
3. github-actions commented on Sep 3, 2026

   

   github-actionsbot

   on Sep 3, 2026 – with GitHub Actions

   Contributor

   More actions

   Potential duplicates detected. Please review them and close your issue if it is a duplicate.

   • GPT-5.6 Sol High consumed ~60% of my 5-hour Plus limit in about 6 minutes #41468
   • Pro 20x weekly Codex quota draining abnormally fast during small VS Code bug fixes #42127
   • [Bug] ChatGPT Plus quota consumption increased ~9x between comparable Codex App tasks #41957
   • Codex quota consumption: same 11% usage band consumed with 52% less observed token-weighted work #42448
   • [Meta] Abnormal Codex usage/quota depletion and usage-accounting inconsistencies — cross-report tracker #41220

   Powered by Codex Action
4. filsta1 commented on Sep 3, 2026

   

   filsta1

   on Sep 3, 2026

   More actions

   I experienced the same issue on September 3, 2026.

   A Codex CLI agent running on my OMV/Docker server consumed a large amount of my 5-hour and weekly usage limits without completing the requested work.

   The incident was caused by orchestration/configuration failures:

   • the agent was started despite being asked to stop;
   • multiple systemd start commands were queued;
   • supervisors received an empty prompt because of shell quoting;
   • the runner incorrectly used grep -qx and rejected valid supervisor reports;
   • Browserless access was not verified before the agent was started;
   • the agent ended with exit code 1 and did not complete the website task.

   The service was limited to 1 CPU, 2.5 GB RAM and a 4-hour timeout, but the issue was unnecessary model/token usage caused by the runner workflow.

   Please investigate the usage accounting and consider restoring the consumed Codex quota. I can provide systemd timestamps and logs privately if needed.
5. bompus commented on Sep 7, 2026

   

   bompus

   on Sep 7, 2026

   More actions

   feedback id 01a07a0c-95f6-7da2-a6bf-113be168b169