Skip to content

Scheduled health-check subsystem hard-fails hourly on invalid ANTHROPIC_API_KEY — no preflight validation at pod startup - #2

Draft
langsmith-engine[bot] wants to merge 1 commit into
mainfrom
issues-agent/8452aae2-97d0-4c0d-97a5-f1b827b5069f
Draft

Scheduled health-check subsystem hard-fails hourly on invalid ANTHROPIC_API_KEY — no preflight validation at pod startup#2
langsmith-engine[bot] wants to merge 1 commit into
mainfrom
issues-agent/8452aae2-97d0-4c0d-97a5-f1b827b5069f

Conversation

@langsmith-engine

Copy link
Copy Markdown

Four consecutive scheduled-health-check runs (02:44–05:44 UTC on 2026-07-02) crashed with anthropic.AuthenticationError: 401 - invalid x-api-key inside scheduler.py::_analyse_with_haiku, producing no health reports for four hours. The API service entrypoint (api.py lifespan) does not call the check_env() preflight defined in main.py, so a rotated or invalid ANTHROPIC_API_KEY is not detected at boot — the scheduler starts, runs every 30 minutes, and fails at the Haiku call each time. Every cycle only surfaces a Slack alert via the outer except Exception in _do_check, so cluster monitoring is dark unless an operator sees the alert.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants