Skip to content

feat: retain independent daily uptime evidence - #79

Merged
simonvanlierde merged 2 commits into
mainfrom
feat/uptime-evidence
Nov 16, 2026
Merged

simonvanlierde merged 2 commits into
mainfrom
feat/uptime-evidence

Conversation

@simonvanlierde

Copy link
Copy Markdown
Contributor

Prometheus retains only 30 days of raw data. This adds an independent probe and daily JSON reports for longer uptime evidence.

  • Dated inventories preserve historical target membership; missed five-minute intervals stay unknown.
  • Immutable observations and reports retain successful and failed intervals without overwriting evidence on retries.
  • HTTP probes enforce a ten-second deadline and collect no response bodies. The guide covers separate-host scheduling and backup.

Tested: just check, followed by final just uptime-test (18 tests), Ruff and ty check. Regression controls detect membership loss, malformed HTTP, incorrect failure exits and missing deadlines. Archived evidence reconstructs at 200 days without raw observations or Prometheus.

Not checked: production probes, scheduler installation or a physical backup destination. Deployment requires an independent Linux host and separate backup storage.

@simonvanlierde
simonvanlierde merged commit fdc4d70 into main Nov 16, 2026
8 checks passed
@simonvanlierde
simonvanlierde deleted the feat/uptime-evidence branch November 16, 2026 01:22
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.

1 participant