Uptime Check #629
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # External uptime watch (GH#665, "See & Hear"). Sentry only reports errors | |
| # from a RUNNING app — a crash-looped backend, dead tunnel, or reclaimed node | |
| # is silent there. This cron probes from GitHub's side and opens an issue | |
| # (label: uptime) when a target stays down across the whole probe window, so | |
| # full-site-down produces an email instead of silence. The issue auto-closes | |
| # on recovery, leaving downtime duration in its timeline. | |
| # | |
| # Probe window: 3 attempts, 50s apart (~100s+ span) — sized to outlast the | |
| # ~90s backend Recreate window during Deploy Dev (#654) so routine deploys | |
| # don't page. GitHub cron is best-effort; expect a few minutes of jitter. | |
| name: Uptime Check | |
| on: | |
| schedule: | |
| - cron: '7,22,37,52 * * * *' | |
| workflow_dispatch: {} | |
| permissions: | |
| issues: write | |
| concurrency: | |
| group: uptime-check | |
| cancel-in-progress: false | |
| jobs: | |
| probe: | |
| runs-on: ubuntu-latest | |
| timeout-minutes: 8 | |
| steps: | |
| - name: Probe endpoints | |
| id: probe | |
| run: | | |
| targets=( | |
| "https://commonly.me" | |
| "https://api.commonly.me/api/health" | |
| ) | |
| failed="" | |
| for target in "${targets[@]}"; do | |
| ok=false | |
| for attempt in 1 2 3; do | |
| code=$(curl -sS -o /dev/null -w "%{http_code}" --max-time 20 \ | |
| -H "User-Agent: commonly-uptime-check/1.0" "$target" || echo "000") | |
| echo "$target attempt $attempt -> $code" | |
| if [ "$code" = "200" ]; then ok=true; break; fi | |
| [ "$attempt" -lt 3 ] && sleep 50 | |
| done | |
| if [ "$ok" = "false" ]; then | |
| failed="$failed $target" | |
| fi | |
| done | |
| echo "failed=${failed# }" >> "$GITHUB_OUTPUT" | |
| - name: Open or update outage issue | |
| if: steps.probe.outputs.failed != '' | |
| env: | |
| GH_TOKEN: ${{ github.token }} | |
| FAILED: ${{ steps.probe.outputs.failed }} | |
| run: | | |
| existing=$(gh issue list --repo "$GITHUB_REPOSITORY" --label uptime \ | |
| --state open --json number --jq '.[0].number // empty') | |
| if [ -z "$existing" ]; then | |
| gh issue create --repo "$GITHUB_REPOSITORY" --label uptime \ | |
| --title "🔴 Uptime: $FAILED not responding" \ | |
| --body "$(printf 'External probe failed 3 consecutive attempts (~100s window) for:\n\n%s\n\nRun: %s/%s/actions/runs/%s\n\nThis issue auto-closes when the probe recovers.' "$FAILED" "$GITHUB_SERVER_URL" "$GITHUB_REPOSITORY" "$GITHUB_RUN_ID")" | |
| else | |
| echo "Outage issue #$existing already open — not duplicating." | |
| fi | |
| - name: Close outage issue on recovery | |
| if: steps.probe.outputs.failed == '' | |
| env: | |
| GH_TOKEN: ${{ github.token }} | |
| run: | | |
| existing=$(gh issue list --repo "$GITHUB_REPOSITORY" --label uptime \ | |
| --state open --json number --jq '.[0].number // empty') | |
| if [ -n "$existing" ]; then | |
| gh issue close "$existing" --repo "$GITHUB_REPOSITORY" \ | |
| --comment "Probe recovered — both endpoints returning 200." | |
| fi |