Decision Service (Policy‑Driven Failover/Burst)

Last updated: 2025-09-22 00:00 UTC

Proves: A policy-governed Decision Service triggers autoscale/failover from SLO + cost/credits + monitor signals.

---

Policy and Orchestration

• What: Decisions come from a versioned policy file; CI/CD executes the chosen scale/failover target.
• Verify: Open the policy commit and the CI run listed in ./links.txt. Confirm the action (target + params) matches the policy.
• Artifacts (./images/): policy_snippet.png, ci_run.png
• KPI tie-in: Autoscale +2@70% (scale-in <40%), DR trigger conditions.

Cloud Signals

• What: Azure Monitor / GCP Monitoring alert → webhook payload → CI run (same correlation_id, aligned timestamps).
• Verify: Follow alert rule → incident → webhook delivery → CI run using ./links.txt. Match timestamps and correlation_id.
• Artifacts (./images/): alert_policy.png, incident.png, webhook_payload.png, ci_run.png, scale_events.png

---

Capture Window (UTC)

• From: YYYY-MM-DDTHH:MMZ To: YYYY-MM-DDTHH:MMZ
• Environment / workspace (non-sensitive): <name>
• Correlation ID(s): CID-YYYYMMDD-XXXX

Redactions

• Mask tenant/subscription/project IDs and any real IPs in screenshots or payloads.
• Keep links read-only; do not expose secrets or tokens.

Navigate: Evidence Map · Proof Archive