Service status

Where to check live status

status.vauchflow.com is in provisioning. Until it is live, incident notifications are emailed to subscribed customers and posted on this page.

To subscribe to incident notifications, email status@vauchflow.com with the subject “Subscribe to status updates”. We will add you to the incident distribution list and notify you of any affecting incidents as they are identified.

What we monitor

Component	Target	What is measured
API uptime (`api.vauchflow.com`)	99.9% availability	`GET /actuator/health` returns 200
Redemption latency p99	< 200ms	Synthetic redemption probe against a test voucher
Webhook delivery success rate	≥ 99% delivered within 1 hour	Percentage of outbox events reaching `delivered` status over a rolling 24-hour window
MCP server (`api.vauchflow.com/v1/mcp`)	99.9% availability	JSON-RPC `initialize` call responds within 1 second
Dashboard (`app.vauchflow.com`)	99.9% availability	`GET /` returns 200
Database backup currency	Last backup within 24 hours	Prometheus metric from the backup job

The 99.9% availability targets apply to Scale and Enterprise plans. Free, Starter, and Growth plans share the same infrastructure but do not carry a contractual SLA — see pricing for per-plan commitments.

How we communicate incidents

We use three incident states:

Investigating — We have detected or been notified of an issue. We acknowledge it publicly and begin investigation. We do not wait for a root cause before posting.

Identified — Root cause is known. We post what broke, which endpoints and plans are affected, any available workaround, and an estimated resolution time.

Resolved — Service is restored. We describe what we did, note any catch-up window for queued operations (such as delayed webhook delivery), and commit to a postmortem timeline.

Update cadence: We post at minimum every 30 minutes during an open incident, even if the status has not changed.

Postmortem commitment: For any incident where a component reached “down” status for more than 15 minutes, a public postmortem is published within 5 business days.

Support hours

Channel	Hours	Plans
Standard support (`hello@vauchflow.com`)	Monday–Friday, 09:00–18:00 CET	All plans
Priority support	Priority response within 4 business hours	Scale
Dedicated support	Custom SLA agreed in contract	Enterprise
Critical-incident on-call	24/7 best-effort	All plans during active incidents

Vauchflow is EU-built. CET is the primary timezone for standard support.

For non-critical questions: hello@vauchflow.com
For incident subscription: status@vauchflow.com

See pricing for full per-plan response commitments.

Architecture commitments

The following guarantees are baked into the platform and verified by automated tests:

EU residency by default. Compute on Hetzner (Falkenstein/Helsinki), metrics and logs on Grafana Cloud EU West, Redis on Upstash Frankfurt region.
Atomic redemption. The redemption engine is verified by a 100-thread concurrent load test: exactly 1 success and 99 denials for a single-use voucher (max_redemptions=1). Over-redemption is prevented at both the SQL layer (atomic UPDATE with row count check) and a database CHECK constraint.
Flash-sale integrity. The high-traffic redemption path is verified by a 1,000-thread load test using a Redis Lua atomic counter as a pre-gate backed by PostgreSQL as the system of record.
PostgreSQL with row-level security ensuring cross-tenant isolation at the database layer. Daily backups; off-host Backblaze B2 EU backups are on the v1.1 roadmap.
Webhook signing secrets encrypted at rest using AES-256-GCM. Secret material is never stored in plaintext.

Where to read more

Getting started — up and running with the Vauchflow API in under 5 minutes.
API reference — full endpoint and permission surface.
Webhooks — event subscription patterns and delivery guarantees.