Availability, support and recovery boundaries

Current service health

Use blinkhost.me/status as the canonical public status page. It currently reports live readiness for the dashboard and control plane. It does not yet provide separate historical availability, regional edge health or per-project health. A green control-plane check is not proof that every deployed application is healthy.

For a production problem, email support@blinkhost.me from an authorized address with the UTC time, workspace and project name, affected public URL, release identifier, visible error and business impact. Do not email passwords, tokens, source code or customer data. BlinkHost does not currently advertise an emergency phone/chat channel or a contractual response time on standard plans.

Service levels

Standard public plans do not include a published uptime SLA, service-credit schedule, RTO or RPO. Enterprise support coverage, response targets, availability commitments and remedies exist only when written into a signed order or agreement. Organizations that require those commitments should complete that agreement before placing the affected workload in production.

What can be recovered

An immutable retained deployment can roll application code back to that retained artifact. This is different from recovering current source, project configuration, secret values or live database rows. Project ZIP and Git exports cover portable source and declarations; they do not contain secret values, provider credentials or live managed-database contents.

Keep a customer-controlled copy of source and secret recovery material. Until a documented live-data export and restore workflow is available for the selected database plan, do not rely on BlinkHost as the only recoverable copy of critical records. Test the actual export, restore and rollback procedure required by the workload before launch.