Service Level Agreement

1. Definitions

• "Service" means the HostStack control plane (dashboard, API, CLI, MCP), customer web services deployed through HostStack, managed databases provisioned through HostStack, and the ingress / Traefik routing layer.
• "Monthly Uptime Percentage" is calculated as (total minutes − downtime minutes) / total minutes × 100, measured in calendar months UTC.
• "Downtime" means a period in which the control-plane API returns 5xx responses, customer services on a healthy replica return 5xx through the HostStack ingress, or a managed database is unreachable from its bound service — for at least two consecutive 60-second status snapshots (visible on the public status page).
• "Excluded Downtime" covers scheduled maintenance announced at least 48 hours in advance, force majeure, third-party DNS or upstream network failures outside our infrastructure, customer-caused issues (broken application code, bad migrations, plan limit exhaustion), and outages affecting only the customer's free-tier resources.

2. Availability targets

The following Monthly Uptime Percentages apply per resource tier:

• Free (Nano): Best-effort, no SLA. Idle resources auto-sleep.
• Pico / Pro Standard: 99.9% (≈ 43.2 minutes of allowed downtime per month).
• Pro Large (CCX dedicated): 99.95% (≈ 21.6 minutes per month).
• HA Database (Patroni): 99.95% for the cluster endpoint; per-node downtime during failover is excluded up to 30 seconds.
• Enterprise: 99.99% (≈ 4.3 minutes per month) with custom commitments available on request.

3. Service credits

If we miss the Monthly Uptime Percentage for a resource tier in a given calendar month, you may request a service credit against the affected resources. Credits are applied against the following month's invoice.

• 99.0% – < tier target: 10% of monthly resource fees for affected resources
• 95.0% – 98.99%: 25% of monthly resource fees for affected resources
• Below 95.0%: 50% of monthly resource fees for affected resources

Credits are capped at 50% of the monthly fee for affected resources and are the sole and exclusive remedy for SLA breaches. To request a credit, email support@hoststack.dev within 30 days of the affected month.

4. Support response times

• Free: Community / docs only. No ticket SLA.
• Pico / Pro Standard: First response within 1 business day for support tickets.
• Pro Large: First response within 4 business hours; 1 hour for Severity 1 incidents.
• Enterprise: 24/7 incident response for Severity 1; 1 business day for general inquiries; dedicated Slack / Telegram channel available on request.

Severity 1 means the Service is completely unavailable for your account or a critical feature is broken with no workaround.

5. Maintenance windows

Routine maintenance is performed during low-traffic windows (typically 02:00 – 04:00 UTC on weekends). We announce planned maintenance at least 48 hours in advance via the status page and by email to billing contacts. Maintenance windows are excluded from uptime calculations.

6. Incident communication

Active incidents are posted to the public status page within 10 minutes of detection. Post-incident reviews for Severity 1 events are published within 5 business days of resolution.

7. Changes to this SLA

We may update this SLA from time to time. Material changes will be communicated by email to billing contacts at least 30 days before they take effect. The current version is always available at this URL.