Service Level Agreement

This Service Level Agreement ("SLA") defines the availability and performance commitments for the Execlave platform. This SLA applies to paid subscription plans only; free-tier accounts are not covered by this SLA.

1. Service Availability

1.1 Enforcement Endpoint

• Target: 99.9% monthly uptime
• Scope: The /api/v1/policies/enforce endpoint and all policy evaluation infrastructure
• Measurement: Calculated as (total_minutes - downtime_minutes) / total_minutes * 100 over each calendar month (UTC)
• Exclusions: Scheduled maintenance windows (see Section 3)

1.2 Dashboard and Analytics

• Target: 99.5% monthly uptime
• Scope: The web dashboard (/dashboard/*), analytics endpoints (/api/v1/analytics/*), and GraphQL API
• Measurement: Same formula as Section 1.1

2. Performance SLOs

2.1 Enforcement Endpoint Latency

• Target: p99 response time below 4000ms
• Measurement: Measured at the load balancer over rolling 5-minute windows
• Scope: End-to-end latency from request receipt to response delivery for the /api/v1/policies/enforce endpoint
• Note: SDK-side policy caching (default 60-second TTL) and circuit breaker behavior operate independently of this SLO

2.2 Trace Ingestion Latency

• Target: p99 response time below 2000ms
• Measurement: Measured at the load balancer for the /api/v1/traces/ingest endpoint

2.3 Dashboard Page Load

• Target: p95 initial page load below 3000ms
• Measurement: Measured via synthetic monitoring from US-East and EU-West regions

3. Scheduled Maintenance

• Window: Sundays, 02:00 to 06:00 UTC
• Notice: Minimum 48 hours advance notice via email and status page
• Frequency: No more than twice per month
• Impact: During maintenance windows, the enforcement endpoint continues to operate in degraded mode. SDK circuit breakers will activate fail_open behavior for clients configured with that mode.

4. Incident Communication

4.1 Detection and Notification

• Detection: Automated monitoring with alerting via Prometheus and Alertmanager
• Initial notification: Status page update and email notification within 30 minutes of incident detection
• Updates: Status page updates at least every 30 minutes during active incidents

4.2 Severity Levels

4.3 Post-Incident

• Root cause analysis (RCA) published within 5 business days for P1 and P2 incidents
• Remediation actions tracked to completion

5. Credit Policy for SLA Breaches

If monthly uptime falls below the committed target, affected customers are eligible for service credits applied to future invoices.

5.1 Enforcement Endpoint Credits

Credits are calculated as 10% for each 0.1% below the 99.9% target, capped at 30%.

5.2 Dashboard and Analytics Credits

5.3 Credit Request Process

1. Submit a credit request via support within 30 days of the affected month
2. Include the affected dates and times
3. Credits are verified against internal monitoring data and applied within one billing cycle

5.4 Exclusions

Service credits do not apply to:

• Downtime during scheduled maintenance windows
• Force majeure events (natural disasters, government actions, widespread internet outages)
• Issues caused by customer-side configuration (e.g., SDK misconfiguration, expired API keys)
• Features in beta or preview status
• Free-tier accounts

6. Support Channels

Contact

For SLA-related questions or to request service credits: