Support & Operations
Incident and Escalation Policy
This page explains how Keldra handles incidents, communication, and escalation for paid customers.
Severity Levels
| Level | Definition | Target |
|---|---|---|
| P1 Critical | Production API unavailable or severe service-wide impact. | Initial response <= 30 minutes |
| P2 Major | Major degradation, elevated errors, or significant latency impact. | Initial response <= 2 hours |
| P3 Minor | Non-critical bug or localized issue with workaround available. | Initial response <= 1 business day |
Incident Workflow
- Detect and page on-call engineer.
- Triage impact, scope, and affected endpoints.
- Mitigate via rollback, isolation, reroute, or scaling.
- Communicate ongoing status and next update ETA.
- Recover and verify service health.
- Publish post-incident review for major incidents.
SLA Exclusions
- Scheduled maintenance announced at least 24 hours in advance
- Emergency security actions needed to protect infrastructure
- Customer-side misconfiguration, misuse, or exhausted quotas/rate limits
- External chain or third-party RPC outages outside Keldra control
- Force majeure events
Escalation Contacts
Operational support and SLA claims: contact@keldra.io
Include account ID, impacted time window (UTC), and request IDs for fast triage.