Antfly Cloud exposes operational controls through first-class Cloud resources so customers can run managed instances without direct Kubernetes access.
Backups
Backups are instance-scoped resources. The control plane stores backup records, then background workers execute the backup against the managed Antfly instance or operator-backed runtime.
Backup status values:
| Status | Meaning |
|---|---|
pending | Backup record created and work queued. |
in_progress | Backup is running. |
completed | Backup finished successfully. |
failed | Backup failed. |
expired | Backup record or object is past retention. |
Backups can be on_demand or scheduled.
Restores
Restores are tracked separately from backups so operators can inspect restore history and status.
Restore jobs move through pending, in_progress, completed, and failed.
Use a completed backup as the source for restore requests. For large restores, expect the instance to report transitional status while data is copied back into the managed runtime.
Backup Schedules
Backup schedules define recurring backup behavior for an instance. Configure schedules for production instances during initial setup, not only after the first incident.
Downloads
Completed backups may expose a generated download URL when the backend storage location supports downloads. Treat download URLs as short-lived sensitive credentials.
Metrics
Cloud metrics expose instance status and usage-oriented values through the Cloud API. Native instance panels should use Cloud endpoints instead of requiring customers to access Kubernetes, Prometheus, or operator internals directly.
Events and Audit Logs
Instance events describe Cloud lifecycle activity such as provisioning, scaling, backups, restores, and failures. Organization audit logs describe administrative activity such as member changes, billing changes, and key management.
Use both views together:
- events answer "what happened to this instance?"
- audit logs answer "who changed Cloud or organization state?"