Resize live instance disk capacity (limited preview)
POST/api/instance/:id/disk-resize
Limited preview; availability requires an enabled organization and an eligible native instance. Set the absolute ext4 root filesystem capacity in MiB in either direction without rebooting. The instance normally remains READY. Track the returned operation: admission is not completion, and the committed instance disk_size changes only on successful finalization. A client timeout does not cancel accepted work. A decrease must retain existing usage, outstanding allocations, and required headroom. Physical storage reclamation and the raw device size are separate from usable filesystem capacity. The plural /api/instances/{id}/disk-resize route is an alias.
Request
Responses
- 200
- 202
- 400
- 401
- 403
- 404
- 409
- 422
- 503
Independently verified same-size no-op or terminal replay. Read status; a terminal replay can preserve a failed result.
Durably accepted pending/running operation, including a replay of unresolved accepted work.
Response Headers
Location string
Inspection route for the original operation.
Invalid body, target, or idempotency key; no resize effects.
Authentication is missing or invalid.
Ownership, organization membership, or account policy denies access.
The instance is not available to this caller.
Conflicting resize or lifecycle operation, stale identity, quota conflict, or a key reused with another target. Inspect error_code before retrying.
The instance does not support managed capacity, or target_below_usage cannot fit usage plus headroom. A measured minimum_target_mib is included when available. No reboot fallback is attempted.
New admission is disabled or a prerequisite is unavailable. Inspection and recovery of previously accepted work remain available.