When To Use Volumes
Quick Example
The Python SDK and CLI snippets below are equivalent ways to create a volume, write data, and mount it into a new machine. Use one control plane unless you are intentionally handing a volume name or machine ID between tools.Mounts Vs Direct File APIs
Volume Tasks
Managing volumes
Create, list, inspect, field-select, and delete volume metadata.
Mounting volumes
Attach volumes at create time or runtime, including read-only and subpath mounts.
Read and write
Read, write, batch-write, move, remove, and edit direct volume files.
File and directory metadata
List directories, stat paths, check existence, search, replace, and watch changes.
Upload data
Upload files, directories, stdin streams, and resumable transfers into volumes.
Download data
Download files, directories, archives, and signed URLs from volumes.
Behavior
Shared volume data is external to VM memory and mutable rootfs snapshots. Hibernate, resume, and fork remount shared volumes; they do not make external storage part of VM memory snapshots. Runtime attachment operations return attachment records withmounted,
released, or failed state. Destroying the machine releases remaining
attachments. Hibernating or pausing a machine releases live mount leases while
the VM is stopped, but it preserves attachment intent and remounts those
volumes on resume.
In hosted private beta, auto-grow is currently disabled and the default cap is
2 GiB per volume unless an operator has applied a tenant-specific override.
There is no public resize API for private beta.