Skip to main content

Storage

Datastores are pools of physical storage capacity in the Cluster Panel: each one is built from disks on one or more compute nodes, and holds the virtual disks (vDisks) that instances actually read and write to — root disks, extra data disks, snapshots, and templates. Virtual disks are the individual storage volumes carved out of that pooled capacity; they're managed both from their parent datastore and from a single fleet-wide Virtual Disks list.

For specific tasks, see Create Datastore under this section.

caution

Draft — needs screenshots. This section hasn't been through the docs-capture pipeline yet, so there are no screenshots. Steps are grounded in the current product code (routes and page components), not a live capture — verify against the running product once this section is captured.

Datastores

Go to Datastores to see every datastore in the cluster, with used/total/ available space per row and cluster-wide Total Storage / Used Storage totals at the top. A Show Reserved Datastores toggle controls whether reserved (system) datastores appear in the list alongside regular ones. Each row's status dot reflects the datastore's state (enabled, deleting, deleted).

From this list, click Create Datastore to launch the datastore-creation wizard (see Create Datastore), or open a datastore's name to view its detail page, which has three tabs:

  • Overview — a storage IOPS chart, a used/total space bar, and an info panel (name, total size, attached resource groups).
  • Physical Disks — the physical disks backing this datastore, shown on a rack/slot diagram with a detail panel for whichever disk is selected.
  • Virtual Disks — the virtual disks currently allocated on this datastore (the same table used by the fleet-wide Virtual Disks page, scoped to this one datastore).

A datastore can be destroyed from its row menu on the index page or from the Manage menu on its detail page. Destroy is only available when the datastore isn't already deleting or deleted, and isn't attached to a resource group; if it still holds orphaned virtual disks, the confirmation warns that deleting the datastore removes those disks too.

Virtual Disks

Go to Virtual Disks for a single list of every virtual disk across all datastores — its type, redundancy health, attached instance (if any), size, parent datastore, and the physical disks it's built from. A Show reserved vdisks toggle reveals system-reserved vdisks alongside regular ones.

There's no dedicated create flow for virtual disks in Storage — they aren't provisioned from here. Selecting physical disks in the datastore wizard adds raw capacity to a datastore's pool; the virtual disks that draw on that capacity come from elsewhere in the product (for example, an instance's root/extra disks), so they simply appear in this list — and on their datastore's Virtual Disks tab — once they exist.

From either list, each virtual disk's row menu can offer:

  • Rename — set a new name.
  • Resize — grow the disk via a slider (up to 5x its current size); disks can't be shrunk.
  • Create Snapshot — copy the disk into a chosen non-reserved datastore that has enough free space.
  • Create Template — save the disk as a reusable instance template.
  • Sync — re-synchronize a redundant disk; only enabled once it has 2+ members and its redundancy health is Degraded or Emergency.
  • Delete — only available for orphaned, non-reserved disks in an Active or Error state.

Redundancy health (Healthy, Syncing, Degraded, Emergency, Enabling, or Disabled) is shown per disk as an icon with a tooltip; configuring redundancy itself isn't part of Storage — that lives on the instance side.