> ## Documentation Index
> Fetch the complete documentation index at: https://docs.coreweave.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Configure storage mounts on SUNK Self-Service

> Size the home directory and add shared storage mounts on a SUNK Self-Service cluster through spec.storage

[SUNK Self-Service](https://docs.coreweave.com/products/sunk/about-self-service-sunk.md)

A SUNK Self-Service cluster provisions shared storage for you from the `spec.storage` field of the `SunkCluster` resource. You define a home directory and any additional mounts by path and size, and the operator creates the underlying storage and mounts it on every login and compute node. This page shows how to size the home directory, add extra mounts, and change storage after the cluster is running.

This is the SUNK-managed storage path. To mount PVCs that you manage yourself, see [Share storage across Slurm nodes on SUNK Standard](https://docs.coreweave.com/products/sunk/manage_sunk/shared-storage.md). For the full storage configuration reference, including field definitions, see the [`SunkCluster` reference](https://docs.coreweave.com/products/sunk/reference/sunkcluster-reference#storageconfig).

## How managed storage works

Each entry in `spec.storage` becomes one shared volume that the operator provisions as a `ReadWriteMany` PVC and mounts at the path you specify. Because the volumes are `ReadWriteMany`, every login and compute node in the cluster mounts the same storage at the same path, so a file written from one node is visible on all of them.

Data persists across pod restarts and node replacement. The operator never deletes a managed volume: the PVCs remain in the namespace after you delete the `SunkCluster` or remove a mount from `spec.storage`, so you can recover or share the data afterward. For how to find and delete retained volumes, see [Delete retained volumes](#delete-retained-volumes).

## Size the home directory

The `spec.storage.homeDir` field sets the shared home directory. If you omit it, the cluster provisions a `2Ti` volume mounted at `/mnt/home`. To use a different size or path, set the field explicitly:

```yaml theme={"system"}
spec:
  storage:
    homeDir:
      path: /mnt/home
      size: 4Ti
```

When you set `homeDir`, both `path` and `size` are required. The `size` is a [Kubernetes quantity](https://kubernetes.io/docs/reference/kubernetes-api/common-definitions/quantity/), such as `2Ti` or `500Gi`.

## Add additional mounts

Use `spec.storage.additionalMounts` to provision shared storage beyond the home directory, such as a volume for datasets or job output. Each entry is a path and a size:

```yaml theme={"system"}
spec:
  storage:
    homeDir:
      path: /mnt/home
      size: 2Ti
    additionalMounts:
      - path: /mnt/data
        size: 10Ti
      - path: /mnt/scratch
        size: 5Ti
```

Each mount becomes its own `ReadWriteMany` volume available at its path on every node. Mount paths must be unique within the cluster.

## Change storage after creation

To change storage on a running cluster, edit `spec.storage` and reapply the `SunkCluster`. The operator reconciles the difference. The following table summarizes what each kind of change does.

| Change | Result |
| - | - |
| Add an entry to `additionalMounts` | The operator provisions a new volume at the new path. |
| Increase a `size` | The operator requests a volume expansion. This succeeds only if the underlying storage class allows expansion. |
| Decrease a `size` | Not supported. Kubernetes does not allow shrinking a volume, and the change is rejected. |
| Change a `path` | The operator treats the new path as a new, empty volume. The volume at the old path is unmounted from the nodes but retained, with its data. |
| Remove an entry from `additionalMounts` | The volume is unmounted from the nodes but retained, with its data. |

> [!WARNING]
>
> The operator identifies a managed volume by its path. If you change a mount `path`, the data stays on the old volume and doesn't appear at the new path. To move data to a new path, add the new mount, copy the data across from a login node, and only then remove the old mount.

Because the operator names each volume after the cluster and the path, adding a removed path back to `spec.storage` mounts the retained volume again. The same applies when you create a new `SunkCluster` with the same name in the same namespace.

## Delete retained volumes

A retained volume still counts against the CKS cluster's [Distributed File Storage quota](https://docs.coreweave.com/products/storage/distributed-file-storage/manage-quotas.md) until you delete it. The operator doesn't delete it for you.

To list the managed volumes for a cluster, including retained ones, replace `[CLUSTER-NAME]` with the name of the `SunkCluster`. These commands use `tenant-slurm`, the namespace the Cloud Console uses; replace it if your `SunkCluster` is in a different namespace:

```bash theme={"system"}
kubectl get pvc -n tenant-slurm -l sunk.coreweave.com/cluster=[CLUSTER-NAME],app.kubernetes.io/component=volume
```

After you copy any data you need, delete a retained volume by name:

```bash theme={"system"}
kubectl delete pvc [PVC-NAME] -n tenant-slurm
```

Deleting the PVC starts the Distributed File Storage soft-delete window: by default, the data is permanently deleted 24 hours later unless you [recover the PVC](https://docs.coreweave.com/products/storage/distributed-file-storage/recover-deleted-pvc.md) first. Don't delete a PVC whose path is still in `spec.storage`. Kubernetes keeps it in `Terminating` while the nodes mount it. Once they release it, the operator creates a new, empty volume at the same path.

## Next steps

* [`SunkCluster` reference](https://docs.coreweave.com/products/sunk/reference/sunkcluster-reference#storageconfig) for the `StorageConfig` and `VolumeSpec` field definitions.
* [Share storage across Slurm nodes on SUNK Standard](https://docs.coreweave.com/products/sunk/manage_sunk/shared-storage.md) to mount PVCs you provision yourself.
* [Create a SUNK Self-Service cluster](https://docs.coreweave.com/products/sunk/deploy_sunk/create-sunk-cluster.md) for the full cluster creation flow.
