Advice for preparing talos machines for ceph backed storage?

[Drakonis]

Hi,

I've been looking into using the set of VolumeConfig resources to partition my Talos workers to deploy Ceph, VolumeConfig can successfully resize the EPHEMERAL partition, however, when i use RawVolumeConfig, it does not produce any raw filesystems that can be consumed by Ceph without any manual intervention, instead producing a generic linux filesystem, is there any way to automate the setup using talos? or would i have to set up a Job or DaemonSet with privileged Pods that can execute a script for this?

I'm using the following manifests for handling the volumes

---
apiVersion: v1alpha1
kind: VolumeConfig
name: EPHEMERAL
provisioning:
    diskSelector:
        match: disk.transport == "virtio"
    minSize: 50GiB
    maxSize: 50GiB
    grow: false
---
apiVersion: v1alpha1
kind: RawVolumeConfig
name: ceph-data
provisioning:
    diskSelector:
        match: disk.transport == "virtio"
    minSize: 10GiB
    grow: true

[smira]

See #11778

Make sure name doesn't contain ceph.

[Drakonis]

Thank you, that did the trick.