Get started

nf ceph

The nf ceph command is the entrypoint for all Ceph cluster management. It provides subcommands for creating and configuring the core Ceph daemons that make up a storage cluster.

Use these commands after nodes have been registered and disks are visible via nf disks list.

Command reference

CommandFlagsDescription
nf ceph fs--name --metadataPool --dataPoolCreate a CephFS filesystem
nf ceph mon--node --diskCreate a Ceph Monitor
nf ceph osd--node --diskCreate an OSD on a disk
nf ceph mds--nodeCreate a Metadata Server
nf ceph mgr--nodeCreate a Manager daemon
nf ceph rgw listList RGW instances
nf ceph rgw deploy--nodeDeploy an RGW instance
nf ceph rgw remove--idRemove an RGW instance
nf ceph rgw user listList RGW users
nf ceph rgw user create--id --name --emailCreate an RGW user

Subcommands

fs

Create a CephFS filesystem with metadata and data pools.

mon

Create a Monitor for cluster consensus and map management.

osd

Create an OSD to provide storage capacity on a disk.

mds

Create a Metadata Server for CephFS.

mgr

Create a Manager for metrics and orchestration.

rgw

Manage RADOS Gateway instances and S3 users.

Typical deployment order

A standard Ceph cluster deployment follows this sequence:

  1. Monitors — deploy 3 or 5 for quorum (nf ceph mon)
  2. Managers — at least 1 for cluster health (nf ceph mgr)
  3. OSDs — one per available disk (nf ceph osd)
  4. MDS — if you need CephFS (nf ceph mds)
  5. CephFS — create the filesystem (nf ceph fs)
  6. RGW — if you need S3 (nf ceph rgw deploy)