ceph-volume

Deploy OSDs with different device technologies like LVM or physical devices using pluggable tools (lvm itself is treated like a plugin) and trying to follow a predictable and robust way of preparing, activating, and starting OSDs.

Overview | Plugin Guide

Command Line Subcommands

There is currently support for lvm, and plain devices (with GPT partitions) that may have been deployed with ceph-disk.

zfs support is available for running a FreeBSD cluster.

Node inventory

The inventory subcommand provides information and metadata about a node’s physical device inventory.

Migrating

ceph-disk was deprecated in the Mimic release and has since been removed. ceph-volume is the supported tool for provisioning and managing OSDs. If your cluster still has OSDs that rely on ceph-disk, there are two migration paths:

Keep OSDs deployed with ceph-disk: the simple command takes over their management and disables the old ceph-disk triggers.
Redeploy existing OSDs with ceph-volume: this is covered in depth in Replacing an OSD.

Note

Adopted ceph-disk OSDs continue to run, but the cephadm orchestrator cannot manage them as it does ceph-volume OSDs. Incremental redeployment with ceph-volume is therefore preferred.

For background on why ceph-disk was replaced, see the Replacing ceph-disk section.

New deployments

For new deployments, lvm is recommended. It can use any logical volume for OSDs, or it can set up a minimal logical volume on a device.

Existing OSDs

If the cluster has OSDs that were provisioned with ceph-disk, then ceph-volume can take over the management of these with simple. A scan is done on the data device or OSD directory, and ceph-disk is fully disabled. Encryption is fully supported.

Brought to you by the Ceph Foundation

The Ceph Documentation is a community resource funded and hosted by the non-profit Ceph Foundation. If you would like to support this and our other efforts, please consider joining now.