MDS Service

Deploy CephFS

One or more MDS daemons is required to use the CephFS file system. These are created automatically if the newer ceph fs volume interface is used to create a new file system. For more information, see FS volumes and subvolumes.

For example:

ceph fs volume create <fs_name> --placement="<placement spec>"

where fs_name is the name of the CephFS and placement is a Daemon Placement. For example, to place MDS daemons for the new foo volume on hosts labeled with mds:

ceph fs volume create foo --placement="label:mds"

You can also update the placement after-the-fact via:

ceph orch apply mds foo 'mds-[012]'

For manually deploying MDS daemons, use this specification:

service_type: mds
service_id: fs_name
placement:
  count: 3
  label: mds

The specification can then be applied using:

ceph orch apply -i mds.yaml

See Stateless services (MDS/RGW/NFS/rbd-mirror/iSCSI) for manually deploying MDS daemons on the CLI.

MDS Service

Deploy CephFS

Further Reading