Cephfs deploy

Skill sserkanml/ceph-cookbook/skills/cephfs-deploy

This skill should be used when the user asks to deploy CephFS or MDS daemons on a cephadm-managed Ceph cluster — e.g. "deploy cephfs", "create a ceph filesystem", "set up MDS", "mount cephfs". Covers pool creation, MDS deployment, filesystem creation, scoped client auth, and mounting.From its SKILL.md

Install

npx -y skills add sserkanml/ceph-cookbook --skill cephfs-deploy

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.3 KB, as published. Nobody here has run it

CephFS Deployment with cephadm

Overview

Deploying CephFS on an existing cephadm-managed Ceph cluster: create the metadata + data pools, deploy MDS daemons via the orchestrator, create the filesystem, grant a least-privilege client identity, then mount it from a client host.

Placeholders

Placeholder	Meaning	Example
`<FS_NAME>`	Filesystem name	`cephfs`
`<META_POOL>`	Metadata pool name	`cephfs_metadata`
`<DATA_POOL>`	Data pool name	`cephfs_data`
`<PG_NUM>`	PG count for both pools	`64`
`<MDS_COUNT>`	Active MDS rank count (cluster also runs standbys)	`1`
`<PLACEMENT>`	cephadm placement spec for MDS daemons	`count:2`
`<CLIENT_ID>`	cephx client id (without `client.` prefix)	`cephfs-client`
`<MON_HOST>`	Mon host(s)/IPs the client connects to	`mon1,mon2,mon3`
`<MOUNT_POINT>`	Client-side mountpoint	`/mnt/cephfs`

Prerequisites

A healthy cephadm-managed cluster with ceph orch working.
Enough OSD capacity for two new pools.
On the client host: ceph-common (kernel mount, via mount -t ceph) or ceph-fuse.

1. Create the Pools

ceph osd pool create <META_POOL> <PG_NUM>
ceph osd pool create <DATA_POOL> <PG_NUM>

2. Create the Filesystem

ceph fs new <FS_NAME> <META_POOL> <DATA_POOL>
ceph fs ls

3. Deploy MDS Daemons

ceph orch apply mds <FS_NAME> --placement="<PLACEMENT>"
ceph orch ps --daemon-type mds
ceph fs status <FS_NAME>   # wait for at least one MDS in state=active

4. Scale Active Ranks (optional, for large workloads)

ceph fs set <FS_NAME> max_mds <MDS_COUNT>

Keep at least one daemon beyond max_mds as standby (covered by <PLACEMENT> above) — multi-active MDS needs a healthy standby for failover of any given rank.

5. Create a Scoped Client

ceph fs authorize <FS_NAME> client.<CLIENT_ID> / rw \
  -o /etc/ceph/ceph.client.<CLIENT_ID>.keyring

6. Mount (kernel client)

grep key /etc/ceph/ceph.client.<CLIENT_ID>.keyring | awk '{print $NF}' > /etc/ceph/ceph.client.<CLIENT_ID>.secret

mkdir -p <MOUNT_POINT>
mount -t ceph <MON_HOST>:/ <MOUNT_POINT> \
  -o name=<CLIENT_ID>,secretfile=/etc/ceph/ceph.client.<CLIENT_ID>.secret

7. Verify

ceph fs status <FS_NAME>
df -h <MOUNT_POINT>
touch <MOUNT_POINT>/test && ls <MOUNT_POINT>