CephFS Deployment Guide on RamNode VPS

Why CephFS on RamNode VPS?

CephFS provides enterprise-grade distributed storage that's particularly valuable for:

Development Teams

Shared storage across multiple environments

High Availability

Applications requiring fault tolerance

No Vendor Lock-in

Avoid cloud storage dependencies

Scalable Storage

Grow storage with demand

Prerequisites and Planning

Before starting, ensure you have:

Server Requirements

• At least 3 RamNode VPS instances
• Minimum 2GB RAM per VPS
• At least 20GB storage per VPS
• Ubuntu 20.04 LTS or newer
• Root access to all instances

→ View Recommended Cloud VPS Plans

Example 5-Node Cluster

• 3 nodes as Ceph monitors + OSDs
• 2 additional nodes as pure OSDs
• All nodes can be CephFS clients
• High availability configuration

Initial VPS Setup

Update all RamNode VPS instances and install essential packages:

Update System on All Nodes

apt update && apt upgrade -y
apt install -y python3 python3-pip curl wget gnupg2

Configure hostnames and update /etc/hosts on all nodes:

Example /etc/hosts Configuration

192.168.1.10 ceph-mon1
192.168.1.11 ceph-mon2
192.168.1.12 ceph-mon3
192.168.1.13 ceph-osd1
192.168.1.14 ceph-osd2

💡 Tip: Set up SSH key authentication between all nodes to simplify cluster management.

Install Ceph

Install Docker on all nodes (required for cephadm):

Install Docker

curl -fsSL https://get.docker.com -o get-docker.sh
sh get-docker.sh
systemctl enable docker
systemctl start docker

Install cephadm on your first monitor node:

Install Cephadm

curl --silent --remote-name --location https://github.com/ceph/ceph/raw/quincy/src/cephadm/cephadm
chmod +x cephadm
./cephadm add-repo --release quincy
./cephadm install

Bootstrap Ceph Cluster

Initialize your Ceph cluster on the first monitor node:

Bootstrap Cluster

cephadm bootstrap --mon-ip 192.168.1.10 --cluster-network 192.168.1.0/24

This command will:

Deploy the first Ceph monitor
Deploy a Ceph manager
Generate cluster keys and configuration
Start the Ceph dashboard on port 8443

🔑 Important: Save the dashboard credentials displayed after bootstrap completion!

Add Additional Nodes

Copy the cluster's public key to other nodes and add them:

Copy SSH Keys

ssh-copy-id root@ceph-mon2
ssh-copy-id root@ceph-mon3
ssh-copy-id root@ceph-osd1
ssh-copy-id root@ceph-osd2

Add Nodes to Cluster

ceph orch host add ceph-mon2 192.168.1.11
ceph orch host add ceph-mon3 192.168.1.12
ceph orch host add ceph-osd1 192.168.1.13
ceph orch host add ceph-osd2 192.168.1.14

Deploy Additional Monitors

ceph orch apply mon "ceph-mon1,ceph-mon2,ceph-mon3"

Configure OSDs (Object Storage Daemons)

OSDs are the storage workhorses of your Ceph cluster.

List Available Devices

ceph orch device ls

Add All Available Devices as OSDs

ceph orch apply osd --all-available-devices

For VPS instances with only the root filesystem, create file-based OSDs:

Create File-Based OSDs

# Create directories for OSD data on each node
mkdir -p /var/lib/ceph/osd-data

# Add file-based OSDs
ceph orch daemon add osd ceph-mon1:/var/lib/ceph/osd-data
ceph orch daemon add osd ceph-mon2:/var/lib/ceph/osd-data
ceph orch daemon add osd ceph-mon3:/var/lib/ceph/osd-data
ceph orch daemon add osd ceph-osd1:/var/lib/ceph/osd-data
ceph orch daemon add osd ceph-osd2:/var/lib/ceph/osd-data

Deploy MDS (Metadata Servers)

CephFS requires metadata servers to manage filesystem metadata. Deploy at least two for redundancy:

Deploy Metadata Servers

ceph orch apply mds cephfs --placement="ceph-mon1,ceph-mon2,ceph-mon3"

📊 High Availability: Multiple MDS servers ensure metadata operations continue even if one server fails.

Create CephFS Filesystem

Create the required pools and filesystem:

Create Data and Metadata Pools

ceph osd pool create cephfs-data 64 64
ceph osd pool create cephfs-metadata 64 64

Create Filesystem

ceph fs new cephfs cephfs-metadata cephfs-data

Verify Filesystem Creation

ceph fs status

✅ Your CephFS filesystem is now created and ready to be mounted!

Set Up CephFS Clients

Install the CephFS client on all nodes that need to mount the filesystem:

Install CephFS Client

apt install -y ceph-fuse

Create a client keyring for mounting:

Create Client Authentication Key

# Create client key
ceph auth get-or-create client.cephfs mon 'allow r' mds 'allow rw' osd 'allow rw pool=cephfs-data, allow rw pool=cephfs-metadata' > /etc/ceph/ceph.client.cephfs.keyring

Mount CephFS

You can mount CephFS using either the kernel client or FUSE client. The kernel client generally offers better performance:

Create Mount Point

mkdir -p /mnt/cephfs

Mount Using Kernel Client

mount -t ceph ceph-mon1:6789,ceph-mon2:6789,ceph-mon3:6789:/ /mnt/cephfs -o name=cephfs,secret=AQC...==

Or Mount Using FUSE Client

ceph-fuse /mnt/cephfs --name client.cephfs

For persistent mounting, add an entry to /etc/fstab:

Add to /etc/fstab

ceph-mon1:6789,ceph-mon2:6789,ceph-mon3:6789:/ /mnt/cephfs ceph name=cephfs,secret=AQC...==,_netdev 0 2

Performance Optimization

Network Optimization

Configure Network Settings

# Edit /etc/ceph/ceph.conf on all nodes
[global]
public_network = 192.168.1.0/24
cluster_network = 192.168.1.0/24
ms_bind_port_min = 6800
ms_bind_port_max = 7300

OSD Optimization

Tune OSD Settings for VPS

ceph config set osd osd_op_threads 2
ceph config set osd osd_disk_threads 1
ceph config set osd osd_recovery_threads 1

Client-Side Caching

Mount with Cache Options

mount -t ceph mon1:6789:/ /mnt/cephfs -o name=cephfs,cache=strict

Monitoring and Maintenance

Using the Ceph Dashboard

Access the web-based dashboard at https://your-monitor-ip:8443 to monitor cluster health, performance metrics, and manage resources.

Command-Line Monitoring

Essential Monitoring Commands

# Check overall cluster health
ceph status

# Monitor OSD status
ceph osd status

# Check filesystem status
ceph fs status

# Monitor cluster performance
ceph -w

Regular Maintenance Tasks

Configure Automated Maintenance

# Enable automatic OSD scrubbing
ceph config set osd osd_scrub_auto_repair true

# Configure deep scrub intervals (weekly)
ceph config set osd osd_deep_scrub_interval 604800

Backup and Recovery Strategies

Snapshot Management

CephFS supports snapshots for point-in-time recovery:

Snapshot Operations

# Create a snapshot
mkdir /mnt/cephfs/.snap/backup-$(date +%Y%m%d)

# List snapshots
ls /mnt/cephfs/.snap/

# Restore from snapshot
cp -r /mnt/cephfs/.snap/backup-20231215/* /mnt/cephfs/restored/

External Backups

Example Backup Script

rsync -av --progress /mnt/cephfs/ /backup/cephfs-$(date +%Y%m%d)/

Scaling Your CephFS Cluster

Adding More OSDs

Add New VPS Node

# Add new VPS node
ceph orch host add ceph-osd3 192.168.1.15

# Deploy OSD on new node
ceph orch daemon add osd ceph-osd3:/var/lib/ceph/osd-data

Adding More MDS Servers

Scale MDS Daemons

# Scale MDS daemons for better metadata performance
ceph orch apply mds cephfs --placement="5"

Conclusion

Deploying CephFS on RamNode VPS instances provides a cost-effective way to build enterprise-grade distributed storage. While the initial setup requires careful planning and configuration, the result is a robust, scalable filesystem that can grow with your needs.

The combination of Ceph's proven technology and RamNode's reliable VPS infrastructure creates an excellent foundation for applications requiring distributed storage. With proper monitoring, maintenance, and security practices, your CephFS cluster will provide years of reliable service.

Remember that distributed storage systems require ongoing attention and maintenance. Regular monitoring, timely updates, and proper backup procedures are essential for maintaining a healthy cluster. Start small, understand the system thoroughly, and scale gradually as your requirements grow.

Whether you're building a development environment, supporting a growing application, or creating a resilient storage solution, CephFS on RamNode VPS offers the flexibility and reliability needed for modern storage requirements.

Related Resources

Cloud VPS PlansHigh-performance VPS for CephFS clusters More Deployment GuidesExplore other infrastructure guides GlusterFS GuideAlternative distributed filesystem Get SupportNeed help? Contact our team

Deploying CephFS on RamNode VPS