MayaNAS On-Prem (qcow2/KVM)

Deploy MayaNAS on your own infrastructure using the qcow2 virtual machine image with KVM, libvirt, or Proxmox.

Prerequisites

Linux host with KVM/QEMU support (kvm-ok should return positive)
libvirt and virt-install (or Proxmox VE)
At least 4 vCPUs, 8 GB RAM for the VM
Network connectivity (bridge or NAT)
A local or network-attached SSD for metadata (recommended)
S3-compatible object storage backend (MinIO, Ceph RGW, or cloud S3/GCS)

Step 1: Download the qcow2 Image

Obtain the MayaNAS qcow2 image from ZettaLane. The image contains the full MayaNAS stack pre-installed.

# Example (replace URL with actual download link)
wget https://releases.zettalane.com/mayanas/mayanas-latest.qcow2

Step 2: Deploy the VM

Option A: virt-install (libvirt)

virt-install \
  --name mayanas-node1 \
  --ram 8192 \
  --vcpus 4 \
  --disk path=/var/lib/libvirt/images/mayanas-node1.qcow2,format=qcow2 \
  --import \
  --os-variant rocky9 \
  --network bridge=br0 \
  --graphics none \
  --console pty,target_type=serial

Option B: Proxmox VE

Upload the qcow2 to Proxmox storage
Create a new VM (Linux, 4+ vCPUs, 8+ GB RAM)

Import the qcow2 as the boot disk:

qm importdisk <VMID> mayanas-latest.qcow2 local-lvm

Attach the imported disk as SCSI and set it as boot device
Start the VM

Option C: QEMU direct

qemu-system-x86_64 \
  -enable-kvm \
  -m 8192 \
  -smp 4 \
  -drive file=mayanas-latest.qcow2,format=qcow2 \
  -nic bridge,br=br0 \
  -nographic

Step 3: First Boot Configuration

On first boot, the VM obtains an IP via DHCP (or configure static IP).

Find the VM IP from your DHCP server or hypervisor console
SSH into the VM:
Terminal window
```
ssh mayanas@<VM_IP>
```
Run initial setup (if not already configured):
Terminal window
```
sudo /opt/mayastor/config/standalone_setup.sh
```

Step 4: Configure Object Storage Backend

MayaNAS needs an S3-compatible backend. Configure it via the Web UI or CLI:

# Example: Configure MinIO backend
export MAYANAS_S3_ACCESS_KEY="minioadmin"
export MAYANAS_S3_SECRET_KEY="minioadmin"
export MAYANAS_S3_BUCKET="mayanas-data"
export MAYANAS_S3_ENDPOINT="http://minio.local:9000"

Supported backends:

MinIO — self-hosted S3-compatible storage
Ceph RGW — S3 gateway for Ceph clusters
AWS S3 — direct cloud backend
GCS — via S3-compatible interop API

Step 5: Access the Web UI

http://<VM_IP>:2020

Step 6: Add Metadata Disk (Optional)

For better performance, attach an SSD to the VM and configure it as the metadata disk:

Attach an SSD disk to the VM (via hypervisor)
The disk appears as /dev/vdb or /dev/sdb
Configure in MayaNAS Web UI under Storage settings

Step 7: Create NFS/SMB Shares

Use the Web UI or CLI to create shares, then mount from clients:

sudo mount -t nfs <VM_IP>:/<POOL>/<SHARE> /mnt/data

HA Deployment (2-Node Cluster)

For high availability, deploy two VMs and configure clustering:

Deploy two VMs following steps 1–4 above
Ensure both VMs can reach each other on the network
Configure clustering via the Web UI or use the cluster setup script:
Terminal window
```
sudo /opt/mayastor/config/cluster_setup.sh
```
The VIP will float between the two nodes for automatic failover

Troubleshooting

# Check MayaNAS logs
tail -f /opt/mayastor/logs/mayanas-terraform-startup.log

# Service status
systemctl status mayastor

# Check ZFS pool status
zpool status

# Check NFS exports
showmount -e localhost