Public Beta

MFConsole Downloads

Two pieces: the console (the control plane you run once) and the node ISO (flashed onto every KVM host it manages).

Console Node ISO

Deploy the Console

MFCloud Console is the control plane for the MFCloud platform — an enterprise KVM hypervisor manager. One image runs the web console, Celery worker, and beat scheduler (selected by command: in the shipped compose files). It manages VM lifecycle and live migration, Ceph/LINSTOR/ZFS storage, OVN networking, Kubernetes clusters, and multi-tenant RBAC, with optional HA via Patroni/TimescaleDB.

Public beta. Runs on the Free edition out of the box (2 nodes / 10 VMs, no license key needed) — paste a license key in Settings → Edition & License later to unlock higher caps and paid features (HA, S3 backup, multi-tenant, LDAP, and more). KVM hosts are added from its UI afterward — see the Node ISO below.

MFConsole 0.45.5.3

Released 2026-09-08

Quick Start

Pull the image, extract the deploy files with cp (no login, no separate download needed), and run the installer — works with Docker or Podman:

docker pull mfconsole/kvm-manager:latest
id=$(docker create mfconsole/kvm-manager:latest)
docker cp "$id":/app/deploy-files ./mfconsole
docker rm "$id"
cd mfconsole
./install.sh

podman pull docker.io/mfconsole/kvm-manager:latest
id=$(podman create docker.io/mfconsole/kvm-manager:latest)
podman cp "$id":/app/deploy-files ./mfconsole
podman rm "$id"
cd mfconsole
./install.sh

install.sh auto-detects Docker or Podman on the host (pass --engine docker or --engine podman to force one) — handles .env setup, secret generation, and bringing the stack up, then prints the console URL and login when it's done. Requires Docker (Compose v2.24+) or Podman (podman compose or podman-compose).

What's New — 0.45.5.3 (2026-09-08)

Fixed a false "needs repair" badge on compute nodes updated by the console's own automatic sync, even when fully up to date. Only the manual Repair Worker button was recording the version it pushed; the automatic path now does too.
Fixed a healthy Ceph cluster showing as unreachable on the dashboard on newer Ceph releases. The cluster itself was healthy the whole time — only the dashboard display was affected.
New AI Assistant chat panel, with per-tenant usage tracking and an optional monthly spend ceiling. Long conversations now compact automatically instead of growing without bound. A follow-up review pass fixed a few edge cases (mid-summarization message loss, an "Explain this task" usage-limit bypass) before wide use.
Fixed a bug where a failed VM creation could permanently strand a reserved GPU, recoverable only by retrying under the exact same VM name. A new admin option manually releases a stuck GPU reservation if needed.
Compute nodes now report their own software version, and the Hosts table flags any node out of sync with the console. Click the badge to run Repair Worker directly instead of inferring it from release notes.
Control-plane backup and restore is now reachable from the out-of-band HA Manager (:9444) — list backups and trigger a full restore even with the main console down.
Automated, verified control-plane backups. The console's own database and config (not just your VMs) now back up nightly to your configured S3 target, encrypted the same way VM backups already are — and every run is proven restorable by an actual test-restore into a throwaway database, not just a check that the backup job exited cleanly.
Every identity/tenancy change is now audited. User creation, role grants, and tenant changes now write a full audit-log entry, answering "who changed this, and when."
Known issue (3-node HA only): losing a control-plane host can still freeze the surviving console replicas for a few minutes. The infrastructure layer fails over correctly every time — VIP, database promotion and workers all recover in seconds. Restarting the affected app container restores service in about 10 seconds. Single-node installs are unaffected; a stack-dump profiler (py-spy) ships in the image to capture the cause on the next occurrence.

First login

install.sh prints the exact URL and login at the end of the run. By default: open https://<your-master-ip>/ (self-signed TLS — accept the browser warning). Username admin, password mfpro — the console forces a password change before the dashboard loads.

Download MFConsole Node ISO

The all-in-one Rocky-based hypervisor installer. Flash it, boot a server, and enroll the node into your MFConsole control plane. Two builds — pick the one matching your hardware.

Standard Hypervisor el9

Rocky Linux 9.8 (el9) · x86-64 · UEFI · For regular KVM hosts (no GPU/SR-IOV passthrough) · Released 2026-07-21 · ~4.89 GB

Download ISO

SHA-256 checksum

c481f2be8e669c20ecb8b4300ec73db8e0cb2bf7bb1f1b5350b69e38b0606166

GPU & SR-IOV Hypervisor (el10, GPU)

Rocky Linux 10.2 (el10) · x86-64 · UEFI · For hosts with GPU passthrough / SR-IOV NICs · Released 2026-06-08 · ~2.94 GB

Download ISO

SHA-256 checksum

2f257bbd5ba3c65ede5b396cf9e4c2202d3dd328a4a4e77faaf467c8450e0d4d

Verify before you flash

Confirm the download wasn't corrupted or tampered with. Run the checksum check for whichever ISO you downloaded.

Checksum (integrity)

sha256sum mfcloud-min-el9.iso        # standard hypervisor
sha256sum mfcloud-v40-el10-gpu.iso   # GPU / SR-IOV hypervisor
# compare the output against the matching SHA-256 shown above

Quick start

1.Write the ISO to a USB stick on Windows with Rufus (see warning below), or on Linux/Mac with balenaEtcher or dd — or mount it directly via iDRAC/iLO virtual media.
2.Boot the target server (UEFI). The installer runs unattended and reboots into the node agent.
3.From the MFConsole UI, Add Host with the node's IP to enroll it into your cluster.
4.Full walkthrough in the Install a Node guide.

Using Rufus? Select "DD Image mode" — not "ISO Image mode"

Rufus defaults to "ISO Image mode (Recommended)" for this ISO, but that repartitions the drive and breaks the installer's boot layout. The symptom is silent: the installer boots and appears to run, then sits frozen on "Installing base system" indefinitely with no error. When Rufus prompts you to choose, pick "DD Image mode" instead. (balenaEtcher and dd always write raw/DD-style by default — this only applies to Rufus.)

Getting Started Browse all docs

Beta builds are lab-mode

This ISO ships with default lab credentials and is intended for trusted lab / LAN environments only — not internet-facing production. Change the default admin password on first login, and restrict node management ports to your control-plane network. See the beta known-issues notes for details.

Hit a problem? Join our Discord or open a thread on the forum — beta feedback goes straight to the team.