One Platform for Bare Metal, Virtualization, Web Hosting and Domains

Most hosting companies do not sell one thing. They sell shared hosting to the small accounts, VPS to the developers, dedicated servers to the people who outgrew VPS, rack space to the ones who bring their own hardware, and domains to nearly everyone. Five product lines, five different fulfilment mechanics, one customer who expects a single invoice.

Almost nobody runs that on one system. The usual shape is a billing platform for invoices, a DCIM product for racks and IP space, a provisioning layer per hypervisor, and a spreadsheet holding the parts none of them agree about. FluxBilling exists to collapse that into one product. This post is about what that goal means concretely, how far along it actually is, and what it costs you to bet on it.

Read this as a statement of intent, not a feature list. FluxBilling is pre-1.0 and we publish what has not been built alongside what has. Every capability described below as shipping is in the product today and listed in the public plugin catalogue; everything described as not shipping genuinely does not exist yet, and no date is promised for it. If a comparison table anywhere claims otherwise on our behalf, this page is the correct version.

Part of the FluxBilling guides to web hosting billing automation software and DCIM software for hosting providers.

The problem is reconciliation, not features

The argument for one platform is not that four tools are four bills. It is that four tools hold four opinions about the same server, and reality only matches one of them.

A dedicated machine is decommissioned. The DCIM record is updated because the technician who pulled it works in DCIM. The billing system was never told, so the invoice keeps going out. Or the reverse: the customer cancels, billing stops, and the machine sits racked and powered for four months because nothing told anyone to pull it. An IP block is released in billing and stays allocated in IPAM, so it never gets reissued. A VPS is resized in the hypervisor and the subscription still bills the old plan.

None of those are bugs in any individual product. They are the predictable output of keeping the same fact in two databases and joining them with a sync job that someone wrote once. Every hosting operator past a certain size has a story like this, and the cost is not the software licence — it is the revenue that quietly stops, or the hardware that quietly costs.

The all-in-one goal is a response to that specific failure: put the invoice line, the rack unit, the IP allocation and the provisioning record in one schema, so that they cannot disagree, because there is only one of them.

What one platform actually has to cover

To bill a hosting business end to end, a single system needs all of these, not most of them:

• Web hosting — accounts created on a control panel, packages mapped to plans, suspension and termination that reach the panel.
• Virtualization — VPS creation, resize, rebuild, suspend and destroy against a hypervisor, with the subscription and the machine staying in step.
• Bare metal — physical inventory, rack and U-space, power, out-of-band access, OS installation, and an allocation step that picks a real machine when an order is paid.
• Colocation — rack units, power circuits and cross-connects sold as recurring products against hardware the customer owns.
• IP space — subnets and VLANs, allocation on provisioning and release on termination, with IPv6 treated as a first-class citizen rather than a text field.
• Domains — registration, renewal, nameservers, transfers, and the registrar operations that go with them.
• The commercial layer — recurring invoices, proration, tax, dunning, credit, refunds, multi-currency, resellers, and an export your accountant can actually open.

Any platform that covers six of those seven leaves you running a second system for the seventh, which puts you back in the reconciliation problem with extra steps.

Where FluxBilling is today

Shipping now, on every plan including the €4.95/month Lite tier:

• Web hosting — cPanel/WHM provisioning.
• Virtualization — Proxmox VE, Virtualizor and SolusVM 2. Game servers via Pterodactyl.
• Bare metal — hardware inventory with out-of-band credentials, SNMP, IPMI and Redfish discovery, rack and U-space management with chassis and blade support, power tracking, OS installation, and an allocation engine that matches a paid order to a physical machine.
• Colocation — rack units and power sold as billable products against the same inventory.
• IPAM — subnets, VLANs, automatic allocation on provisioning and release on termination, IPv4 and IPv6.
• Domains — eNom, with availability checks, registration, renewal, nameserver management, registrar lock and unlock, WHOIS privacy, EPP and auth-code retrieval, suspension and termination.
• Commercial — proration on upgrades and downgrades, dunning, credit and refunds, multi-currency, EU VAT handling, resellers and sub-accounts, ticketing, and accounting exports in XLSX, CSV, SAGA XML and bulk PDF.
• Payments — fifteen gateway integrations, from cards and direct debit to bank transfer and cryptocurrency.
• Integration — a visual plugin builder, so a provider with an HTTP API can be wired up without writing a module.

The part that matters for the thesis: those are not separate products with connectors. Rack, IP, inventory and invoice are the same database. When a service terminates, the IP returns to the pool and the rack unit frees because they are rows related to the service, not records in another system waiting for a sync.

Where it is not, stated plainly

The goal is all-in-one. The current state is not, and pretending otherwise would make everything else on this page worth less.

• SSL certificate reselling does not ship. There is no bundled certificate authority integration and no certificate product type. You can bill a certificate as a product and wire ordering to a CA through the plugin builder, but that is work you do.
• Software and control-panel licence reselling does not ship. If you resell panel licences as a line item, there is no dedicated licence product type or vendor integration for it today. This is on the roadmap and it is genuinely not built.
• Several major control planes have no module. Plesk, DirectAdmin, OnApp, CloudStack, OpenStack and VMware are not in the catalogue. If you provision from any of those, an established platform serves you today and we do not.
• One registrar, not many. Domains means eNom. If you are with a different registrar, that is a plugin-builder project.
• The catalogue is small. Six infrastructure integrations against ecosystems with hundreds. We covered the honest version of that in control panel and hypervisor support compared.
• Pre-1.0. The roadmap lists what was deferred, and the changelog includes the rollbacks.

What all-in-one costs you

Consolidation is a trade, not a free win, and the case against it is real enough to state.

You give up best-of-breed depth. A specialist DCIM product built by people who do nothing else will have features an integrated module does not. If hardware visibility is the centre of your operation rather than an input to billing, the specialist is the right buy and we said so in EasyDCIM vs HostBill.

You concentrate vendor risk. Four vendors means four things that can fail independently. One vendor means one, and it holds your invoices and your infrastructure records together. That is a legitimate reason to want an export path and a licence you control, which is why self-hosting exists on the Business tier and why the accounting exports are plain formats rather than something proprietary.

You inherit our gaps. With four tools you can replace the weak one. With one, a missing integration is a missing integration. That is the honest cost of the list in the previous section.

Why we think the trade is worth making

Because the reconciliation failures at the top of this page are not edge cases, and they do not get better as you grow — they get more expensive. A hosting business that cannot answer "which physical machine is this invoice line for, and is it still racked" is losing money in a direction nobody is measuring. Every integration we have not built yet is a real gap. But the gap between an invoice and a rack unit is a worse one, and it is the one that most stacks in this category leave open by design.

That is the goal: not more features than the incumbents, which would take years and would not be true. One system where the commercial record and the physical record cannot drift apart, sold at a price that does not require a second product to be useful.

How the plans work

Lite is €4.95/month for 10 clients and 5 servers, Starter €24.95/month for 250 clients and 50 servers, Professional €34.95/month for 500 clients and 100 servers, and Business €44.95/month for unlimited clients and 150 servers. Additional servers are €0.75 each per month. White-label branding is a €50/month add-on on any tier. Self-hosting is available on the Business tier as a €500 one-time fee on top of the subscription. DCIM, IPAM, provisioning, every plugin and every payment gateway are in the base plan on all four tiers — there is no infrastructure edition to upgrade into. There is no free trial; every plan carries a 14-day money-back guarantee.

See DCIM, IPAM, provisioning, the plugin catalogue, the roadmap and current pricing.

Closing thoughts

If you run one product line and one control panel, you do not have the problem this platform is built to solve, and a mature specialist with a decade of modules is probably the better buy. If you sell across bare metal, virtualization, web hosting and colocation at once, and you have ever had an invoice disagree with a rack, this is the argument — and the list of what is missing is on this page rather than buried, so you can price the bet properly.

All third-party product names above are trademarks of their respective owners and are referenced for identification only. FluxBilling is not affiliated with, endorsed by, or reviewed by any of them.