Trade Desk
Node apps you have to keep running for other people
A Node site is not a website with different syntax — it is a process that can die at four in the morning, and the client's first call comes to you.
Straight answer first
Use the Node support on our Business hosting for a standard client app, and move to a VPS running PM2 behind nginx when you want the whole process tree answering to you — the second option is the one you can put an uptime promise against.
The commercial point underneath: a PHP site fails by returning an error, and a Node app fails by not being there at all. Whoever owns the restart owns the relationship, so decide whether that is the platform, you, or the client's own developer, and make sure the hosting you buy matches the answer you have already sold.
Written by the Hosting Seller staff · Checked 4 August 2026
NVMe
Storage under every account
Free
Domain, year one on annual orders
99.9%
Uptime target behind your promise
Flat
Price at renewal, per account
For studios and resellers who have taken on a JavaScript app — an API, a bot, a server-rendered front end — and are now responsible for it in a way nobody quite wrote down. The build was scoped. The keeping-alive rarely is.
You learn how good Node hosting is at the moment something crashes. Managed process support brings it back on its own; a runtime bolted onto a PHP-shaped platform means you hear about the outage from your client, which is the most expensive monitoring system ever devised.
What the customer sees when it stops
A dead PHP site shows an error page. A dead Node app shows a browser timeout, or worse, whatever the layer in front of it is configured to show — which on carelessly arranged hosting is a default page carrying somebody else's name. That is your white-label leaking at precisely the moment the client is paying attention.
Fix it before it happens: a branded holding page in front of the app, a health check that restarts rather than merely reports, and a status message you can send within minutes. None of that is expensive. All of it is invisible until the one day it is the only thing anybody remembers.
Shared account or a server of your own
A standard Express or Next app on the Node support in our Business plans keeps the convenience of a managed account and a panel you can hand to a colleague. That is the right home for most client work, and it costs one fixed monthly figure billed annually rather than anything that moves with the traffic.
Custom daemons, websockets at volume, or a runtime version the client insists on all argue for a VPS with root, where PM2, the firewall and the reverse proxy are yours. Charge for it. A managed server is a service you are performing, not a hosting line you are passing through.
Pricing an app that needs watching
Hosting is a number; attention is a rate. Split the invoice so the client can see both, because a single blended figure teaches them that the two-in-the-morning restart was included and always will be.
A workable shape: the platform at cost plus your margin, a fixed monthly allowance of maintenance time, and an out-of-hours rate that exists mostly so nobody uses it casually. The clients who object to seeing it written down are precisely the ones who would have consumed it.
What you should be able to do without raising a ticket
Read the log, restart the process, roll back to the previous release, change an environment variable. If any of those requires a ticket with your supplier, you cannot honestly offer a client a response time, because your response time is really somebody else's queue.
Over SSH those four jobs are minutes of work. That is the commercial reason shell access matters rather than merely the technical one: it is the difference between a support promise you control and one you are relaying.

The supplier writing this would like the account
This is a host recommending hosting, said out loud. There is no affiliate arrangement behind the recommendation and no fee for the position it takes, which is why the rest of the page can afford to be blunt about where the limits sit.
Mailboxes on the client's own domain are part of the plan rather than an extra you have to itemise and defend.
- Tests you can run before you commit a client
- One trade price, held at renewal
- Free migration for the sites you already host
- A refund policy with no small print
Why Hosting Seller
On every plan, as standard
Node without a PHP-shaped platform
Managed process support on Business plans, rather than a runtime dropped on the box and left to fend for itself.
Shell access as a support tool
Logs, restarts, environment variables and rollbacks are yours to run, which is what makes a quoted response time honest.
Matched to this verdict
The Business Standard plan is this page's recommendation made concrete — one flat price, essentials included, upgradeable in place.
Free client migrations
Bring an app you already host elsewhere and our staff move it across at no charge, usually inside 24 hours, with nothing going dark.
Root when the app earns it
KVM virtual servers with full root, DDoS filtering in front and one flat price, bought from the same counter as everything else.
A supplier with a company number
IGI Security Services Ltd, registered in England and Wales — a public record, which is what you would want your own supplier to have.
First Steps
From choosing to live
- 1
Name who restarts it
Before anything else, write down whose job it is to bring the process back and within how long. Every other decision on this page follows from that one sentence.
- 2
Put your own page in front of a failure
A branded holding page and a health check that restarts rather than merely alerts stop an outage from doubling as an introduction to your supplier.
- 3
Rehearse the rollback
Deploy, break it deliberately on staging, roll back, and time yourself. Do it inside the refund window on a real account so the number you quote is one you have actually measured.
In the Box
Packed with every plan
- SSH on the plans built for developers, for logs and restarts
- Git deploys, so a rollback is a command rather than an upload
- cPanel, the panel any developer you hire already knows
- Node.js, Python and Ruby runtimes on Business plans
- Staging copies so a release is proved before a client meets it
- Upgrades applied to the account in place, with no move between plans
- Free site migration handled by our staff
- Mailboxes on the client's own domain, included
- Root access on KVM virtual servers when an app needs it
- Real people on the counter, every hour of every day
Across the Counter
Things people ask us all the time
Can I promise a client an uptime figure on a Node app?
Only for the part you control. The platform carries a 99.9% target with a pro-rated credit on any month that falls short through a fault at our end. The application process is a separate question, and to promise anything about it you need either managed process support or a VPS where the supervisor is yours.
How do I keep my supplier out of sight when the app is down?
Put your own page in front of it. A default error page from whatever sits ahead of an application is the commonest way a white-label arrangement gives itself away, and it costs nothing to replace with a holding page carrying your name and a support address.
Is it worth putting several client apps on one account?
For quiet apps, yes, and the allowance supports it. The caution is blast radius: several clients on one account means one runaway process is four unhappy phone calls. Busy or business-critical apps deserve their own resources, and that is a straightforward thing to charge for.
What does your support actually cover for a reseller?
People at the counter at every hour, with a scope that takes in the practical work: the server, the network, the panel, mailbox and DNS questions, restores. It stops at your client's application code, which is where your own service begins — ask us something hard before you order and the reply is a fair sample of what follows.
Read next
Website Builder vs WordPress
Two ways to give a client a site, compared on what each one leaves you able to support afterwards.
News Site Hosting
Publishing workloads and sudden traffic, weighed up by somebody who has to keep them online for a customer.
Node.js Hosting
Keep Node.js apps beside your websites, with SSH and Git access.
VPS Hosting
KVM virtual servers with root access, DDoS filtering and one flat price.
Moving your site to another host? Start with this checklist.
A step-by-step order of work for a move your visitors never spot: which files to copy first, how to carry email across without losing one message, the right moment to repoint DNS, and the two mistakes behind nearly every hour of downtime people ring us about.
Take on the app with your eyes open.
Shell access, staging, free migrations and root available the day a client outgrows shared. One price, held at renewal.
See Node.js Hosting plans