Skip to content

UpdateSome plans are temporarily out of stock while we add memory. Running servers are not affected.

Passim IT
Menu

Why Passim IT exists, and what it is not

Most hosting company "about" pages are written to make a small company look like a big one. This one goes the other way, because the thing worth knowing about Passim IT is that it is small on purpose.

It started because good hosts are hard to find

I have always been an IT nerd. Long before any of this was a company I ran into the same problem everyone else runs into when they want to host something: finding a provider that actually listens. Not one that takes your money, files your ticket, and answers three days later with a copy-pasted paragraph that does not address what you asked.

That experience is the whole reason this exists. The people I host for have usually been through the same thing. What I am trying to be is the host a friend would be, if your friend happened to own the hardware.

Six years before it was a business

I started hosting Minecraft servers for friends and people I knew through Discord. That went on for close to six years, on my own, before any of it was official.

It grew from there. Websites. A simple VPS, meaning SSH root access and nothing in your way. A VPN for people who could not reach the sites they needed, or who wanted to host something without exposing their home connection to the internet. None of that was planned as a product line. Somebody asked whether I could do it, I could, and it became a thing I did.

Passim IT was registered in the Netherlands in June 2025, as an eenmanszaak. KVK 97583685. The registration changed the paperwork, not the way any of it works.

Why I own the hardware

I could rent a machine. I already use Hetzner for off-site backup storage, so I have no objection to renting in principle.

I own it because putting a system together is the part of IT I actually love. Picking the parts, building it, racking it, running the cables, configuring the network until the routing is exactly right. That is not a cost I tolerate to get a server. That is the reason I do this at all.

The practical consequence is worth stating plainly: everything runs on a Proxmox cluster I assembled and rack myself, Dell PowerEdge R640s with registered ECC memory, backed by seven Samsung SSDs in ZFS raidz2. It started as one Threadripper, which now does nothing but hold the storage pool. When something breaks, the person who fixes it is the person who built it, and there is no ticket queue between the two. There is a longer write-up of that migration, including the two things about it that got worse.

What has gone wrong

Every hosting page tells you what goes right, so here is one that did not.

My worst self-inflicted outages have all come from the same habit: making the scope too big. Segmenting every last thing on the network, routing traffic through a VPS because it seemed more secure and more reliable, and then discovering I had forgotten a route. The design was not wrong so much as over-ambitious for one person to hold in their head.

That class of mistake has never taken me down for more than a few hours. It has also never been anyone's fault but mine, and I would rather write that down than pretend the uptime graph has always been flat. If you are choosing a host, the useful question is not whether they have ever broken something. It is whether they will tell you when they do.

How big this actually is

Around five customers. About eight people hosting things, once you count the ones I host for who are not paying customers.

Every one of them arrived through networking or word of mouth. A hosting friend from Discord decided to stop hosting and their users needed somewhere to go. Someone told someone else. There has never been an advertising budget, and at this size that is not a gap in the plan, it is the plan.

I am telling you the number because most hosts will not, and because it is the context for everything else on this site. When the specifications page says your RAM is reserved and never oversold, that claim is easy to make at this scale and easy to verify. There is a live figure on the home page showing how much of the node is committed.

What this is not

  • It is not 24/7. One person answers, roughly 07:00 to 00:00 CET. Outside those hours nobody is reading your message. Every host that says "24/7 support" at this size is either lying or outsourcing to somebody who has never seen your server.
  • It is not a team. There is no first line, no escalation path and no account manager. There is me.
  • It is not the cheapest. Reserving memory 1:1 instead of overselling it caps how many customers fit on the node, and that shows up in the price.
  • It does not claim DDoS protection it does not have. Where a filtering service is not live, this site says so on the page where you would look for it.

If you want a company with a support portal, a sales team and a status badge that says 99.99%, there are plenty. If you want to know exactly what your server runs on and who picks up when it breaks, that part is all published.


Questions about any of this go to[email protected]. You reach the person who wrote it.