USE-NY Server Migrations

A Huge Start to the Year

This year has already been massive for us at Kinetic. We hit our end-of-year client goal at the start of February—something we never expected so early! It’s been incredible to see this growth, and we owe a huge thank you to everyone who has checked out Kinetic Hosting and placed an order.

With this growth, we’ve had to more than double our server inventory in a short time. Like most hosting companies, we work with multiple data center partners worldwide, renting hardware and installing our own. We’ve relied on our core partners for years, knowing they provide a stable and reliable service.

Scaling Challenges in the USA

In the USA, we have one key partner. Over the past few months, they’ve struggled to get Ryzen stock, meaning we had to pre-order servers months in advance. That worked fine—until January, when we experienced an unexpected surge in orders. Several of our modpack partners, like Beyond Depth, took off, leading to a sudden spike in demand.

Our usual data center partner was already facing stock shortages. While we managed to get some servers from them, they couldn’t keep up with our needs. This forced us to look for additional data centers to partner with. While our Performance and Budget systems in the USA continued to be supplied by our main partner, our Impact systems were placed with a new provider. Initially, we had no issues, so when we suddenly needed more servers, we worked with this provider to get new Performance systems in stock.

Reliability Issues

Although we had used this partner for months without problems, as we expanded our inventory, we began encountering serious reliability issues. Due to the rapid growth, we relied on rented hardware. We soon found that for every three servers we ordered, at least one had a hardware fault. While most issues were caught in setup testing, some slipped through and affected live servers. This resulted in unexpected outages and forced us to move clients off faulty hardware.

Hardware failures aren’t uncommon—that’s why we run checks—but this level of issues was something we had never seen before with any other partner. In hindsight, it was a warning sign of things to come.

Power Issues

At the start of this month, six of our servers with the new provider randomly went offline. After reaching out to the data center and waiting four hours, we were told it was a PDU failure. A PDU (Power Distribution Unit) is essentially a power strip for servers.

The next day, at around the same time, the same servers went offline again. Once again, after another four-hour wait, we were told it was a PDU failure. And then, the day after, it happened again. The response? Another power issue, now supposedly resolved.

Hardware failures happen—but three days in a row, with no proactive communication from the data center, led to unacceptable downtime for you guys. We pride ourselves on high uptime, and we understand that a server being down for four hours can mean losing an entire evening of gameplay.

This wasn’t the only power issue. Several times, systems would randomly go offline, only for us to later be told that someone had accidentally unseated the power supply—essentially, they had unplugged the server by mistake.

Network Issues

Beyond hardware and power failures, we also experienced frequent network outages and unexplained server restarts. These disruptions further impacted our ability to provide a reliable service.

Migrations

After weeks of work, we’ve finally secured additional stock with our primary USA provider. As new servers come online, we’ll be migrating affected clients to more reliable hardware. Keep an eye on our status page for updates on when your node will be moved.

Each migration will involve minimal downtime—no more than 30 minutes, with most servers moving in under 5. The only change you’ll notice is a new IP address.

Our Apology

While these issues only affected a small portion of our clients—and we worked quickly to move users off faulty hardware—this still caused disruptions for some of you, and for that, we sincerely apologize.

We did our due diligence when selecting this provider, as we had used them before without issues. Unfortunately, they let us down this time. We’re committed to ensuring this doesn’t happen again and appreciate your patience as we work to get everything back on track.

Thank you for your continued trust in Kinetic Hosting!