Nocix, 1530 Swift, Wholesale Internet, and formerly DataShack are all the same company.
A Kansas City, Missouri (yes, strange) budget-dedicated server and colocation provider.
If you review their terms of service and go over to the Kansas business entity search, you can see they were formed way back in 2003 and opened up for business as Wholesale Internet, INC.
- Nocix, LLC (formerly DataShack)
- Wholesale Internet, INC
- 1530 Swift, LLC
All share the same business address of 201 E. 16th, North Kansas City, MO 64116.
DataShack and Wholesale Internet were more of their previous brand names:
“Datashack / Nocix / Wholesale Internet….. Have they fixed their frequent power outage issues yet?”
One of the biggest issues that’s haunted budget-friendly Nocix and crew since the beginning is frequent power issues.
You know, the power randomly going out (for whatever reason) and taking down your server, because there wasn’t much redundancy… and then maybe starting after the power comes back, hopefully with no disk errors. That kind of thing.
Here’s what a thread way back in 2016 said:
When comments pointed out Nocix was known for being a budget host, the author of the post replied:
I knew I would get replies like this….. so basically, I’m a big idiot for expecting Datashack/Nocix to provide reliable service because they’re a “budget host,” and budget hosts all suck and “you get what you pay for” and all of that. Right?
Well as I said in my original post, they’re service has been good. And I’ve been satisfied overall. For a “budget host” I think they’ve performed adequately over the last 2 years.
But WHY shouldn’t I expect a company – even a “budget” company – to fix their rudimentary power supply issues? They’ve had these power issues for a long time. I don’t think it’s unreasonable (in 2016) to expect even a budget host to supply constant, consistent power to my boxes. It’s not rocket science.
That’s why I asked the question in the OP.
I know they will fix it eventually. I’m just curious if anyone has any inside info about their progress.
Semi-related note… Over the last 2.5 years I’ve been paying $120/month for a server that they sell today for $50. They’ve made good money on me (and thousands of other people like me). So yes, they’re a “budget” provider, but they still have serious cashflow running through that place. Don’t tell me that they won’t sometime/someday fix their power issues once and for all.
He wasn’t alone, either.
There are lots of other threads complaining about DataShack and Wholesale Internet power outages in the mid-2010s (or even before).
Eventually, it seems Wholesale Internet, INC realized the DataShack brand had grown an increasingly negative perception and dropped it at that point.
The Wholesale Internet site also appears to be seriously outdated. The last update to its terms of service was back in late 2016, right around the same time DataShack died and Nocix was born.
Definitely looks like it hasn’t been updated since the early 2000s.
The Birth of Nocix
After the death of DataShack, Nocix is launched. A more modern take:
They ran pretty good deals, honestly.
Dated hardware, low-quality bandwidth blends, and not the best of uptimes, but nonetheless… a solid deal depending on the use case.
They also introduced their new colocation brand…
1530 Swift: “The Most Connected Datacenter In North America”
Big moves. For a while, it seemed they were quietly working on growing and improving their infrastructure.
That Was, of Course, Until the Power Went Out Again
Back on July 5th, 2023, @windytime90 posted a thread over on LowEndTalk.
With the following Tweet being shared from the primary energy provider in the KC area:
‼️ OUTAGE UPDATE ‼️
Strong storms and high winds came through our service area last night causing more than 55,000 customers to lose power. Customers in the Kansas City area were most impacted. We’ve restored power for more than 80% of customers already. pic.twitter.com/QsI6g8jxuF
— Evergy (@evergypower) July 5, 2023
Quite a few budget VPS and hosting providers in general felt the heat. Nocix provides cheap services, and that comes with… quirks.
Some users were down for days. On its website, Nocix advertises the following:
Our facility’s power is fed with dual feeds from dual power grids on separate substations. Even buildings in downtown KC fed from multiple power grids are only fed from a single substation.
Temporary power is supplied by modular, online UPS systems. Our UPS systems are all N+1 redundant. The modular nature of the systems allows us to perform maintenance and correct UPS issues without the need to cut power to the servers. We have enough capacity to assure continuity of power to the entire facility. In addition, a separate UPS system maintains all critical networking infrastructure.
Backup power comes from an onsite generator farm. We operate both natural gas and diesel powered generators. In the case of diesel the generators can run for 48 hours without refueling. We maintain fuel contracts with multiple local companies to supply fuel. In the case of catastrophic event, we have access to over 20,000 gallons of fuel stored in close proximity to the DC.
If you have generators capable of running for over 48 hours, how does a bad storm knock your customers offline for days?
Another thread was posted onto WebHostingTalk right around the same time where Aaron, one of the founders of WSI, commented the following:
First off. I apologize for the recent outage. It was a set of circumstances that should never happen.
While we’re still working on figuring out all the details the short story is that power to the facility was knocked out by storm damage in the area. When this happened, generator one, which had just been tested the day before, started, then threw an overcrank error and shut down. Generator two started and ran as expected for a short time before we believe it was struck by lightning, damaging the alternator. This caused the phase to ground voltages to crazy, throwing alarms on the UPS’ which then dropped load causing a complete outage to our entire facility.
This is the first complete outage we’ve experienced in our “new” facility (it’s over 10 years old now). The last one I can think of was in 2008 or maybe 2010 when we were still downtown. That’s not to say its the first time we’ve lost power. We lose power every once in awhile, like anywhere, and the generators run, UPS’s work and no one ever knows.
So the issue becomes, when you have over 15,000 servers running, sometimes for years, along with PDUs, switches, etc, sometimes running for over 10 years, that suddenly get shut off they don’t necessarily come back when the power comes back. Power supplies fail, drives fail, PDUs die, all sorts of bad things happen and it’s not as easy as just getting the lights back on.
All the downed switches and PDUs were replaced by mid day yesterday. We’re down to just the individual machines that have hardware issues that need to be fixed. When I say “just” I mean we’re down to under 500 machines still offline.
I understand that getting responses to tickets that seem canned or short can be frustrating but every one wants a response and if we’re spending all our time writing individual responses then we’re not getting things turned back on. In addition, the technician answering your ticket may not have all the facts and we don’t want them assuming things and giving out incorrect information. That leads to more confusion and just fosters distrust in ALL the information being disseminated.
As for excuses, I don’t have any. What happened happened and we will look into way to improve as we always do. If you are still down I would ask that you work with our people in the original ticket you submitted. Submitting multiple new tickets just slows down the whole system for everyone and will actually increase your resolution time, not shorten it. Also, if you are not a direct customer please do not submit a ticket to our system. Work through your reseller. We don’t have any information on you and therefor can’t answer your questions.
Again, all overtime for our technical staff is approved and they are all here working as fast as they can. We expect to be down to a handful of servers by the end of the day and back to normal for the weekend.
The “new” facility he’s mentioning I would assume is 1530 Swift:
Here’s what the building actually looks like (an old warehouse maybe):
They also have “Nocix Clay” which looks like an office building and is right across the street:
This is Clay right here:
(1530 Swift is the building seen on the left.)
As someone that’s very familiar with Kansas City, I can tell you that this area is pretty dated. It’s an older area of town, mostly businesses, warehouses – that kind of thing.
In Kansas nine out of ten times when you see streets with parking like the following:
You’re looking at a dated area of town, with likely dated infrastructure, including electricity.
If you go to the Kansas City power and light district, for instance, you can instantly tell the difference in the age of the area and the infrastructure.
The founder’s reply to his reoccurring power issue was quite simple, “as for excuses, I don’t have any. What happened happened and we will look into a way to improve as we always do.”
It’s cheaper to repurpose an old large building into a data center than build a new data center, so I get it. Especially for a budget brand.
Does It Really Matter Nocix Has a History of Going Down?
For the majority of people, probably not. Nocix is a budget provider.
Hopefully, people going into Nocix have done their research and know what they’re getting into.
1530 Swift literally sells a full 42U colocation cabinet for $400 per month. There are not many competing at that price.
Depending on the use case and if reliability is a big concern for you or not, Nocix or 1530 Swift could still totally work for you.
A little downtime here and there to save hundreds is something to consider, depending on the project.
And hey, maybe Nocix fixes their power outage issues once and for all, and we never see any power outages again.
Who knows?
Related Posts:
- Path.net Seeks Discord Subpoena for Ex-Employee and Competitor’s Data - December 12, 2023
- Interviewing Christopher Cantwell - December 6, 2023
- Uptime Kuma: Your New Favorite Self-Hosted Uptime Monitoring Tool - December 1, 2023
Leave a Reply