When good servers go bad

Posted on December 1st, 2008 by fizzle in Rants

Servers are peculiar objects. They can sit there and do millions of calculations and operations per second, for months and months and months on end.  They will dutifully perform their tasks until they are interupted by some stupid user asking it to do something it doesn’t want to do. Until the power goes out due to a few inches of freshly fallen snow. The power outage should be no big deal, right? Datacenters are the most equipped buildings known to man that are able to deal with mother nature and her wicked ways, right? Sure. Datacenters have backup generators to keep all of the machines and power on and running flawlessly while the electricians work to restore power. What happens when a water main bursts and floods the room where the backup generator is and main circuitboards, causing the entire building to lose power?

Well, if you’re the webhosting company sectorlink, located in Flint, MI, you’re entirely shit out of luck. What if you’re a customer of sectorlink’s? You’re definitely shit out of luck, and losing money by the boatload, while fielding angry calls from your clients, supervisors, and what have you.

This was my day. I got in at my usual 15 minute late marker of 9:15am, worked and got things settled after the long holiday weekend, and then I had some things to take care of out of the office around 10:15 or so. I come back to the office around 11 and my project manager tells me “Sectorlink is down”. Being that they are who they are, it’s no suprise. They go down once every month or so for anywhere from 30 seconds to 2 hours. That’s why it’s our development server, because its simply unreliable. Well, now it’s become much more than unreliable, it’s become a liability.

We are no longer going to be using their dedicated server services, or any other services for that matter. I understand that things happen outside of the control of the hosting company. I understand that they probably worked their asses off to get it restored at around 3:30pm or so, for a total downtime of nearly 5 hours. That’s all well and good. It was just the straw that broke the camels face. Oh, and the fact that their “datacenter” looks like nothing more than a 1,300 square foot ready to least office space. Located in an alley. Literally. Their address is 512 Buckham Alley.

We had always joked that they were running their entire datacenter on a Comcast cat5 line, but it seems pretty damn likely that they just might be. At any rate, we’re just another dissatisfied customer with another dissatisfying local service. Today we purchased a nice new fancy shiny dedicated server from 1and1, which has twice the storage, bandwidth, ram, processor power (quadruple CPU power, actually), for half the price of sucktorlink. Yes, I am the first person to ever call them sucktorlink. I’m so damned witty and original

TL;DR: our server went down because of a power outage and then a burst water main took out their generator. Owned we were, and owned they are now that they’re losing a ton of customers because of it.

Tags: datacenters, hosting companies, power outages, servers, water main breaks

This entry was posted on Monday, December 1st, 2008 at 8:56 pm and is filed under Rants. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.