From Contractors to Cleaners, There’s No Shortage of Threats to Your Server Rooms and Wiring Closets

This audio was created using Microsoft Azure Speech Services

An hour delay in the discovery of a cooling failure can make the difference in avoiding a complete outage, yet far too many companies don’t have what it takes to detect such a failure in their unmanned server rooms and wiring closets: a monitoring and automation system for their data center infrastructure.

We assembled a list of horror stories to demonstrate just how easy it is for something to go terribly wrong in a server room or wiring closet, without you knowing it.  In a related post, we tell you how to avoid disasters like these.

1. Without informing the data center operator, contractors hired to renovate a server room wrap servers in shrink-wrap to keep dust away from them.  The servers overheat and shut themselves down.
2. A senior business manager is having trouble connecting to the Internet so takes matters into his own hands. He goes into the server room and connects his laptop directly to the router, thus bypassing all firewall services and encryption and exposing the entire system to outside viruses and other malware.
3. A plumber drills a hole in the ceiling directly above an Exchange server as he’s repairing a pipe. His fix doesn’t hold and in the middle of the night, water begins leaking from the pipe, through the hole in the ceiling and onto the Exchange server below, causing permanent damage to the server.
4. Cleaning people sent into the server room see pockets of dust around the server racks and inside them as well.  The do what comes naturally to them: clean inside the racks and inside of the servers using window cleaner.
5. A UPS overheats because packages of toilet paper were piled high on top of the unit, obstructing air circulation.
6. Employees at a retailer routinely plug devices such as space heaters and fans into circuits protected by a UPS that supports point of sale systems, causing an overload that kept bringing the POS systems down.
7. A rack of servers is lost because an IT administrator unintentionally overloads an already maxed out power strip.

If any of these stories sound a little too plausible or close to home, and they should, you should investigate a data center monitoring system.  Check out our related post to learn more about the key components of such a system, including:  video surveillance, sensors, intelligent rack outlets, and monitoring and automation software.

Or, learn more by downloading the APC by Schneider Electric white paper, “How Monitoring Systems Reduce Human Error in Distributed Server Rooms and Remote Wiring Closets.”  You’ll be glad you did.