The Year In Downtime

1sockchuck writes "Lightning, floods, car crashes, and coding snafus had starring roles in major Internet outages of 2010. Data Center Knowledge reviews the year's business downtime, including outages for banking and e-commerce sites and several incidents that knocked state government services offline. Meanwhile, Pingdom focuses on downtime for major social media sites and Wikileaks. Then there's the guy who got drunk and shot up a server."

Comments

Lack of funding

[DigiShaman]

And then you have the cheap bastards and/or business without money. Time and materials are often needed to perform preventive maintenance. So, I'd say were seeing a lot more reactive vs proactive support as a result.

Re:Lack of funding

[DigiShaman]

If you think firmware on modern devices is perfect, you're sorely mistaken. Just as with any other software, firmware on shipped servers can be buggy and lead to data loss. That can range from a nuisance to extended downtime. Things like BIOS, back-plane, RAID controllers, and HDDs may require updates to prevent data corruption or total loss. These are all things can be scheduled ahead of time in a controlled manor rather than be blindsided with issues during normal business operation.

And that's just strictly hardware. OS, backup job review, and patching are all part of the preventive maintenance routine. If you wait till it's too late, you could be kicking yourself later and wishing you HAD spent the money over time rather than one lump sum in man hours with extended down time.