Sometimes You can’t Prevent Failure

This is one of those blogs that you won’t often hear from the likes of me. The topic of this blog is the difference between what you can prevent and what you can’t. I think many people like to believe their Web Hosts are all powerful, and we get disappointed in them when we find out that they fail just like everyone else… but unlike everyone else we pull our sites from them and go somewhere else.

Hardware fails. People can have bad days. And then there is always those problems that are completely out of your hands… a semi slamming into power lines, wire thieves, random gunfire, worse snowstorm of the century, or even the ultra-curious, but not too bright, racoon.

Peter over at Pingdom (for the website, www.pingdom.com, and their blog, royal.pingdom.com) sent me an interesting blog on the number of freak accidents that happened in one week and caused downtime or power failure that could lead to downtime.

Some of these are absolutely funny. You can read it for yourself here: http://royal.pingdom.com/?p=231 .

As an example of my own brief searches. A bird struck a transformer in El Paso and tripped a circuit breaker killing power for roughly 3,000 customers.

What was the most surprising (other than the apparent large number of suicide fauna) is that many back up systems did not work as intended. Even companies who have extensive maintenance routines had problems getting their backup systems up and running.

Going back to what was said earlier, I am no different. When one of my sites dies I start looking for a new host. Doesn’t matter if it was entirely out of their hands, which is a sad commentary really. We want 100% uptime, but we want it for $4.95. Not only is it unrealistic in the sense of money value, but we hardly put such pressure for performance on ourselves. If I were to look at the number of typos I have on a first run draft, I’m sure I would probably be at 90%.

That’s not bad, but it is a long way from 100% isn’t it?