"Wait a sec," Ove N. said, "your current network support guy charges a monthly retainer and hourly fees? You know, it's supposed to be one or the other; not both."
Ove's client-to-be looked a bit dismayed, "that's what we thought; but he said that this is just how it's done!"
It was a pretty easy sell for Ove; not only could he save the company $400/month in retainer fees, but his hourly support rate was $25 less than the current support guy. After signing the support contract, Ove offered to do a Network Assessment, just to make sure things were properly licensed, backups were functioning, and so on. The company declined, saying that they had been running smoothly for the past two years and only needed someone in case of an emergency. Maybe in a year or so they'd look to revamp the whole network. Ove was just fine with that.
Things were very quiet over the next few months, with Ove remoting in once or twice to add a new user. And then he received a panicked call: a week's worth of email and faxes had vanished, and week-old messages were suddenly showing up as unread. This called for an on-site visit.
When Ove arrived onsite, he was lead -- for the fist time -- over to the server room. And by "server room" I mean a vacant office with desktop and tower computers stacked on top of each other, cables strewn all over the place, and plugged-in monitors sitting in random places on the floor. Unventilated, it was at least 90°F in there. Ove questioned the setup, to which his client replied, "it did seem a little off; but the last network guy said that this is just how it's done."
Ove looked around the server room some more and noticed that there were six hard drives laying on a desk, each mounted in a hot-swap case with a weekday name written on the front. He asked what they were all about. "Those are our backup drives," the custom replied, "at the end of each day, we swap the current drive with the one that has tomorrow's day written on it; each Friday, we take one drive home and bring back another to the office. That way, we always have a full backup off-site, not more than a week old."
Being familiar with backup routines, this rotation scheme came as no surprise to Ove. The thing that worried him was that these were hard drives, not backup tapes. He logged on to the server and opened the RAID-1 administration console. A pop-up message with a red exclamation mark greeted him: "Warning: RAID-1 is degraded!"
He ducked under the desk and, sure enough, one of the disks had a red light above it. The bad disk was labeled "master." The other disk was labeled "Wednesday."
Then it dawned on him exactly what kind of backup scheme this was: one drive (the "master drive") was left permanently in the server and each day the secondary drive was pulled out and replaced with another. The RAID-1 controller would then see the array as failed and rebuild it using the "master" drive. A few hours later, the second drive would be rebuilt.
Now, it would seem that the "master" drive was not too happy about being abused like this, and after a couple years of constantly being used to rebuild the other drive, it seized up. Not noticing the red flashing light above the "master" drive, the customer removed the only functioning drive from the array and replaced it with a week-old drive. Hence the "week-old emails" problem. Ove explained the problem and how wrong this set-up was. Like a broken record, the client pleaded " but he said that this is just how it's done."
Fortunately, only half a day of emails and faxes were lost. Ove simply replaced the broken "master" drive with the previous days, and things were just like they the night before. It didn't take much for Ove to convince his client to not only get a tape backup system, but to go ahead and do that Network Assessment. And each step of the way, he was reminded that the set-up was "just how it's done."