Not the kind of redundancy we really needed here


It's a few decades back, and this technical rep pilot fish -- who works for a big telecom switch company -- is awakened from a sound sleep by his beeper going off.

"Upon calling my office, I learned that the entire campus system for one of my accounts was down and was told to get over there immediately," says fish.

"I frantically drove the 15 miles across town to the college campus, to find everyone in the computer room in a panic. But there was no fire, no flood, and everything looked normal. Another tech had been called in and he couldn't find anything wrong either."

In fact, it all looks pretty much like it did the day before, when fish completed a major configuration update to the main switch.

The big difference: Everything was still working fine when fish finished up by popping the oversize floppy disk out of the drive for one of the two redundant CPUs, writing the date on its jacket with a marker, then reinserting it in the drive with a satisfying click and going home.

But after an hour, fish is baffled, the entire two-campus complex is still without voice or data communications, and the campus telecom manager is losing patience with both fish and his employer.

Still, fish keeps looking. "At one spot I could faintly hear this repeated clicking noise," he says. "I called my colleague over, and when we got nearer to the CPU's drive unit we could hear it better.

"I popped out the redundant CPU's diskette that I had placed in the afternoon before, clicked it back in, and all of a sudden everything came to life."

The other tech looks at fish in disbelief. So does the telecom manager, who's happy the system is working, but wants to know what went wrong.

Fish knows the workload automatically switches between CPUs every 12 hours, to prevent a physical meltdown from the heat generated by all the discrete components in the switch's 1970s-era computers.

He tells the telecom manager his best guess: Somehow the disk wasn't seated correctly in the non-active CPU's drive. And when the load switched over, sometime past midnight, the CPU couldn't read anything from the floppy -- a critical part of startup -- and shut itself down.

"My colleague said he had never seen that before, and ventured that the odds would be about one in 10,000 that a diskette once it's clicked into the slot would not be seated properly," says fish.

"He went to the inactive CPU side of the big phone switch and popped the other diskette in and out about a dozen times, to prove the point to the now-much-calmer telecom manager."

Communicate with Sharky. Send me your true tale of IT life at You'll snag a snazzy Shark shirt if I use it. Add your comments below, and read some great old tales in the Sharkives.

Get your daily dose of out-takes from the IT Theater of the Absurd delivered directly to your Inbox. Subscribe now to the Daily Shark Newsletter.

Why is Apple letting Macs rot on the tree?
View Comments
Join the discussion
Be the first to comment on this article. Our Commenting Policies