Night-shift horror a screamer

Virtually unsupervised, the “new guy” got a thorough initiation when faced with a night-shift full of spun-down drives

Because of the way our machine room was configured, the disk farm was located on the floor directly below the main computer. The unmounted disk packs were stored in the same area on shelves along the walls

When a user wanted data that was not online, he would go downstairs, spin down an appropriate disk drive, remove the 40-lb. disk pack, mount the new disk pack, spin the drive up and tell the OS it was now available. The machine-room staff performed this procedure numerous times a day, all three shifts (we ran 24/7).

John, the new staffer in the machine room, had been given the worst shifts — midnight to 8:00 a.m. Unfortunately, this shift was also the least supervised, but that did not bother John — he was young, enthusiastic and eager to learn.

One early morning about 3:00 a.m., one of the disk drives went offline. When the alarm sounded at the console, John was dispatched to investigate. Upon arriving at the offending drive he found it was indeed spun down. First, he spun the drive up. It came up, displayed some status lights, then spun down again. John was puzzled. He didn’t quite understand the meaning of all those status lights yet, but he did know that there was really not that much to the things — just a drive and a disk pack. His supervisor was not available, but John figured he had a way to fix this problem.

It must be a faulty disk pack, he decided. No problem. He knew where the test packs were, so he carted one out and mounted it. Same thing. He knew that test packs didn’t always perform correctly, so he decided to try another. Still no success.

Now John was no dummy. If it wasn’t the disk pack, he reasoned, it had to be the drive. He proceeded to take one of the test disk packs he had tried over to one of the spare drives, mount it, and spin it up. Down it came. Really strange. So he got another disk pack. Still no change, whereupon he tried another idle drive. Now totally confused but not about to quit, John then swapped disk packs and drives more or less at random. By the time his supervisor showed up to see what the problem was, John had “discovered” that there were some five drives and at least nine disk packs that just would not spin up.

By now the old-timers reading this are probably busting a gut because they know what had happened. The problem was that the original drive had shed a drive head, damaged the original disk pack, and spun down. That’s what the status lights (which John could not read) said. When John put the now-damaged original disk pack in a new drive, it immediately destroyed the heads on the new drive and the drive spun down. And so on, and so on. Result: the loss of many tens of thousands of 1960s dollars’ worth of hardware.

But this tale does have a happier ending. When all this was reported to the operations manager the next morning by a shame-faced John and his equally embarrassed supervisor, instead of exploding, the manager took this as a chance to change procedures for both operations and training. John was not fired. As a matter of fact, many years later he was promoted to the position of operations manager.

Throughout his career John never forgot the lesson he had learned and often told the story himself when instructing new staff.