It’s big iron days at a military base, recalls pilot fish who was there, and a new, bigger and faster mainframe computer is installed in a somewhat temporary computer center.
Installation goes great and everything checks out OK.
Until a day later, when the new mainframe dies. Shortly after, it is brought up and it runs fine again.
Until the next day, when it dies once again. Engineers are on site and check things out. All diagnostics run with no errors; the system is good.
Until the next day … same thing.
Meanwhile, the old system sitting right next to the new system hasn’t even burped once. Fish is called in with other engineers to try to figure out what is going on.
After a lot of detailed evaluation, it emerges that the system only goes down in the afternoon within a two-to-three-hour time period.
That leads to a lot of head scratching and theorizing and coming up empty. No possible explanation can be ignored, and so the engineers take note when a member of the military staff makes the offhand observation that the mainframe is going down right around the time that helicopters return to the base every afternoon after daily flight training.
It’s true that the base radar is turned during that activity to sweep for incoming helicopters. But why would that affect one mainframe and not the other?
Turns out the old mainframe had what was called “ROS retry,” and the new mainframe did not. The radar caused ROS errors, which were retried by the old model and corrected before the next sweep of the radar. The new model couldn’t make those corrections.
The mainframes were moved out of that temporary computer center sooner than planned.
Your true tale of IT life is on Sharky’s radar. Send it to me at firstname.lastname@example.org. You can also subscribe to the Daily Shark Newsletter and read some great old tales in the Sharkives.
[ Related: Wireless charging from A to Z: What you need to know ]