Unintended consequences

I’m not sure how far back to go, to get to the beginning.

Could be the time, a few years back, when the townhouse complex’s main water supply, after 30 years of flawless operation, was “upgraded.”  This, of course, inevitably resulted, a couple of years later, in some very odd variations in water pressure.  Some of the time we had little more than a trickle of water in the taps, and occasionally the washing machine took forever to fill.  (The “upgrade” may also have been responsible for the Great Flood of Aught-Nine, out on the main road.  But I digress.)

This year the main pressure regulator for the complex was replaced, and water was back to full pressure.  As a matter of fact, it was back to significantly higher than full pressure.  Filling the washer (or sink) is much quicker than it used to be.  You have to be careful not to turn the kitchen sink on full blast, or much of the counter around it gets sprayed.

A couple of day ago, the upstairs toilet stopped working.  Well, it would still flush, if the tank was full, but refilling slowed to a stream of drips.  (Hypothesis: the intake valve in the tank has blown from the higher water pressure.)  The manager happens to be away this weekend (of course), so we’ve been muddling through.

This morning, while attempting to refill the tank manually, I discovered that, if the tank was in the process of filling itself, and you turned on the bath tap full blast, the toilet would start filling normally.  Further experimentation determined that it had to be full blast: half or even three quarters wasn’t good enough.  (Revised hypothesis: the valve is partly damaged, and reducing the pressure allows it to function, temporarily.)


Anyway, it reminded me: if a system as simple as a toilet, and household plumbing, can have these sorts of effects, what makes you think your incredibly complicated IT system, and its protective elements, is working as you think it should be?