DIMM Techies Werent Allowed To Leave The Building Until Proven To Not Be Pilferers

On Call Another Friday is upon us, and The Register understands some of you would rather not retain memories of the last week. That's why we offer another instalment of On Call, our reader-contributed column that revives your happier recollections of wreaking revenge on colleagues who caused you tech support trauma.

This week, meet a reader we'll Regomize as "Harry" who once worked for a supergiant global tech services company as it helped a major UK government agency to drag its tech into the 21st century.

"I was one of those responsible for digital security in two large datacenters, which in turn supported multiple sites," Harry explained.

As the project progressed, Harry learned that servers in the datacenters regularly ran out of memory, slowed down, then crashed.

"Server memory was expensive, so critical servers were prioritized to receive extra memory," Harry told On Call.

That upgrade program did not produce the expected performance improvements.

"The odd thing that repeatedly occurred was that we would upgrade a machine that crashed for lack of memory, then it would start crashing again a few weeks later at which point we would find the memory had not in fact been upgraded."

This was a very odd situation, so Harry dug into the paperwork and discovered the memory modules had been ordered, paid for, and delivered.

This being a government job, Harry's employer handled part of the gig, and another contractor did the memory upgrades.

Harry checked with that contractor, who insisted the upgrades had been made, and commented that perhaps the memory was being stolen.

• Techie cleaned up criminally bad tech support that was probably also an actual crime
• I was told to make backups, not test them. Why does that make you look so worried?
• User said he did nothing that explained his dead PC – does a new motherboard count?
• Techie fluked a fix and found himself the abusive boss's best friend

That was an interesting prospect, so Harry and his colleagues wrote a script that ran when a techie logged a complete memory upgrade job in the change management system.

The script looked up the quantity of memory the server should contain after the upgrade, then interrogated the machine to check of the quantity of installed RAM matched records.

"If that did not happen, the change could not be closed, the engineer wouldn't be allowed to leave the site, and people would be alerted."

When those alerts arrived, Harry's approach was to ask the techie if they could just check the memory was seated correctly – a cunning way of telling them they'd been rumbled as DIMMs sometimes need a little extra encouragement to connect all their finicky little pins to memory sockets.

"From then on, we never had issues with memory upgrades," Harry told On Call.

How have you detected, and deterred, light-fingered colleagues? Get your own light fingers moving across a keyboard and click here to send On Call an email so we can tell your story on a future Friday. ®