Working as a field technician for a now-defunct local tech support organization, I was required to support a variety of clients in about a 75-mile (more or less) radius of our home office. My duties included standard support, installation, and configuration of the then-current Windows 95/98 and Windows NT/NT Server operating systems, as well as hardware builds and upgrades -- pretty standard stuff for a field tech.
In addition to the usual one-off calls for support, we had a few clients who used us for help desk and desktop support (one of our "products" was a 10-hour block of support time for a reduced price). One of these clients, who happened to be an MCSE Network Specialist at a local publishing company, needed what were being called "data collectors": stripped-down Windows 98 clients connected to a Windows NT-based network, which allowed only the input of specific production data for transmittal to the server. The order came in for three clients, and I had them ready, tested, and delivered within the week.
[ Got amazing IT tales, real-life experiences, lessons learned the hard way, or war stories from the trenches? Submit it to InfoWorld's Off the Record blog. If we publish your story, we'll send you a $50 American Express gift card. | Join or start an IT conversation in InfoWorld's Adventures in IT discussion area ]
Fast-forward to the middle of the next week. I was at a site at the far southern reaches of my territory performing an upgrade of two DOS/Windows systems to Windows 95 via floppy drive and a concurrent upgrade of a DOS-based accounting system running QuickBooks 1.0 for DOS to a new Windows 98 "server" running QuickBooks 3.0 for Windows. I was knee-deep in floppies, network cards, and programs when I received an emergency page from my boss: One of the newly built data collectors would not communicate with the server, and a production line was down. The faulting system had been replaced with a second system that also would not communicate I was instructed to drop what I was doing and go to the publishing company at once. After explaining to my current client what had occurred and promising to return as soon as I was able, I left and started the 90-mile drive to the distressed publisher.
Now this publisher (not the network specialist, but the boss) was problematic to work with on the best of days; today he was as angry and short-tempered as I'd ever seen him (he had my cell phone number and contacted me a few times on the 90-minute drive to his office, and with each contact he grew angrier and shorter-tempered: where was I, didn't I know that his production line was down, how soon would I get there, how could I perform such shoddy work -- well, you get the picture).
Arriving on site an hour and a half (more or less) after the first contact, I was met at the front door by the network guy, who proceeded to ream me up and down about the "crappy stuff being pawned off" on him. Proceeding to the workstation, I signed in and attempted to connect to the network: no connection. I tried pinging the loopback: all was OK. I pinged the server: no reply. I checked network properties and Device Manager, but saw nothing wrong. The hardware gave every indication of being good, and the operating system was installed from an image that had been used successfully before.
OK, then, what to do next -- thinking that the network adapter was going bad, I prepared to do a field replacement. Moving around to the back of the system, I thought something looked funny -- there weren't enough cables plugged in. A closer inspection revealed the network cable coiled up neatly and duct taped to the support column behind the computer.
Let's see now: 90 miles at $0.285/mile, 1.5 hours of travel time (debited against the support block) at $75/hour, half an hour of actual support time, and let's not forget travel expenses back to the interrupted upgrade. That's about $200 to "plug it in, plug it in." Talk about an unhappy camper -- but at least at the end it was not directed toward me.