(Oh no, not again.)
Fans are a little like nails. Well, one kind of nail anyway.
If you wear the right hat.
But I’m getting way ahead of myself.
At the end of ‘Installing‘ I said
Until they did it to us again the next year.
which sort of implies a sequel to that story. (Typical sequelitis. Just because someone read a sixteen month old post last week doesn’t mean another chapter is called for. Oh well, some sequels don’t suck so maybe this one won’t either. Laugh laugh, joke joke.)
At the end of ‘Installing’ we had a lab (not). All things considered, it was a pretty good lab (if I do say so myself) and saw heavy use for a number of years. (Except for that time it flooded. It was only on the second floor, after all.) It was clear, though, that there was ‘room’ for another one — perhaps one with another dozen or so seats, maybe slightly higher-end ones for some of the (mostly smaller) upper-year courses that had more elaborate computing requirements. Or for training. Or for the odd bit of research, perhaps. (Or for banging one’s head against, but that wasn’t exactly Official.) There wasn’t room in the budget, of course (there seldom is) but there was another computer company that could be approached for swag. (“One of your competitors gave us some cool stuff last year. How about you?”)
As it happens (not) the timing of the request was fortuitous because this particular computer company had just finished a trade-in program; they had offered heavy discounts on new hardware if used (but still new-ish) hardware was traded in. It was a popular program and meant that when the university came calling asking for donations, they were practically awash with potentially useful but used equipment. They could sell it for pennies on the dollar, of course, or they could give it away for free and claim full retail value on their corporate taxes.
Of course ‘newish’ doesn’t necessarily mean ‘in good working order’ — if they had the room, a lot of people kept broken hardware around just for situations like this. Trade-in programs often resulted in some truly horrifying gear being dragged out of storerooms and closets.
But we got lucky. Twice. The computer company that has never been explicitly named not only agreed to our proposal and gave us a bunch of stuff but what they gave us was all useful and in working order. (Of course they also wrote some truly ludicrous press releases and made some positively mind-boggling tax claims but corporations do that sort of thing as a matter of course. We were still happy.) They gave use thirteen (not) almost identical systems — eight with pretty good graphics cards, four with darned good ones and one without any at all to use as a server. (No problem — everybody had a spare frame buffer or six lying around, probably in that closet I mentioned.)
What they didn’t send us was storage, because disk drives weren’t part of the trade-in program. So we had fourteen (the ‘server’ had two) 424 megabyte disk drives spread over thirteen (not) machines which had to hold multiple copies of the operating system plus applications and those pesky user files.
Couldn’t be done. (That’s not entirely true. I’m fairly sure it could have been done but it would have made the setup overwhelmingly complex. ‘Overwhelmingly complex’ is a 5-point disad but we didn’t need the points. Or the headaches.)
But we did need storage. (Just like in ‘Installing’ but for a slightly different kind of vendor buttheadedness.) But unlike ‘Installing’ we had a bit of time so we (not actually Ms. Rose and I, but you know what I mean) had time to buy a small pizza box with a (for the day) large disk in it. Lots of space. All in one place. We were set.
First came names. As I’ve mentioned before, naming is important and a lot of people do it wrong. So we looked at the relevant standards document(s), reflected on what we had done the previous year and consulted widely (well, the guy in the next office anyway) before settling on a bunch of cartoon characters. Second came the operating system; I think we chose the one that maximized the number of sub-versions and sub-sub-versions in the version number. And since we had ‘enough’ time and ‘enough’ disk space, no middle-of-the-night epiphanies were required. Then came the apps but again, with time and space there were no significant crises (although we did learn that some software vendors take great pride in ensuring that installing their products is AS DIFFICULT AND TIME-CONSUMING AS POSSIBLE but that’s not an important part of the story).
And at last we reach the actual start of the story.
The school year started. The first pizza-fueled all-nighters (against the rules) happened. The first video games were installed (also against the rules). The bickering over usage and scheduling (guess) approached a crescendo. Thousands of pages of printouts were left on the floor. And one day one of the students came to me and told me that there was a funny noise.
Oh. That wasn’t good. Well, maybe that wasn’t good. I mean, computers make noise — in the case of servers, sometimes quite a lot — but all thirteen (not) of ‘ours’ (including the one being used as a server) were intended for desktop use in an office or lab environment so weren’t particularly noisy. Certainly they weren’t supposed to make funny noises. So I investigated. (It’s wise to confirm things for oneself; sometimes people don’t interpret things in exactly the same way you do. One day, for example, a grad student came to me and told me that his computer was behaving strangely. By this he meant that it had CAUGHT FIRE.)
When I investigated, I found… a funny noise. Well, not that funny, but definitely a noise. A mechanical noise. A loud mechanical noise permeating the entire lab but originating in the server closet. (Not completely unlike…)
“That’s a fan” I concluded. (Sometimes I can
see hear the obvious with the best of them.) I opened the closet. A fan on the server. It had to be a fan on the server.
But it wasn’t.
Well it was, of course, because the server was the only thing in the server closet, but it wasn’t on the server per se. It was coming from the not-really-a-pizza-box containing the external disk drive, the one ‘we’ had bought.
Those fans never break. Except that this one had. In particular, one of the blades had snapped off (which never happens). Given the speed of the fan’s rotation, the resulting imbalance resulted in some heavy-duty vibration (the case was threatening to dance across the table it was on) and, yes, a funny noise. Well, not that funny.
Well, I could get it replaced but in the meantime it would be nice to do something about the noise and keep the disk drive from hurling itself to the floor. That was easy — all I had to do was snap off the opposite blade and that would calm (and quiet) things down.
Or not. With a seven-bladed fan there is no opposite blade. So I piled a bunch of books on top of it and went to log the call with the brain trust.
It didn’t go well.
I got a moderately helpful guy at one of the vendor’s call centers that wasn’t in Cape Breton. “But those things never break” he said before asking what group I thought this call should be sent to.
“Storage, I guess” I said. It seemed reasonable to me. So he transferred me to the storage group. The storage group told me two things:
- Those things never break.
- This isn’t the right place for this call. We don’t do fans.
“Where should the call go?” I asked. I forget his answer but it was something unexpected. Systems, perhaps. He transferred me.
“Those things never break. And this isn’t the right place for this call.”
Round and round and round. At every step I was told that ‘those things never break’ and ‘let me transfer you to a different group.’ This went on for most of the next two days. (During that time I checked on the pile of books periodically to make sure the disk drive hadn’t hurled itself to its doom.) Finally I got someone who said he could help. After the obligatory preliminaries (“but those things never break”) he said that he could help me. Yay.
Just give me the FRU for that part.
Some background: field service is all about FRUs. A FRU (Field Replaceable Unit) is the part that ‘they’ send you to make your broken thingy non-broken. Often the same part will have multiple identifiers, each assigned by a different group. Sales might give it one number, for example, Service another. You need the part number that the service arm knows this part by. I needed a fan for my not-really-a-pizza-box so he was asking me how he could identify that part and arrange to have it sent to me. FRUs/part numbers are on stickers attached to the part in question. Unfortunately, the fan didn’t have a sticker on it. Nothing in the case had a sticker on it — except for the case itself.
“Umm, I don’t have one.”
That was bad. “That’s bad.” Turns out that he could have sent me an entire enclosure except that I had already made the mistake of telling him (and a number of his co-workers in other groups) that it wasn’t the enclosure that had failed, but only a specific part of it. And he wasn’t allowed to send me ‘more part than I needed.’ Oh. “Perhaps you could look up the part number for the fan?” He tried. Apparently there wasn’t one. Because, you know, they never break.
To be fair, it wasn’t his fault — he was bound by unassailable corporate logic: I was allowed to have any part I needed. I just couldn’t have an enclosure because I didn’t need one and they couldn’t send me a fan because they had no part number for it. Helleresque in its simplicity and finality. And downstairs the damaged fan continued to whine and vibrate.
There appeared to be no way out. But wait! I hadn’t tried the guy named Mike.
Over the years I’ve dealt with service personnel from many of the major players in the computer industry. I’ve dealt with a Brian, a Tim, a Ken, a Dale, a Nancy and a few others but for some reason a disproportionate number of service personnel have been named ‘Mike.’ My local guy named Mike even married a woman named Mike and they had a daughter named Mike. For reasons I’ve never understood, Mike is an important name in the service industry.
I called Mike. Could he help me with my little fan problem? Of course he could.
It turned out that the solution was to consult a list of FRUs for the not-really-a-pizza-box (a list that was apparently not easy to find) and find a FRU that they were allowed to send me that also contained the fan.
Piece of cake.
So a couple of days later I received a box. It contained a chunk of metal, a couple of SCSI interfaces, some cables, a red LED and… a fan.
So after N phone calls, several service departments, some
judicious rule-bending creative reinterpretation of corporate Procedures and no more than a week, things were calm once again.
Until someone walked off with the printer the day after I installed a new toner cartridge in it.