This morning the last of the hard drives arrived. Yippee! I gleefully unwrapped them, placed them into the Proliant DL380 chassis and tapped the Power button.

Let me state that the Proliant DL380 rackmount server is one of the noisiest chassis I’ve ever heard. It sounds like there’s a small aircraft experiencing engine trouble in my office.

After a couple of false starts (in other words, I tried to do things my way instead of the right way) I ran the SmartStart ™ CD-ROM and told the machine I wanted to install Netware 5.x.

But, you see, when the BIOS options says “Netware 5.x/6.0,” what it really means is “Netware 5.1/6.0,” and my copy of Netware 5.0 is system-non-grata.

Then I tried the so-called “manual” method, just booting from the Netware CD and letting things fly. Alas, it wasn’t meant to be.

That’s right, folks. It is impossible for Netware 5.0’s installer to recognize the array controller of a HP/Compaq Proliant DL380 G3. (Does Apple know that they’re using the “Gx” nomenclature, by the way?)

As things now stand, I’m sitting at my desk waiting for a quote from CDW on the cost of Netware 5.1 plus upgrade licenses for 150 connections. If the price tag doesn’t a) cause heart failures along Officer’s Row and/or b) get me fired, it’ll still be a couple of days before I can start on this project.

Again.

The best-case scenario, right now, is that I’m going to lose my Memorial Day weekend for this stupid project. I don’t want to think about the worst-case scenarios.

Let’s recap the story so far.

Almost exactly a year ago, pressed for network storage space, we evaluated two solutions. One involved expanding the existing server’s capacity, the other involved putting our faith in new, unproven technology. Naturally we went with the unproven technology.

July of last year, we suffered the first crash of the Snap! server. Considering it a fluke, we soldiered on blithely.

Fall of last year, the monthly crashes began. Stress mounted.

Two months ago, the decision to go with the original “other” plan is reached, and we order the replacement drives only to find that we can’t upgrade the original server without also destroying the operating system partition in the process, thus losing us several days of productivity.

A month ago, we came up with the final possible option, replacement of the main server. The clever part was that we’d use the drives that couldn’t be safely used in the old server, thus getting our money’s worth out of ’em.

Today, the new server showed up.

Guess what? We’re still not out of the woods. You see, we can’t use the drives we purchased for the other server. Turns out, upon investigation, that you have to buy special “Hot Plug” drives for this kind of server. Lovely. That’s just effing wonderful.

My greatest fear right now is two-fold: One, that someone is going to wake up and realize what a mess I’ve caused and boot my ass out the door; two, that it’s not over yet, and even after we get the new drives in and the machine is running, something else will go horribly awry.

It’s a gift I have, this knack for calamity. Too bad I haven’t figured out how to turn my talent for disaster to my actual benefit, eh?

Bad news: Just like on Friday, the Snap! server crashed this morning, costing me an entire productive day of work. The symptoms were identical to Friday’s, which were atypical for the monthly crashes we’ve grown used to.

Of course, when I say “grown used to” I mean “grown increasingly fearful of each and every month, to the point where jaws clench and stomach acids churn and tempers fray like cheap twine.” Or something to that effect.

Good news: We’ve finally ordered the parts for what will become the new central office server, replacing Tokimi (the Snap! server was named after the “evil” goddess from the Tenchi Muyo OAVs, yes) and Arthur (our current main Netware server, named according to Ben’s “Holy Grail” theme). It’ll only take me three solid days and nights of work to make the transition… if nothing goes wrong.

Starting either tomorrow or Wednesday I’ll probably be living at the office for a few days. Anybody have a cot I can borrow?

I’ve had a dark, unhappy day at the office, and it’s not over yet. Left for the evening are the tasks of upgrading RAM in the traffic department’s workstations and upgrading Tapscan (sales software for radio). Both tasks have to wait until the respective departments are fully done for the day, so… for the moment I have time to write.

I won’t bore you with the details of the day. Suffice to say that I had more excitement with Corporate about the new server we need, there were lots of weird little fires of which about half were manageable, and I have at least one full workstation rebuild in my immediate future. And how lucky am I that one of the dead PCs is a KNRK studio workstation? Yep, I’m just that lucky.

Oh, did I mention that because of this evening’s oh-so-necessary jobs, I have to miss the planned Wednesday-night viewing of Cowboy Bebop and blow off my family, who were planning to see a baseball game tonight? I could only do one or the other, but now I can’t do either.

If you need me, I’ll be in my office.

Let’s skip the “ado” and go right to the email I sent to my Corporate Boss a few minutes ago:

The story so far: Last summer, strapped for network storage space on the main office network, our hero researched alternatives and chose a Network Attached Storage device (the Snap! server) over attempting to upgrade the second RAID array in his existing Netware server. This solution cost about $2500.

Late last year problems arose; the Snap! server began a monthly cycle of crashes, costing the office a day of downtime on every occurrence. This became increasingly unacceptable, and Quantum’s tech support was lacking. (“We have no idea what’s wrong with your machine. We’d send you a replacement, but we don’t have any.”)

After one crash too many, more accurately two crashes on the same day, our hero swallowed his pride and admitted his mistake in choosing the Snap! appliance instead of expanding the Dell Poweredge server’s storage capacity. Hard drives were ordered, to the tune of about $1600.

Then the other shoe dropped. It is impossible, according to Dell, to delete only one array from a Poweredge 2300. One must remove all arrays, then recreate. This would mean fully reinstalling Netware onto the core office network server. All of a sudden, the “upgrading the Netware array” solution has become utterly unworkable.

The problem as it stands: Live with a day of downtime every month, in addition to the other problems the Snap! appliance is causing? Spend somewhere between two and five days attempting to recreate the Netware environment we currently enjoy, with no guarantee of completion date or even success of any kind? Drop another big bundle of cash on Some Other Solution?

I’m at a loss, and so I’m punting this up to the Powers That Be at the home office. I look forward* to discussing this at some very near date.

Yours, etc.

( * Okay, that’s a lie. I don’t look forward to anything involved in this giant ugly snafu anymore. I’m ashamed and disheartened at the whole damned affair. But we need a solution, so I’ll carry on until we have one.)

There are days when just being me is enough to wear me out. This is one of them.