I was interviewed a few days ago by a reporter from the Sun-Sentinel. Surprisingly, it had nothing to do with either space or blogs. He was just looking for people who had recently moved here, to get a newcomer’s perspective on hurricanes. He got my name (actually, Patricia’s, but she was at work when he called, and I was home) from our next-door neighbor, who is apparently involved with the local Welcome Wagon.

Anyway, here’s the result.

The Big Trunk has a tribute to the late and justly lamented Ray Charles on the first anniversary of his birthday since his recent death.

The Big Trunk has a tribute to the late and justly lamented Ray Charles on the first anniversary of his birthday since his recent death.

The Big Trunk has a tribute to the late and justly lamented Ray Charles on the first anniversary of his birthday since his recent death.

In another consequence of the hurricane, we had a thousand-dollar repair bill on the BMW. Though actually, in a sense, the hurricane may have simply made us aware of a problem that had been ongoing.

As you may recall, I was blogging with power out by running my modem and laptop off a voltage inverter hooked up to the car’s battery. On the Saturday morning before we got full power back (even with partial power, I couldn’t get the DSL modem to work from the house current), I was letting it idle in the driveway to recharge the battery, while I watched the (infuriating) Michigan-Notre Dame game.

Suddenly, I heard a loud hissing sound in the driveway. I ran outside and the car resembled a steam locomotive, with its hood obscured by all of the dangerous DHMO in gas phase. I looked at the dash, and the temperature gauge was pegged. I shut the engine off, and let it sit.

The next day, I tried topping it off, and the water was pouring out as fast as it was going in, through a crack in the filler tank that had apparently ruptured.

I tried driving it, and while it ran smoothly, it had no power (top speed about ten MPH), which really started to concern me, because I was afraid that I’d warped or cracked the heads on the V-6 (though that didn’t make sense, given how smoothly it was running).

I also couldn’t figure out how I’d managed to drive it across the country two weeks previously, through the Southwest in the hottest part of summer, with no problems at all, but then have it overheat idling in the driveway.

Then, of course, the little cartoon lightbulb went on over my head. It has an electric fan to pull air through the radiator when the car isn’t moving. Most likely scenario–the fan had failed sometime in the past, and I hadn’t noticed it because I’d rarely let the car idle motionless for that long previously.

Sure enough, when we took the car to the repair shop, that was exactly what happened–a resistor had gone bad and the fan had quit fanning. Of course, the resistor isn’t replaceable–you have to buy the whole fan unit from Wolfsburg, at over three hundred dollars. Also, it was a cascading failure–the incident, in addition to rupturing the plastic fill container, wiped out the water pump by running the bearing dry, and the thermostat. All told, about a thousand bucks, including labor.

The mechanic told us that he hadn’t seen this happen before, but it didn’t surprise him, because BMW had gone to a single, non-redundant fan about that time. I’m not sure why they don’t just drive it off a belt like in days of yore, but I guess most modern car manufacturers prefer to only run it when it’s needed, perhaps to not be a useless power drag, since it’s rarely needed. I know that I have one on my eighteen-year-old Accord that’s never had a problem. And of course, this would have been avoided if I’d been sitting in the car while it was idling, because I probably would have noticed the temperature creeping up (which would have been a much less costly way of discovering the problem than the catastrophic failure that it actually endured). But there’s no telling how long it hasn’t been working, or how long it would have been before I discovered it, if it hadn’t been for Frances.

The good news is that the engine is all right. The power problem wasn’t caused by a lack of compression, but by a slight warping of the throttle body so that the valve couldn’t open properly. After cleaning it, they got it working again.

The storm apparently did a U-turn up in the mid-Atlantic and headed back down here. It’s been dumping rain on us overnight, and through the morning, on its way west. It’s headed across Florida and back to the Gulf, where it may reform. Keep an eye out for it, Texas.

That’s one of my pet peeves. It’s very frustrating to get someone in tech support who a) has no knowledge at all other than what’s on the checklist in front of them and b) certainly doesn’t know as much as you do, particularly when it comes to first principles or logic and c) doesn’t know that they don’t know, and answer your questions with gibberish, often in a condescending way as though you’re the idiot.

This happens all too frequently, and it happened again today with DirecTV.

I’ve got a weird problem. Intermittently, I’ll lose signal on some channels, starting with pixelation and audio breakup, deteriorating into complete loss of satellite signal. The last time I was having a problem like that, in California, it turned out to be a bad LNB, so I went out and bought a new one, and installed it.

The problem persists. Now here’s the really strange part.

When I run a test on the individual transponders while it’s acting up, the odd-numbered ones are fine, with signal strengths in the nineties. The even-numbered ones are zero across the board. What kind of failure would cause this kind of selective behavior? What’s different between odd-numbered and even-numbered transponders that would cause one to be fine and the other useless? I thought it might be something in the logic programming of the receiver, but I hooked another one up, and saw exactly the same behavior.

It seemed like an intriguing problem to me, and I figured that if I talked to DirecTV about it, they’d have some kind of ready explanation. And indeed they might, if you could actually talk to someone who understands how the system works, instead of a drone with a checklist, who not only couldn’t explain it, but didn’t seem to think it remarkable. He simply kept leading me through his check list. When I explained to him that even was bad and odd was good, he could only repeat, “that means it’s seeking signal,” as though that actually meant something significant and useful.

The bottom line was that he said he’d send someone out to look at it. On October 12th. I’m tempted to make another attempt to see if this time I can at least get someone with a little intellectual curiousity, and ability to think, but I’m wondering if anyone out there has any insight.

It’s not the receiver, it’s not the LNB. It could be dish aim, but the problem with this, as with all hypotheses, is that it doesn’t explain why I have a perfect signal on odd transponders and zero on the evens. Same thing with a bad cable, which is the only other thing that I can try at my end.

If someone described the symptoms to me, and I had no other knowledge, my first guess was a problem with the transponders on the satellite itself. But that implies that everyone else would be having the same problem, and it’s hard to imagine that occurring for long without DirecTV doing something about it.

Anyway, I guess I’ll try swapping cables, just because there’s nothing else I haven’t tried, but if that’s the problem, I’ll be very interested to understand what kind of cable failure would affect half the transponders, and only those with even numbers.

[Update on Sunday morning]

Thanks for the input. I finally did get in touch with someone at DirecTV who knew what they were talking about, and he told me that the most common reason for this behavior was cabling, with receiver second, and LNB a distant third. He couldn’t explain the physics of it, but said that in his experience, it was usually a bent pin on a cable, or some similar problem.

I went out this morning, and started tracing the wire. I found a corroded connector where it goes into the house on the active line. When I moved it to the other side (on a line we aren’t currently using), which wasn’t corroded, the problem seems to have gone away. I’ll have to watch for a while to see if it recurs, but that looks like it was the culprit. I’ll have to go to Home Depot and get a replacement for it, and seal it back up out of the weather (it only had electrical tape wrapped around it, which is probably why it went south).

Here comes Jeanne, pointed right at us.

My server is having trouble with DNS. It can ping the internet by IP, and when I ping by name, it makes an attempt and echos the IP, but cannot ping. My /etc/resolv.conf file contains only “nameserver 192.168.2.1” which is the address of the wireless router.

Am I doing something wrong?

[Update]

I finally found the DNS servers for AT&T. When I put them in resolv.conf, everything seems to work fine. Thanks for the suggestions.

I’ve got a weird (at least to me, I’m hoping that it will be obvious to someone else) problem with my network.

I’m running an 802.11(g) router. I’ve got my server and my desktop ethernetted to it, and I’m using my laptop in wireless mode. All machines are seeing the router fine (and getting Internet access), but the laptop is not able to see the other two machines and vice versa on the network. The laptop can ping everyone, but the laptop cannot be pinged. I’m running in encrypted mode on the wireless connection, and wondering if that might be the problem. It doesn’t seem like it should be, because I would think that the signal is being decrypted by the router and available through ethernet.

Does anyone have any ideas?

[Update]

D’oh!

I’d changed my network from 192.168.1.* to 192.168.2.* and forgot to tell Zone Alarm on the laptop.

Problem solved.