What we have here is a failure to define. Is it:[ul][li]A unit that stops working, then repairs itself, orAn intermittent failure?[/ul][/li][/QUOTE]
Very hard to tell without knowing the nature of the defect. Did it fail for a week, heal itself, work for a week, and then break again, or was it intermittent with a long period of working/non-working?
The first IC failure I mentioned truly healed. The second was caused by coupling between a power bump and a signal line. That never healed per se, but it sometimes took a long time to sensitize the failure. (Days even in the lab.) I’m not sure what you’d call that one.
Intermittent is just a symptom, not a definition. When you diagnose a failure to its root cause, I’ve never seen a real intermittent one.
I fixed my issue. It started grinding. Against my better judgment I put a DVD in for a few seconds of grinding, ejected it, and tried again. With the third repetition, out came a foam disc used to separate DVD-/+Rs. I guess it stuck to the bottom of one of the discs and got jammed in the gears. Seems to be working fine now.
Whether hardware or software programming, it is often more of a relief to know it is truly broken then to slap a bandaid on it and assume it works.
I wrote a column about rebooting, inspired from the maintenance guy having to reboot the TV set in a hotel room. And someone told me that she had to frequently reboot her digital picture frame.
Rebooting has moved from an esoteric strategy that only computer scientists knew to the number one thing to try when your DVR goes out.
This is probably true to be honest. I was going to point out that blue screens do not always indicate hardware failure. Device drivers can cause blue screens, and the flaws in the drivers can be subtle enough that on a computer with healthy hardware you still get a couple bluescreens a year, give or take. Such rare failures are very difficult for the authors of the software to ever pin down so they don’t.
However, in your friend’s case, a blue screen right at turnon is a case where the drivers are in a known state at bootup and should be perfect for that simple use case. Most likely that is a hardware fault.
Right. Either the unit passed a final test before being shipped, and became defective during shipment, or it wasn’t tested at all before being shipped.
And the replacement Drobo I mentioned in post #10 is still working, so I guess I’ll keep 'er.
I had a hard drive that seemed to be entering its death throes – every few hours it would lock up, make a weird whining/grinding noise, followed by a “thrashing” sound before it settled back into place. (Accompanied by a blue screen or program crash, though it never totally hard-locked my computer.) After some investigation I reasoned that it wasn’t actually an impending failure, but a combination of background programs, plus Minecraft, which were overtaxing the drive and causing it to hit some kind of failsafe and reset. So I uninstalled and/or reconfigured the programs that were causing the problem, as well as avoiding things that were making the drive thrash in the first place.
Naturally, I figured it was just a band-aid, and only matter of time before the drive failed completely. But no – the noises eventually disappeared completely, and it worked perfectly for another 6 to 7 years until I mothballed the whole computer for an upgraded system. Not sure if I’d count this as a magical “self-repair” though, since I did take steps to try and fix it.
When a HD starts filling up, areas of the drive that had seen little or no use but are flakey will cause the noise. In the old days it would be the OS that notices it and maps the sectors as faulty, moves the data to somewhere else and then the bad sectors remain unused.
Now, it’s the drives themselves that notice the problem and do their own remapping. That goes by the label SMART. Spare sectors are set aside initially to replace the ones that are detected as going bad. Once the spares are used up, then the OS takes over like before but the HD is basically toast anyway. You can check the SMART status of your drives with tools from the makers or in the BIOS sometimes.
Since several sectors in a given area might be bad, you get the noise for each one as they get mapped out until they are all marked bad. And then you’re good for a while.
JUST went through this. My Samsung all in one desktop pretty much wouldn’t do anything when I fired it up last Monday. The “home” desktop screen would come up but trying to run any program (app, whatever) wouldn’t work. When I clicked the icons at the bottom of the screen they would light up like they were going to open but then go back out. Tried powering all the way down then back up, nothing. I was able to eventually get the control panel to open and tried to do a system restore but it would time out after hours of seemingly running. After a minute, when my screensaver would normally come on it went to a black screen. I messed around with it for three days and was getting nowhere. Thursday night I messed around with it for an hour or more, no luck. I went to get dinner started, took a shower and intended to go back and yank the cables and remove it to make room for it’s replacement. Walked back to my office and my screensaver was running. Bumped the mouse and started clicking stuff and all was and is working normally. Between the time it malfunctioned and recovered it had been rebooted numerous times but also left running for 24+ hours.
I still don’t trust it and am planning to get a backup, backup drive and maybe new desktop just in case.
Not always. Does it boot in safe mode? When my Vista PC got a blue screen at turn on, after tons of debugging I found that the root cause was flaky antivirus software. Uninstalling that and replacing it fixed the problem.
Hardware in general is much more reliable than software, so your first paragraph is correct. It was always fun when our hardware was accused of being buggy and it turned out to be a firmware issue instead.
A bunch of computer programmers are driving home from a conference when they get a flat tire. They get out of the car to look at it, and one of them opens the trunk and starts digging for a jack and the spare.
“Wait a minute,” one of the other ones says: “Before we do that, let’s all get back in the car, get out again, and see if the tire’s still flat.”
I have an iPhone app that sometimes won’t execute a task when you push that button, and pushing it a few times will get it to work. I’d observed occasionally I could push the button 20 or 30 times without it working, and then it would work. Once I got curious and stood there push push pushing the button, and counting, and got all the way up to 110 attempts, and then it worked. People will say it’s crazy to try pushing a button 110 times, that it obviously just does not work, but occasionally the 111th attempt is the successful one.
People may find this astonishing: when I took a computer science course in the late 1970s, they actually taught that computers were deterministic systems, and could be counted on to do exactly the same thing every time. That was actually the way they thought back then.
They are supposed to be deterministic, but underlying them is analog circuitry that is not. What you are describing sounds like an analog power supply failure of some sort - reason it worked on #110 was probably a temperature change from the previous attempts.
Also note that unlike the late 1970s, their software is so immensely complex that it can feel non-deterministic.
And, technically, since we use multi-threading and other hardware devices that wobble a little in timing, it is very easy to make a program that isn’t completely deterministic, actually.
I was programming in the 1970’s and knowing that was true often led us to find out why it wasn’t doing the same thing each time. Some unknown factor was responsible (hardware or software), and it was my job to find out what that was and fix it. Sometime before that rocket took a wrong turn and started WW3.
But our first inclination was to blame cosmic rays. Much like Chicken Man, They’re everywhere.
Yep, this is true. I got the laptop home and I was able to play a DVD - although it took a while to get to the title menu and then the first few minutes of the movie were stuttering, after that it played OK, and I was able to play a second DVD after that. Last night, however, neither of those DVDs would play, none of the others would play. I really wish the computer repair guys had just gone with my initial request to replace the drive.
Another possible reason as to why it worked when you took it in is due to the cold. While the computer was in transit to the shop, it may have been colder than the normal operating temperature in your house. Seeing that you are MN, that’s probably a true statement. Plus, if you had the computer actually turned off during the trip, then the computer’s temperature would also have dropped. It is possible there might be a bad solder connection that gives intermittent results when heated up. If it starts acting up again, try turning it off for 15-30 minutes to see if the problem comes back immediately after you start it up or if it happens after the computer has been on for a while.
My computer went out completely, blank screen. I messed with it a few days and then went and bought a new one. I plugged it in a few months later and now it seems to work fine.
Frequently automotive problems “fix themselves” by the time you get the car into the shop (or such would seem to be the explanation for the mechanics not being able to reproduce them).
Human malfunctions disappear by the time you get to the doctor’s office. We tend to intermittently go out of whack or self-heal.
