As was foretold, we've added advertisements to the forums! If you have questions, or if you encounter any bugs, please visit this thread: https://forums.penny-arcade.com/discussion/240191/forum-advertisement-faq-and-reports-thread/
Options

Bizzare computer crashes, different from anything I've seen before.

DigitoDigito Registered User regular
I generally consider myself a pretty tech-savvy person but this one's got me properly stumped right now. So this only started earlier today, but already my machine has crashed a good four or five times in a very unusual way. The one consistent thing about it is that the first sign it's about to crash is Explorer shuts off with no warning, no toolbar at the bottom, etc etc. Other programs seem to cut out mostly at random, sometimes they do sometimes they don't, and then a very short moment later the whole thing crashes. Sometimes it bluescreens, sometimes it crashes to a black screen then reboots, more often it crashes to a black screen and does absolutely nothing until I manually shut the machine off then back on. Last time the mouse cursor stayed on screen and remained responsive, and the keyboard still seemed to respond as well (The various -lock lights would turn on and off.)

Not gonna lie, I'm perplexed by this one. Virus scans have come up clean, I just tried resetting the RAM as I had a hunch I might have jostled it loose the last time I moved my computer, but it hasn't been long enough for me to tell if that fixed it or not. The computer seems to run fine for a couple of hours with no obvious issues before suddenly giving up the ghost, and there seems to be no obvious link between computer load and the odds of a crash.

Anyone have any ideas?

Posts

  • Options
    CambiataCambiata Commander Shepard The likes of which even GAWD has never seenRegistered User regular
    Do you know what temperature your CPU is running at? Has your CPU fan or any of your case fans stopped working?

    "If you divide the whole world into just enemies and friends, you'll end up destroying everything" --Nausicaa of the Valley of Wind
  • Options
    SkeithSkeith Registered User regular
    There's a program called Speccy that'll tell you what temp your CPU is at if you don't have any software like that already.

    aTBDrQE.jpg
  • Options
    DigitoDigito Registered User regular
    Case fans still seem to be cooperating as is the CPU fan, and my CPU's idling at around 36c. Given it's crashed when completely idle with no apparent heat spikes to go with it, overheating doesn't seem to be a likely culprit right now, though I'll try to keep a close eye on that next time I'm doing things that involve heavy loads.

  • Options
    HevachHevach Registered User regular
    edited June 2015
    Does it still happen in safe mode?

    If so, it's most likely a hardware issue (though Windows itself being trashed to an unusual extent is not out of the question). Memory, hard drive, and CPU are all big ones for weird crashes. If not, drivers or malware not showing up on scans are likely culprits. Check task scheduler and msconfig (or the startup tab of task manager for windows 8) for anything you don't recognize, and event viewer for any service or driver errors that seem to match up to the crashes.

    Hevach on
  • Options
    ReznikReznik Registered User regular
    Have you been able to grab the stop code when it blue screens? If you can, I'd google that. It'll point you towards something specific (unless it's a completely wacky error)

    Do... Re.... Mi... Ti... La...
    Do... Re... Mi... So... Fa.... Do... Re.... Do...
    Forget it...
  • Options
    Dunadan019Dunadan019 Registered User regular
    honestly I suspect graphics card since you could use your mouse/keyboard.

    mouse is controlled by windows/cpu/mobo.
    keyboard is cpu/mobo

    video card going bad can cause random program crashes and crashes to black screen.

    also, check your event log.

  • Options
    DisruptedCapitalistDisruptedCapitalist I swear! Registered User regular
    I had a situation about a year ago caused by a faulty PSU. All the memory tests and temperatures came back normal. Even the voltage monitors were reporting normal. It would just freeze with no error messages or log entries. So just on a hunch I bought a new PSU and it started working fine again.

    "Simple, real stupidity beats artificial intelligence every time." -Mustrum Ridcully in Terry Pratchett's Hogfather p. 142 (HarperPrism 1996)
  • Options
    JohnnyCacheJohnnyCache Starting Defense Place at the tableRegistered User regular
    I had a similar situation once and it was bad GPU memory. I was lucky in that it was one card of an SLI setup and I just ditched one card until the machine was obsolete.

  • Options
    DigitoDigito Registered User regular
    The only thing under critical events in the event log (although I'm not too familiar with its use, so I may be missing something.) is the computer complaining about powering down unexpectedly, and I should mention while I was still able to move the mouse and keyboard ONCE, another time it crashed both became completely unresponsive. It hasn't been consistent as to if they'll stay working or not when the crash happens. Furthermore, even when they were responsive I could still tell the computer wasn't doing anything in the background. (There was no sound, no programs responding to keyboard input, nothing.)

    It wasn't happening at all yesterday after I removed, dusted (with compressed air) and re-seated the RAM, so I thought for a bit that it was fixed only for it to happen again just earlier today. I also ran an overnight memory test with no issues whatsoever, which leads me to think the memory is probably not at fault. The fact the crashes happen gradually, if quickly, also leads me to suspect the power supply is not likely as I'm under the impression that a PSU-triggered crash is functionally instant, and the machine never actually loses power until I manually power it down. The only problem with testing to see if it crashes in safe mode is that this time, it took over a day to crash again, and I'm not really in a position where I have enough time to sit the machine in safe mode for that long seeing if it crashes or not.

  • Options
    bowenbowen How you doin'? Registered User regular
    PSUs are finicky things. They can provide power randomly and start and stop randomly. Components like clean power, so if you've got a capacitor that's leaky or just old and delivers bad power, it can manifest in odd or random ways.

    Random usually indicates power.

    RAM and CPUs usually fail in predictable ways (writing to a certain portion of it or working with a certain register). Drivers can be random, but are often hit with something predictable (always crashes when you do a certain task).

    You could have a motherboard issue (leaky capacitors can manifest here too). It's unlikely. Switch out the PSU.

    In terms of things you can troubleshoot, this is one of the cheaper ones to do it with.

    Also it could be a corrupt windows install. Malware does shit like this. But let's rule out the PSU first, and then reformat windows.

    not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
  • Options
    DigitoDigito Registered User regular
    edited June 2015
    Well here's the problem, as it stands I'm A: Broke, and B: I have no other PSU's sitting around to switch out to except for a MUCH older one that I know for a fact doesn't have the horsepower to drive this machine, I'd be astonished if it even managed to boot in the first place with it in. Incidentally, of all the crashes today it seems to be a much more consistent error than before, just saying that "the system has shut down because a process crucial to system operation has been unexpectedly terminated or exited". wasn't able to get the stop code yet, it looked far too big to easily write down, but next time it happens maybe I can snag a picture of it with my crappy old cellphone and then google it that way.

    I will say going by that error message my gut instinct says Windows? Nothing else has seemed amiss, but it's been a fairly long time since I last reformatted this thing. (Some number of years) Maybe updating my drivers might not be a bad idea either, just to see if that does anything.

    Incidentally, is there anything I could look for on the computer/PSU itself to identify potential capacitor problems?

    EDIT: Got the stop code, or part of it at least. it's something in the 0x000000F4 range, and some googling indicates this is very likely a failing part somewhere. And yet a memory test didn't turn up any errors...? But it's odd that reseating the RAM fixed the problem for a day or so, I wonder if the slot is failing to hold the chip entirely secure?

    EDIT 2: Could it be a USB device causing it? I just realized this only seemed to begin once I switched to a USB wireless router, though lacking literally any other way to get online at this apartment that's going to be a difficult one to test...

    EDIT 3: Currently running on a single stick of memory to see if the problem persists. One thing making it annoying to track down is nothing seems to cause it, and re-seating the RAM does appear to have some effect in significantly delaying how long it takes before problems happen again. Unless that's an incredibly strange coincidence, this is leading me to think bad memory, but it's gonna take a few days to hammer down the time to test both sticks together, test them on their own, and then test different slots. How many passes should I let memtest86+ do its thing for to be reasonably certain?

    Digito on
  • Options
    EnigmedicEnigmedic Registered User regular
    My wife's computer was having weird crashing problems too a few weeks ago. After messing with it for a while we ended up actually taking it to the computer place by our house and they fiddled with it. It turned out to be that the hard drive had pretty much died. They put in a new one and everything has been fine with it since. Luckily we could get into safe mode and throw everything important onto a usb drive so really we just had to pay for a new HD and a little labor.

  • Options
    DigitoDigito Registered User regular
    I'm hoping that's not it, but that's beginning to be my suspicion. So I ran memtest86+ overnight and after about nine and a half hours there were no errors and it hadn't crashed. Based on that, it's probably not the memory and I feel like it's not likely to be the power supply or CPU either. And yet it crashed again. This time it crashed when I tried to tab back into my browser and it locked up, then when I tried to Control/Alt/Delete it stalled out trying to load security settings on the full screen that pops up on Windows 7 when you hit that, and several seconds later bluescreened on me with the same error I've come to know and loathe.

    Based on that I feel like it's far more likely to be either Windows itself or possibly the SSD that's dying but I'm not 100% sure, any way I could potentially narrow it down further? Replacing the drive would cost money at a time when money is painfully tight, so I want to be absolutely certain I know what's failing before I try to replace something. Either way, this feels like a good time for a backup. One last thing I thought was kind of odd, this only really seemed to start happening when I moved my computer from one room to another, but it doesn't seem like I jostled anything loose or damaged anything...

  • Options
    bowenbowen How you doin'? Registered User regular
    What brand of SSD do you have?

    not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
  • Options
    bowenbowen How you doin'? Registered User regular
    Either way it's a good idea to reformat, this will clear all issues with programs or drivers and get you back to a clean slate.

    I did find some interesting tidbit that turned up, that 4F can be caused by a dying CMOS battery on your motherboard. Maybe replace that?

    not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
  • Options
    DigitoDigito Registered User regular
    A dying CMOS battery you say? You actually just reminded me my CMOS battery died some time back and since I didn't have a replacement on-hand, I had to use an old one dug out of my previous computer that still had some juice left in it. The battery in there right now has been through two different machines, and I think replacing it with a fresh one is going to be my next move. I believe we have one laying around here somewhere.

    For the record my SSD is one of the Samsung 840's, though I forget which one offhand. According to Samsung's own tool though, my SSD is in good health and isn't showing anything abnormal which reinforces the possibility of it being the CMOS battery. I'll get that changed and see if it crashes again.

  • Options
    bowenbowen How you doin'? Registered User regular
    Yeah 4F can also represent some sort of driver issue from Samsung SSDs too. No idea if that's the case.

    not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
  • Options
    DigitoDigito Registered User regular
    edited June 2015
    That might be the case after all, even with a brand new CMOS battery it just crashed again. This time I could actually see programs and services click off one by one before the system rolled to a halt and bluescreened again, which was pretty surreal. I'm gonna have to look into if it's a driver issue then, I'm beginning to run out of ideas. Notably, it doesn't seem to crash when I'm running things that aren't on Windows, but I can't tell if it's Windows or the SSD that's screwing up here.

    EDIT: Well, that sucked. Reformatted the machine, had a crash before I'd even gotten the internet working again. Still haven't gotten my old wireless adapter to work (Windows outright won't recognize it, stuck using a far slower one), the crashes aren't fixed, and now I'm scratching my head trying to figure out where to go from here. I wonder how to properly update the driver on an SSD? I'm running out of things to try.

    EDIT 2: Wow, not only did I not fix the problem but now I actually have a new one. So my TP-Link TN-WN722N worked fine before the reformat, but after reformatting Windows literally refuses to see it at all. I even installed the drivers and nothing, if I plug it into a USB slot windows doesn't even acknowledged that I've plugged in a device. I'm stuck using a backup adapter right now, but I'm peaking at just under 3Mbits downstream and that's when the connection is stable enough to do anything. Anybody know how to get my old adapter working again? Last time it was plug-and-play, now nothing I do seems to get Windows to see the device at all. In fact, when I plug it in Windows doesn't even make the "you plugged in a device" sound. It just flat out doesn't react.

    (Actually, come to think of it, should I give my newfound wireless adapter problems their own thread?)

    Digito on
  • Options
    SeñorAmorSeñorAmor !!! Registered User regular
    Digito wrote: »
    EDIT 3: Currently running on a single stick of memory to see if the problem persists. One thing making it annoying to track down is nothing seems to cause it, and re-seating the RAM does appear to have some effect in significantly delaying how long it takes before problems happen again. Unless that's an incredibly strange coincidence, this is leading me to think bad memory, but it's gonna take a few days to hammer down the time to test both sticks together, test them on their own, and then test different slots. How many passes should I let memtest86+ do its thing for to be reasonably certain?

    Apologies if I missed it, but what were the results of your memory test?

    Typically one pass of memtest is sufficient, but you can let it run twice if you want to make absolute sure. IMO anything over that is overkill.

  • Options
    bowenbowen How you doin'? Registered User regular
    Is that tplink thing a USB wifi adapter?

    Try removing it completely and see if that helps. Get your computer down to the bare essentials.

    Hard drive, CPU, motherboard, graphics on the motherboard if possible. Maybe CDRom.

    I'm still suspecting power supply, but I wouldn't rule out motherboard either at this point.

    not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
  • Options
    DigitoDigito Registered User regular
    edited June 2015
    SeñorAmor wrote: »
    Digito wrote: »
    EDIT 3: Currently running on a single stick of memory to see if the problem persists. One thing making it annoying to track down is nothing seems to cause it, and re-seating the RAM does appear to have some effect in significantly delaying how long it takes before problems happen again. Unless that's an incredibly strange coincidence, this is leading me to think bad memory, but it's gonna take a few days to hammer down the time to test both sticks together, test them on their own, and then test different slots. How many passes should I let memtest86+ do its thing for to be reasonably certain?

    Apologies if I missed it, but what were the results of your memory test?

    Typically one pass of memtest is sufficient, but you can let it run twice if you want to make absolute sure. IMO anything over that is overkill.

    Four passes of memtest over around nine hours (four and a half, technically, but I needed my computer to get some job hunt stuff done so that was all the time I could give it) with zero errors and the machine not crashing. And yeah, the TPLink thing is a USB wifi adapter, I forgot to clarify about that. I was pretty out of it last night, sorry about that. I did update my SSD's firmware just in case that may be contributing.

    The only tricky thing with stripping the computer down to the essentials and seeing if it crashes is a matter of time, though they generally happen every two to three hours sometimes they can take longer, and at the moment circumstances make it difficult to go that long without really using the machine for anything. I'm gonna see if I can arrange some things so I can get things done without relying on this machine so I can better carry out those tests though, or maybe pull out half the guts and let the machine idle overnight and see if it's alive come morning. With the SSD's firmware updated it hasn't crashed again yet, but I'm gonna work on some things throughout the day and see if it crashes again, being as I booted it up roughly three hours ago I suspect I'll more than likely be editing this post within the next hour or two with an "Oh, it crashed again" update, and then I'm gonna try to pull the machine apart and start tracing down exactly what part's causing it because it'd have to be hardware from there.

    Oh! One thing I should clarify, though I initially suspected the adapter had a hand in the crashes as my errors only really started when I began to use it, the first crash I had was before I even got the adapters reinstalled or plugged back in, while I was trying to find the discs to get the motherboard's own drivers up and running. So at this point the wireless adapters have been ruled out as a culprit.

    Lemme know if I need to clarify anything, I'm writing this post on about an hour's sleep so it's probably a little more stream-of-consciousness than most of mine usually are.

    EDIT: It feels a little early to celebrate but it hasn't crashed on me yet since I upgraded the SSD firmware, though I'm going to continue to keep my eyes open for that. In the meantime I'm gonna focus on trying to get my actually decent USB wireless adapter working again... any idea what could possibly cause Windows 7 to just flat out refuse to recognize it? Both of my other adapters work fine and without issue, but neither of them are good enough adapters to hold a fast connection at this distance from the router while my TL-WN772n gave me far and away the best connection. It works without incident if I plug it into a friends computer, but my machine refuses to recognize it at all no matter what drivers I've got installed or what port it's in, even though it's supposed to be plug and play apparently. Being as it's the only adapter I have that gives me a stable connection, this has turned into something of a roadblock to getting the rest of my computer back up and running.

    EDIT 2: Well, on the bright side the crash has been solved! Turns out it was the SSD Firmware, as after updating it my computer was able to run in Windows overnight with no issues. No crashes yesterday, overnight, or today. Thanks @bowen ! Gonna take the adapter problem to the tech forum and see what they've got to say about it, but I think we're pretty much done here as the original problem's fixed.

    Screeching Halt EDIT 3: I swear whatever's causing this is sentient and was just waiting for me to say that. After two days of everything working fine, no signs of anything being amiss? It just crashed again, same crash as before which rules out the SSD's firmware. And now we're in the ballpark of things that are going to be very difficult to nail down, because apparently this can sometimes work fine for literally north of 24+ hours straight without crashing, which makes running it on a barebones setup until it crashes extremely difficult.

    Digito on
Sign In or Register to comment.