|
Computer freezes randomly
It doesnt matter what distro i use it still freezes completely and no input works
freeze happens low cpu/ram and high cpu/ram logs doesnt show anything during the time it freezes its not hdd cause ive used both ssd and hdd still freezes I tried torture testing my ram and cpu nothing frequency is completely random from 1h to 24h+ today when it froze it popped up this(only way i could save this is to take picture with my phone cause computer was frozen with this screen http://www.upload.ee/image/5193163/imgo__1_.jpg http://www.upload.ee/image/5193166/imgo.jpg |
That looks like a motherboard or CPU problem to me, and if I had to guess the cause I would say heat.
Check the cpu cooling fins, are they tight? Is the cooling paste still pasty, or is it hard and brittle? It will not hurt to pull the cooling off, clean everything, apply new paste, and reseat and tighten the cooling head. While in there, when was the last time you cleaned in there. Get the dust bunnies out and check the airflow. Have you a fan down? Are the vents clear, or have you blocked some? Once you do your best to eliminate heat, opt for a total burn-in diagnostic run. It takes hours, but a lot less gets past the longer testing. You do not mention what hardware you have, or the vendor(s) involved. Some vendors have hardware specific diagnostic and testing software for their products that may help. When did you last flash/update your firmware? Does your vendor have a firmware update for your product? If so, it may be one that detects or even solves your problem. It is possible your machine is just going to fail, but using these techniques we have pulled many machines back from the brink! |
Updates are available for BIOS gotta see how to install them
The freeze messages on pictures are from virtualbox i can reproduce the freeze just booting on virtualbox and in a middle of a boot computer freezes and i get the error log pictures above.Does any1 know what is causing this? I opened my computer case and it wasnt that dusty cleaned it and left case open. |
If you can replicate the behavior under Virtualbox, then it is not a real hardware problem. You cannot make a virtual machine have a heat issue without the host having a heat issue.
I have never seen anything that could cause different distributions to have random appearing freeze issues under virtualbox. Providing the wrong hardware or inadequate resource comes to mind as a possible. A corrupted virtualbox install is the only other thing that comes to mind. Hit my limit. Seems very odd, and not like anything I have encountered. BTW: I would not leave the case open for long. The case closed, if well designed, channels the air flow past specific potential hot spots. |
Htop might reveal something causing the freeze if you leave it open and showing.
But being intermittent. Sounds like a pain to trouble shoot if dmesg | tail is not not saying anything in /var/log/dmesg.1.tgz through like dmesg.4.tgz after they are extracted and looked at somewhere else like ~/Documents/dmesg/copied dmesg.tgz sitting here. |
Well seems like im getting closer to the cause
today when i wanted to start virtualbox to check if it freezes again well it didnt booted up on virtualbox everything was fine untill i opened up chromium rebooted and started again only virtualbox everything works fine but whenever i start chromium i get freeze immediately error on screen says fatal exception or something like that funny thing is that ive installed chromium on every single linux distribution ive used so that might be the case?but why does it only freezes on my computer |
im now using firefox and everything seems okay virtualbox running fine using nearly 2k ram so it has to be chromium whats causing this:(
it only happens when virtualbox and chromium are running together so im not entirely sure that chromium causing random freezes aswell edit:also at the top of the error message theres something about failed sync ,fatal exception |
It is occurring as the CPU does some memory operations, flushing pages out of the TLB. Have no idea why though. And the memory operations could technically be because of the panic.
|
If my machine panics, it's normally out of RAM and can't find SWAP. Which happens mostly when you run an app for a duration that's bad at memory management. Routinely ending that bad app (like a browser) and restarting the app will free up most of that mess and start fresh. One way to avoid damnation while still playing with fire. After hard crashes like that you should fsck your filesystem. This is sometimes automated in some distros based on the configuration of /etc/fstab, but that's less and less the case these days. Otherwise you might need to wipe out the chrome/chromium history to get it to work again.
$ mv ~/.config/chromium ~/.config/BAD_chromium or $ rm -rf ~/.config/chromium Running debian tends to avoid this kind of behavior most of the time. Since the lengthy vetting process of X number of days without "new" bug reports keeps most of the Junk away from the hands of users. |
Quote:
|
Hmmm.
Not at all random, just not very predictable. (different thing)
How much memory do you have, and how much swap are you giving it? |
Quote:
i rarely get over 2gb ram usage so im sure thats not the issue im monitoring my ram all the time |
Just so I understand the problem - the freezes happen both when you run virtualbox and also on occasion when you are just running your linux distro. The freezes also occur across several different distros you've tried.
As has been requested once before, post your hardware configuration(ram, motherboard, cpu, power supply). Certain hardware vendors are of known low quality and have a penchant for producing intermittent lockup problems. Quote:
In my experience, intermittent problems like you are reporting are most likely to be caused by either a problem with your ram or a problem with your power supply. Further down the list of suspects would be a bad motherboard, typically caused by faulty capacitors. You can check your board to see if you find any bulging capacitors. Memtest should pick up any ram problems. Unless you have specialized equipment, the only way to test for a problem power supply is to swap in a know good one and see if the problem goes away. |
If you have another machine, you might setup ssh on that one. It sounds like a typical X crash, which is mostly caused by bad video card drivers. The HID devices stop responding but the machine is still working in cases like that. Which means that you can ssh into it and kill X which might bring it back to a usable state. If that is indeed the case, then video drivers or X version changes might be a good starting point. Although if you have bad hardware, there's not much you can do software wise to fix that.
If it's hardware, and it seems to happen near a cold boot, then you might have mismatched parts. Like a PSU that isn't enough W's to drive your original system plus an after market GPU. PSU's also tend to age poorly, so if it's a couple years old, a starting point. I always seemed to need to replace the PSU (power supply unit) once every 3-ish years on my older desktop. I always went with one +100W more than what my system actually needed, but I also cheaped out since I knew I'd eventually have to replace it again. |
Could this be display only?
IF it IS just an x-freeze and not a system freeze, you might try Ctrl_Alt-Backspace (which should kill X). This does not always work if X is frozen, and can take a minute or two to work.
Another option is to try ctrl-Alt-F2 to open terminal 2, log in and kill from there. This assumes that the normal and original multiple console terminal setup is in place. It is possible to make a distro that does not support this, but such are rare. This is the original multiterminal support designed in by Linus himself, so it is unlikley to go away any time soon. The ssh trick already mentioned above is a good one, but these options MAY (should) work even before you change anything! Frankly, I work in GUI so little that it had not occurred to me that you could be talking about a problem with X-Windows: my bad. I would ask about your video circuit, monitor, and settings but if something is misbehaving and not acting as designed then that may not matter, I do hope that your problem IS just X-Windows, as that would mean that the basic OS underneath should be fine and still running normally. It would explain a bit about this happening even if the Linux instance is running under VirtualBox, if the host hardware or driver triggers the crash condition as a result of some perfectly normal call that should not cause a problem: but does due to unexpected hardware behavior. We will see. |
| All times are GMT -5. The time now is 11:35 PM. |