Linux - NewbieThis Linux forum is for members that are new to Linux.
Just starting out and have a question?
If it is not in the man pages or the how-to's this is the place!
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I was hoping to install Fedora 15 x64 on my desktop on a seperate disk from Windows. I have tried about 15 times so far and keep getting errors. The same type of errors occur when trying Ubuntu 11.04 x64.
This is the spec of my PC:
Core i7 920 oc'ed to 4.00ghz (stable in Windows with temps under 60 degrees celcius at load)
Asus Rampage II Extreme
Sapphire Radeon 4890
6gb Ram
Install disk is an empty WD Caviar Black 1TB
(Also, not sure if this is relevant, two soundcards - M-Audio Audiophile & Asus Xonar)
This is the error (I can't get it all written down because it reboots too quickly, so I've written down a selection):
And lower down the error: No human readable MCE decoding support on this processor type (Googling this hasn't helped much, though there are a lot of results)
I suspect this is due to overheating because when I boot back into windows the cpu temp has risen to 80+.
Is there going to be any way to get Linux on this machine without frying the processor?
Thanks for your suggestion, but I'm not actually sure that the overclock is the problem which is why I came here. It could be that it is getting confused by the reported clock speed and actual clock speed being different, or it could be that for some reason the CPU fan stops running during the install which causes the kernel panic, I don't know, I'm just hypothesising. I was wondering if someone could help me work out exactly what the issue is.
It could be that it is getting confused by the reported clock speed and actual clock speed being different, or it could be that for some reason the CPU fan stops running during the install which causes the kernel panic, I don't know, I'm just hypothesising.
So... In the first case you just gave, the overclocking is causing the problem, no? It's probably best if you follow Wim Sturkenboom's suggestion (ie. disabling overclocking, and see if you can install it like that) unless you have compelling evidence that it's not the overclocking causing the problem.
However. you mentioned "for some reason the CPU fan stops running" - is this one of your hypotheses, or does it definitely happen? Either way, there might be some information somewhere in /var/log that might help... Certainly, I grepped through mine and found a couple of interesting things:
Code:
[joshua:~/test]$ grep CPU /var/log -ir 2>/dev/null | egrep -v packages
/var/log/setup/setup.setconsolefont:egrep -q "SheevaPlug" /proc/cpuinfo > /dev/null 2>&1 && exit
/var/log/dmesg:CPU: Physical Processor ID: 0
/var/log/dmesg:CPU: Processor Core ID: 0
/var/log/dmesg:mce: CPU supports 6 MCE banks
/var/log/dmesg:CPU0: Thermal monitoring enabled (TM2)/var/log/dmesg:CPU0: Intel(R) Core(TM)2 Duo CPU E7600 @ 3.06GHz stepping 0a
/var/log/dmesg:Brought up 2 CPUs
/var/log/dmesg:HPET: 4 timers in total, 0 timers will be used for per-cpu timer
/var/log/dmesg:cpuidle: using governor ladder
/var/log/dmesg:cpuidle: using governor menu
/var/log/dmesg:ACPI: SSDT 00000000cfee8100 0022A (v01 PmRef Cpu0Ist 00003000 INTL 20040311)
/var/log/dmesg:ACPI: SSDT 00000000cfee85c0 00152 (v01 PmRef Cpu1Ist 00003000 INTL 20040311)
/var/log/nvidia-installer.log: CPU0: Temperature above threshold, cpu clock throttled (total events = 1)/var/log/nvidia-installer.log: CPU0: Temperature/speed normal
I just highlighted a couple of the lines which looked vaguely interesting
it is not like the 4.00ghz will make an install all that much faster
anaconda will be only using 1 cpu ( single threaded ) on that i7
drop it back down to the ?? what 1.9 ghz that was stock
the installer might not know what to do with the fan but after install the OS should
Thanks for all your suggestions. I see how going back to stock might fix it, the trouble is it took me quite a while to oc it and get it stable at that speed so I want to do that as a last resort - I'm a bit scared of messing it up.
Snark1994 - do you have any idea how I could access the logs if the installer has a fatal error? Would using the nokill boot option work so that I have more than 30 seconds before the machine reboots after the error? Or do I have to set up a network location for logs to be saved like it says in the documentation? Can you drop to a shell after a kernel panic during install?
JohnVV - if the installer doesn't know but the OS could I be load some updated modules during install to mitigate the problem?
Also does anyone know of any way to throttle the cpu during install so the temp doesn't rise so dramatically?
Based upon your error code, it really looks to me like your over clocking IS the problem. See this link. From the article:
Quote:
The error usually occurs due to failure or overstressing of hardware components where the error cannot be more specifically identified with a different error message. .... Normal causes for MCE errors include overheating and/or incorrect hardware installation. Some specific manually induced causes could include overclocking (naturally increases heat output)
That's really helpful thanks, I didn't notice that when I read the article before. I thought it might be bad for the processor to be oc'ed for 18 months then suddenly revert back but according to Google it's ok.
Nonethless I'd love to know if anyone can think of a way to throttle the processor during install of Fedora. I've had the same problem in the past with laptops (not oc'ed) getting way too hot while installing Fedora, it just didn't result in a kernel panic. I've been able to find out how to do it with an installed system but not with Anaconda parameters. Any ideas?
Most installers offer a backdoor at least to terminal. If you must, it may be possible to set cpu speeds there then install. Might be still possible to use boot time kernel options.
Otherwise I guess you could use a live cd to install either some image or make it as you go.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.