[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Lost interrupt, page allocation failure, and kernel oops




On Mar 17, 2006, at 9:49 PM, Kaspar Fischer wrote:

On 17.03.2006, at 19:35, Michael Schmitz wrote:
Bad RAM, perhaps? Or other hardware dying?


As to RAM, how can I test it? http://www.memtest86.com/ seems
to be for Intel architectures only.

I wish I knew.

I guess this need not be a linux-specific test. Checking the RAM
with TechTool from MacOS X would do the job, too, right? (Dumb
question, I know, but I want to be an the save side...)

TechTool should give the RAM a workout. It may work for disk too. Also, Linux itself is a pretty stiff RAM workout. I've heard it said that a Linux kernel build will find RAM problems faster than any commercial test -- it's just that the diagnostics are less specific as to the exact cause of the failure.

My advice: If you suspect a flakey RAM, try to find something (or a combination of somethings when run simultaneously) that tickles the problem reliably (it may involve running a suite of programs for a few hours before it breaks, but it should reliably break eventually). Then open the box and remove one RAM SIMM and try again. If it still fails, replace the SIMM and remove a different one. Repeat this process until the problem disappears. Once you are certain that you have found the bad SIMM (e.g. put it back in and verify that the problem returns) physically destroy it so that nobody can ever try to use it again! I find that pounding on it with an appropriately sized hammer is effective -- and a good way to relieve the accumulated stress the @#$% thing has caused me.


HTH

Rick



Reply to: