Just because Memtest86 went 30% without failure does not mean your RAM is 30% OK. It should run for at least a full loop, and I typically run it for at least 2 loops (though only once have I seen a system suddenly start spewing errors it didn't have during the first loop, and I've run it on a lot of systems). It does take a while, so you probably don't want to sit around watching it. I usually let it run overnight while I sleep.
----- Original Message ---- From: Peter Davoust <[EMAIL PROTECTED]> To: [email protected] Sent: Monday, May 14, 2007 2:08:42 PM Subject: Re: [gentoo-amd64] Gentoo crashing? Ok, so I compiled a new kernel, and it seemed to work. I booted the new kernel, I was able to unzip and install dhcpcd, ndiswrapper, wireless-tools, and cab extract. Then, once the wireless was working, I tried emerge --search dhcpcd, because gentoo apparently doesn't like my manually configured dhcpcd.... CRASH! I ran memtest86+, as suggested, and it got to at least 30% without a failure. I'll try it again, but at least 30% of my memory is in tact. I'll try to emerge some other things, and see how it goes. -Peter On 5/14/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: On Mon, May 14, 2007 at 02:04:30AM -0400, Peter Davoust wrote: > Ok, first, while I appreciate your advice, this is a brand new laptop > and there's no way I'm running bonnie++ (that's prime95, right?), or > anything with the words "cpu" and "burn" in the same sentence on this > thing. Memtest86 might be an option as long as it has no potential to > kill anything. I agree, it could be the heat, and that was the first > thing that came to my mind, but Vista boots and runs for long periods > of time with no issues. I'll check it out with the new kernel in the > morning and see what it does. Any new laptop should have the hardware smarts not to smoke itself, or something really is broken. It may shut down "unexpectedly" (which I also consider a design bug), but actually causing damage is unlikely. That said, this really sounds like a RAM problem, so I would run memtest86 first. Memtest86 has zero chance of smoking any system that has passed a factory QA check. I had a Gentoo system (a server) that pretty much ran (to be honest, it was a heavily used database server that stayed up for a good 3 months in this state). However, its clock was skewed something like 10m/hour (I now think this was due to lost ticks during processing of memory faults). I tried all the various kernel flags, largemem, etc., only to find out that the problem was (as others on this thread have posted) incompatible RAM. I point this out only to say that bad RAM can cause *very* unusual problems (not just the segfaults you'd expect), and to say that lots of complex operations (like Vista, for example) can continue to run just fine in such a broken environment. Dustin -- [EMAIL PROTECTED] mailing list
