<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Dec 17, 2015 at 9:06 PM, Mike Hodson <span dir="ltr"><<a href="mailto:mystica@gmail.com" target="_blank">mystica@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
And if youve tried memtest86 for maybe 1 pass... errors in ram can<br>
sometimes elude detection for many many passes, yet still cause daily<br>
problems. Literally it took 17 hours for my system to show even 1<br>
error. another 3 hours to show the second. Of note, trying to compile<br>
GCC caused the compile to fail in many differing points of the<br>
compile, nearly always internal compiler exceptions; a few times the<br>
system hung.<br></blockquote><div><br></div><div>I had a RAM issue with my old Athlon system way way back. I had started with 256 MB of RAM, and had slowly added and replaced until I ended up with 768 MB. It was a lot back then :)</div><div><br></div><div>The various RAM modules didn't play nicely together. The system ran rock stable for three weeks at a time (I rarely rebooted), and then would suddenly start having issues with memory corruption (applications crashing, etc). I had to power off, then back on and I was good for another three weeks.</div><div><br></div><div>It took a while to finally figure that out though. Memtest86 would run clean if it was run before the three week mark. After the three weeks, when the corruption problems appeared, running Memtest86 would show all sorts of RAM issues. I took the RAM in to the computer shop and (of course) it tested clean in their rig....</div><div><br></div><div>aaron</div></div></div></div>