Siward wrote: > here's my 2c , just a long shot ... > did you consider that it could be a hardware problem, > e.g. in the videocard ? > i heard of thermal defects that can do things like this, > e.g. a cracked solderjoint opening/closing a connection > depending on temperature > if you take your harddisk to another box, > then does problem appear at that box ? > anything unusual about 30-31 oct ? > (all bugs went on strike to protest harassment by debian developers ? :-) I suppose that this could be a hardware problem. Since my hardware is a laptop, I have not tried something like running the hard drive elsewhere and it's more complicated than isolating a potential hardware problem on a desktop. I live in a very thermally varied environment, but there has really been little special about the days it failed vs the days it did not; more significantly it has failed at all times of day, which involves temperature ranges of up +-40 degrees here. For example, on 3 Nov, the 11 am failure would have been around 70 degrees; the 11 pm failure more like 45 degrees. Perhaps I should try a proprietary X server, a different X driver (VESA)? Or install X in a stable chroot. If I can reproduce the problems in any of those, it would be a hardware or kernel problem, right now my money is still on the X server. (FWIW, I can reproduce the problem running X programs such as xterm, from stable, on another computer with the DISPLAY set back to my laptop.) -- see shy jo
Attachment:
signature.asc
Description: Digital signature