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

Bug#286213: reopen 286213



I cannot exactly reproduce this problem, but I am suffering from (what
seems to be) the same problem: X freezes, *except* *for* *the* *mouse*
*pointer*; keyboard and mouse buttons are completely inoperative, but
all other processes seem to be running normally.  In my case this can
happen at any moment; in any application, including the screen saver;
and frequently, sometimes within 15 minutes after boot.  And the
problem is present in an up-to-date 'Ubuntu 6.06 LTS', kernel
2.6.15-27-386, as well as in Debian etch, kernel 2.6.17-2-486.

I did some web research on this problem, and found two suspected but
unproven causes.  Because of the frequent reboots I don't have any
bookmarks, unfortunately.  The first suspect were ntpd and related
utilities.  The X system was supposed to be unable with negative time
adjustments while it is running, suggesting an unsigned int being used
for the difference between 'now' and 'then'.  In my case the problem
persists even with these programs disabled.  And negative adjustments
should not happen that often, if at all.

The second suspect was that someone observed (using strace, iirc) that
X was flooded with SIGALRM signals.  The reason for this flood was
unknown.

This gave me reason to think. I'm running on what calls itself an
'Athlon XP (0.18) 1261MHz' CPU on an 'ASUS A7V333 rev 1.02'
motherboard, which consideres itself running on the border of
overheating, according to the sensors utility.  And, if my impressions
are correct, the freezes are more likely after more activity, when the
CPU is more likely to be hot.

The next thing I have to research is if X is somehow trying to
terminate because it received some hardware(/BIOS?/kernel?) signal,
and having that termination hit by the same signal again.

What I would like to know is how the mouse pointer escapes the
freeze. :-)





Reply to: