I'm running the latest Ubuntu on my X61 tablet, and I have a strange
suspend/resume problem.  Most of the time, suspend/resume works fine
-- closing the lid puts the thing to sleep, and opening it causes it
to resume to the previous state.

However, every morning the following happens:

   *  I resume it in it's ultrabase, disable the screensaver lock, and
eject it from the ultrabase (using the button and level, as directed)
   * Close the lid, let it get sleeping (waiting for the green
moon-crescent to lite up)
   * stow it in a sleeve in my messenger bag
   * travel 15 min by bus to work
   * pull the tablet out, and open the lid.

Now, without fail, the tablet will not wake up fully.  The bezel
lights flash around as they normally do when waking, finally settling
on the battery and the power lights
(http://www.thinkwiki.org/wiki/Image:PowerLED.png).  Normally, it
seems that the wifi, bluetooth, power, and battery lights turn on.

The screen, however, never wakes up, and the computer appears
completely unresponsive.  I've only found two things that will make it
do anything:  plugging it back into an ultrabase causes it to reboot
immediately, and holding the power button for a few seconds will make
it turn off.  This has happened once or twice under other
circumstances, but it's extremely rare aside from this trip *to* work
-- taking the tablet home involves almost exactly the same sequence of
events, but has never caused this problem.

Normally, I would assume that the screen is simply not waking up,
however, I happen to have a script that runs and writes a timestamp to
a file in my home directory every time the laptop is resumed, and that
is not running when I resume the machine -- so it's (probably)
something more nefarious than just a display issue.

I've also not been able to find anything of interest in the
/var/log/messages, syslog or kern.log -- everything looks normal right
up until the "oh, we're starting!" messages begin.  I don't actually
see any indication of the system going to sleep in there, though.

The only environmental difference I know of is that there is no wifi
at work (so I've tried disabling networking before leaving home, and
I've tried letting the system sit at work with it's black screen to
see if it's waiting for a network time out.  Nothing happens until the
battery dies.)

Since it may be relevant:

   * I'm not using gnome / kde / xfce -- I run Enlightenment DR17, and
invoke the gnome-power-manager to handle suspend/resume (it runs in
the background, and is started by my .xsession)
   * This happens with both the 4-cell and 8-cell batteries.
   * I just went through 6 suspend/resume cycles here at work (after
doing the obligatory reboot when I got here) and everything worked
fine, although gnome-power-manager stopped responding and wouldn't put
it to sleep a 7th time until I killed it and started that program back
up.  The 8th suspend-resume worked fine too.  I rarely put the machine
to sleep more than 4 times per day, for reference.
   * I have an ultrabase at home and at work, and when at those
locations, it spends most of the time in the ultrabases. (which are
connected to the wall for power.)

I would *really* like some suggestions on how I can debug this
problem.  It's driving me nuts, since I loose all my state every time
I go to work, and the only way I've been able to reproduce the problem
is by traveling to work.  (I've spent a lot of time doing
suspend/resume cycles to fix a different bug -- relating to the stylus
calibration, and this only happened a few times out of the hundreds of
cycles I did then.)


Thanks!
Rogan
_______________________________________________
PLUG mailing list
[email protected]
http://lists.pdxlinux.org/mailman/listinfo/plug

Reply via email to