screen not refreshing after wake up from long suspension.

This is suspend to RAM? What hardware do you have?

(In reply to comment #1)
> This is suspend to RAM? What hardware do you have?

Yes, it is suspend to RAM.
Hardware is DELL 700M laptop as follows.

Pentium M 755 -- 2.0G
RAM: 1G
VIDEO: intel 855GME

Interesting. Does it help if you comment out the POST_VIDEO line in
/etc/default/acpi-support?

(In reply to comment #3)
> Interesting. Does it help if you comment out the POST_VIDEO line in
> /etc/default/acpi-support?

I've been using the suspension as much as I can since the POST_VIDEO was
commented out. I haven't met any problems.
But there are updates on the Xorg packages. So I don't know which fixes the bug.

I've tried the uncomment the POST_VIDEO line, it seems OK.

I uncommentted it, because a few days ago, in the first few minutes of wakeup
from suspension,
my mouse cursor suddenly disappeared, shortly after that the screen blank out
except that
I can see the busy mouse cursor every 10ish seconds. Looks like Xserver kept
trying to restart,
but failed each time. Keyboard short cut stop working. Ctrl+Alt+F? couldn't
switch to console.
I have to hold the poweroff button to force a poweroff.

I was then suspected it may be the POST_VIDEO. But jut a few hours ago, this
happened again
after I uncommented POST_VIDEO while I was browsing BBC.

I feel this is a severe bug. Unfortunately I don't know where to look for error
messages. As far
as I known, it's been there since July.

This bug could lead to loss of important unsaved work. Breezy should kindly warn
users of it
unless it got fixed before official release.

OK now I've got more debugging information and hope it's useful.

This time the screen doesn't blank out. But nautilus segmentation fault. When I
log out, gdm crashed with a warning dialog: "The greeter application appears to
be crashing. Attempting to use a different one". And then gdm start with a plain
gnome gdm theme without background image. I start X11 from console using the
following command 'startx -- :1'. Nautilus is still crashing. I've captured the
gdm log, lsmod logs and a screenshot in the following directory. N.B. Going to
recovery mode by issuing ' sudo telinit 1' and back doesn't help.

http://people.pwf.cam.ac.uk/sl392/ubuntu/

By the way, I think this bug may relates to Xserver, kernel and acpi-support.
How can I inform all the related developers?

It seems suspend to ram is getting stabler these days. But once in a while X
crashes.
I found 'sudo /etc/init.d/dbus restart' solves the problem always.

I just wanted to confirm that I have this bug with the missing redraw/refresh of the screen also.

I got a IBM x40, output from `lspci`:

    00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)

I noticed too that the possibility for the bug to occur is much bigger the longer you wait until you resume. I have tried to suspend and resume many times in a row... nothing happens. But often when I suspend at work when I resume 15min or 2 hours later this happens.

I good example is when I resume my laptop the only thing I can see is the text field for my password, the rest is just black. When I enter it I can see my desktop, but the background of the icons from gnome-power-manager and NetworkManager in my Notification Area is now black instead of the same color as the panel.

If I try scroll on a web page in firefox I can see some flickering in the top and bottom of the visible part of the homepage within the browser, but mostly the page appears to not scroll (only the flickering and the scrollbar actually moving indicates that I'm trying to scroll).

When I choose to quit to restart my laptop (so far the only way I found to get it all working again), I can't see any of the buttons in the "Quit..."-dialog. When I hover the mouse over Suspend, Hibernate or Restart, the buttons will appear (but nothing else). When I use tab to jump between the selected button I can make all the buttons appear. Don't know why the buttons redraw differently.

I'll give POST_VIDEO a spin... and I'll be back if it doesn't work ;)

Is POST_VIDEO considered a fix or just a workaround until the bug is actually fixed?

Is there anything else I can do?

I found that I can reproduce this behavior every time by addind "vga=791" to the kernel boot parameters.
As soon as I remove this it seems to work just fine every time.

Thank you for reporting this issue.

Breezy is in End of Life status. Please update and repost detailed error report.

Closing because of no response.