questions about "Suspend"

Michal Jaegermann michal at harddata.com
Wed Jun 14 22:31:12 UTC 2006


I meant to ask for some time what actually "Suspend" is supposed
to do on the current development setup.

I have in gnome menus two different "Suspend" entries and they do
not behave exactly the same.  The main function of the one directly
in a "Desktop" menu seems to be to kill my video and my keyboard.
If you will risk trying that one then you will get a blank black
screen and not too many options to proceed.  My power button will
generate an ACPI event to do 'shutdown -h' in an orderly manner
and that is about it.

With a network connection it is also posible to login from a remote
but the "minimal" way to restore some sanity seems to be 'telinit 3'
followed by 'telinit 5'.  That works.  With other things I tried to
restore at least some video I can see, from a remote, corresponding
processes running, and nothing seems to be amiss, but my screen and my
keyboard are still dead.

There is another "Suspend" button under "Shut Down...".  That one
actually causes a bunch of text console messages to scroll and
after that I am back to my original session screen but with
a totally blank alert (I think) box in the middle.  If you will let
it sit it will timeout after a while.  If you will try to close it
(there is a "close" button on an otherwise empty title bar) then
you will get another one which says "Application window not
responding", or something similar to that effect, followed by dire
warnings.  If you ignore these then both boxes will close and you
can continue as usual; the same if you will allow this "alert"
to time-out.

With both ways to "Suspend" after such experiment I can find in logs
the roughly the same:


gnome-power-manager: Hibernating computer because the DBUS method Hibernate() was invoked

(this line is different in the second case and it says 

gnome-power-manager: Suspending computer because the DBUS method Suspend() was invoked

with the rest of the stuff beeing mostly the same - no "Shrinking
memory" line)

gpm[2182]: *** info [mice.c(1766)]:
gpm[2182]: imps2: Auto-detected intellimouse PS/2
kernel: end_request: I/O error, dev fd0, sector 0
kernel: Freezing cpus ...
kernel: Stopping tasks: ========================================================================================|
kernel: Shrinking memory... done (47547 pages freed)
kernel: pnp: Device 00:0c disabled.
kernel: pnp: Device 00:08 disabled.
kernel: pci_set_power_state(): 0000:00:00.0: state=3, current state=5
kernel: Extended CMOS year: 20
kernel: Class driver suspend failed for cpu0
kernel: Extended CMOS year: 20
kernel: Could not power down device <NULL>: error -22
kernel: Some devices failed to power down, aborting suspend

... and now we are getting back.

It seems to a bit weird trying to suspend starting from "cpu0",
and I am not sure why to try to power down device <NULL> and why
failure here is an error, but whatever.  Results are what they are.

It still looks to me that to have two different actions with the
same looking "Suspend" description is somewhat, ahem, confusing.

Should I put that into bugzilla?  If yes, then which package?

  Michal




More information about the test mailing list