Odd machine behaviour

Amichai Rotman amichai at iglu.org.il
Sat Dec 1 12:06:45 UTC 2012

Try to SSH to the machine from a console on the remote box. That will tell
you if the flaky machine is still alive and it's actually a PS/2 problem.

You can also find out about the the motherboard and BIOS by entering at the

sudo dmidecode -t bios

Then you can look for the key combination online and enter it repeatedly
right after starting the computer.

You can also troubleshoot the hibernation issue by following the following


Note the link to the official documentation at the beging oif the answer.

Good luck!

On Sat, Dec 1, 2012 at 11:09 AM, Colin Law <clanlaw at googlemail.com> wrote:

> On 1 December 2012 00:12, Phil Dobbin <bukowskiscat at gmail.com> wrote:
> > On 11/30/2012 10:26 AM, Liam Proven wrote:
> >
> >> On 30 November 2012 01:15, Phil Dobbin <bukowskiscat at gmail.com> wrote:
> >>> I have a AMD Athlon 2GHz box with 2 GB's of RAM running Debian Wheezy
> >>> with all the latest patches pulled in from the repos. It runs fine
> >>> (after the initial glitches with Gnome 3 which settled down about a
> >>> month ago) but it has a nasty habit of just "dying". By dying I mean,
> >>> I'll be logged into it from another machine using Byobu with tmux as
> the
> >>> backend & I'll get a broken pipe error message & sure enough, when I
> >>> cross the other side of the room, the screens gone blank (it's running
> X
> >>> Windows) & the keyboard & mouse are completely unresponsive.
> >>>
> >>> I can see that the mouse still has life (it's an optical one) & with it
> >>> being on a KVM switch, the light still shows activity on the switch &
> >>> the computer tower itself has a green LED showing it's running & if I
> >>> strain, I can still hear it running.
> That just tells you the fan is running.
> >>
> >> This is a bit beyond my level of expertise, but a couple of ideas:
> >>
> >> * attach an actual physical keyboard to the flakey machine, and then
> >> either or both of:
> >>
> >> * enable X-clobbering and see if Ctrl-Alt-Backspace terminates X.org
> >> and drops you to a console, in which case, you know it's still running
> >> & it might be a graphics-related issue
> >>
> >> * try the Linux system-reset key combination - if it works, that will
> >> tell you if the kernel is still responsive
> >>
> >> If the keyboard is USB then you can have multiple ones. I suggest a
> >> local connection in case the issue is KVM-related; indeed, IIWY & if
> >> possible, I'd remove the KVM for testing purposes.
> >
> >
> > I've already had her off the KVM: no difference (the reason she's on a
> > KVM in the first place is where she is situated is my workbench for new
> > machine installs & maintenance; it's easier than unplugging the VGA plug
> > on this particular monitor. I think it was designed by a sadist).
> >
> > So she actually has a keyboard & mouse attached (PS2) & a minimal Gnome
> > desktop running. Unfortunately, the USB ports have been removed from the
> > machine as they were no longer working so it's just a PS2 interface. I
> > remotely log in 'cause I'm lazy; saves me walking across the room.
> >
> > Strangely enough, whatever key combination I use, I can't access the
> > BIOS (&, believe me, I've tried 'em all). Also she boots like a rocket
> > so there's virtually no BIOS boot screen.
> >
> > No key combinations work at all. First thing I did was try dropping into
> > a console. No luck. Keyboard completely unresponsive but as I've said
> > whether on KVM or off, the optical mouse is still lit up & I can hear
> > the machine running (also there's a message on reboot saying filesystem
> > was not cleanly unmounted when I pulled the plug so she was still live).
> No, that just tells you it stoped without cleanly unmounting.  The
> processor might have stopped or crashed any time before you unplugged it.
> >
> > This is the message that spewed forth just before last she "died":
> >
> > 181 Nov 22 18:16:48 wheezy kernel: [    0.939731 ] PM: Hibernation image
> > not present or could not be loaded.
> The 0.939... here is the number of seconds since starting to boot, so
> this message is from the reboot after the failure, not from the
> failure itself.  It is checking to see whether it is resuming from
> hibernation.  Look further back in the log to the start of the kernel
> messages that show [0.000] and another few messages before that is the
> start of the reboot.  You should see a jump in the actual time there.
> To make it easier if you are doing it again then leave it switched off
> for ten minutes before restart so there is a definite gap.  Then post
> the last couple of dozen messages there.
> Colin
> --
> ubuntu-users mailing list
> ubuntu-users at lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/ubuntu-users/attachments/20121201/e29851a4/attachment.html>

More information about the ubuntu-users mailing list