----- Forwarded message from dave@owenville.net -----
From: dave@owenville.net
Date: Tue, 10 Aug 2004 00:06:19 -0700
To: Joey Hess <joeyh@debian.org>
Subject: Re: Bug#264715: Package: installation-reports
User-Agent: Internet Messaging Program (IMP) 3.0
Added init=/bin/sh to my kernel parameters (it being grub and all) and started
looking through the log files. Here's an interesting thing:
The last entries of dmesg before I did a reboot via reset button look like this:
Aug 9 22:57:45 localhost /usr/sbin/cron[984]: (CRON) INFO (pidfile fd = 3)
Aug 9 22:57:45 localhost /usr/sbin/cron[985]: (CRON) STARTUP (fork ok)
Aug 9 22:57:45 localhost /usr/sbin/cron[985]: (CRON) INFO (Running @reboot jobs)
So far, so good, since that matches the last entries in my log now, with my
current (emergency) session running. But then we immediately get:
Aug 9 22:57:45 localhost shutdown[998]: shutting down for system reboot
Aug 9 22:57:45 localhost init: Switching to runlevel: 6
Aug 9 22:57:49 localhost kernel: Kernel logging (proc) stopped
Aug 9 22:57:49 localhost kernel: Kernel log daemon terminating.
Aug 9 22:57:49 localhost exiting on signal 15
Time passes, then I get up and hit the reset switch. The next entry is:
Aug 9 23:46:30 localhost syslogd 1.4.1#15: restart.
So it appears to be spontaneously going into reboot mode, although the hardware
isn't up to the task of actually rebooting until I hit the switch. No errors in
the boot messages, no errors in the log files. Seems that the monitor isn't
misconfigured, as I originally suspected -- it's just shutting off as part of
this rebooting that the machine initiates (but that doesn't actually reboot the
hardware). There is a bit of disk activity after all this, but only for a few
seconds. Odd.
If there is a particular file I can send to flesh out this bug report, or an
additional avenue that you would recommend I pursue, please let me know. I'm
going to bed (work in the morning), but I'll pick this up when I get home
tomorrow evening. Thanks again for your help so far, even if you can't help
further. :)
-Dave
Quoting Joey Hess <joeyh@debian.org>:
> dave@owenville.net wrote:
> > 1. Booted, boot messages were going by, right after hardware detection
> the
> > screen went blank. No recovery possible. Manually rebooted.
> >
> > 2. Booted, boot messages were going by, right after hardware detection
> the
> > screen went blank. No recovery possible. Manually rebooted. Vowed to
> remember
> > recovery mode next time. ;)
> >
> > 3. Booted recovery mode, boot messages were going by, very quickly screen
> went
> > blank. No recovery possible. Manually rebooted.
> >
> > 4. Booted into Windows to make sure my other partitions were still okay.
> They
> > were. Are. Whew.
> >
> > I am currently assuming that the issue is my NVidia GeForce MX200
> DVI-output
> > video card and Gateway FPD1500 LCD monitor, which many Linux distributions
> have
> > historically had a problem with (as the monitor reports its' settings
> > incorrectly). XandrOS and Linspire have no difficulties with this video
> > hardware, but Suse 8.2 and Debian woody did (both required manual
> modification
> > of X config to work). I assume that if I were able to boot Debian into
> console
> > mode, I would be able to work around this issue -- sadly, recovery mode
> does not
> > appear to allow this. Would be grateful to learn that I am wrong.
>
> You should be able to boot it into emergency mode ("linux emergency").
> You'll then need to remount root read-write and try to see what is
> causing this. Try starting hotplug and see if that loads a module that
> kills yor display.
>
> --
> see shy jo
>
----- End forwarded message -----
--
see shy jo
Attachment:
signature.asc
Description: Digital signature