[ltp] Urgent:No boot a tp600x bios update!

Tod Harter linux-thinkpad@www.bm-soft.com
Mon, 8 Apr 2002 09:07:18 -0500


My guess would be that you are running into a problem with getty attaching to 
your virtual console device. Normally a running Linux system with virtual 
consoles will have some process like so running:

root      2275  0.0  0.0  1392    4 tty2     S    Apr01   0:00 /sbin/mingetty 
tty2
root      2276  0.0  0.0  1392    4 tty3     S    Apr01   0:00 /sbin/mingetty 
tty3
root      2277  0.0  0.0  1392    4 tty4     S    Apr01   0:00 /sbin/mingetty 
tty4
root      2278  0.0  0.0  1392    4 tty5     S    Apr01   0:00 /sbin/mingetty 
tty5
root      2279  0.0  0.0  1392    4 tty6     S    Apr01   0:00 /sbin/mingetty 
tty6

(this is from a Mandrake system, its using "mingetty", sometimes systems use 
agetty or just plain getty). Getty just listens on a serial device and 
generally spawns a login process when the device signals a connection is 
established. With virtual consols tty1-ttyn are simulated serial ports, the 
virtual console driver does all the mapping to the physical console and 
translating of keyboard input to what looks like a serial device to getty.

Those "respawning too fast" messages you are getting are getty crashing as 
soon as it tries to talk to the serial console device. init will attempt to 
restart the getty when it sees that it isn't running anymore, but there is a 
"safety" feature in there so that if a process controlled by init is 
restarting too fast it will report an error and not run it anymore, which is 
what you're seeing. 

The reason you can start services by hand is that the system itself is 
working OK. The problems arise because services that should have been started 
by init (which is a whole different kettle of fish from your /etc/rc3.d/... 
type stuff) are not all running.  The file that controls that stuff is 
/etc/inittab, which has lines like so in it:

# Run gettys in standard runlevels
1:2345:respawn:/sbin/mingetty tty1
2:2345:respawn:/sbin/mingetty tty2
3:2345:respawn:/sbin/mingetty tty3
4:2345:respawn:/sbin/mingetty tty4
5:2345:respawn:/sbin/mingetty tty5
6:2345:respawn:/sbin/mingetty tty6

These are the lines in my inittab that start those gettys I referred to 
earlier. Each line is several fields, the 2nd field containing the "2345" in 
this example is saying that init should pay attention to these lines in run 
levels 2-5. "respawn" tells init to rerun these processes if they ever halt.

Generally run level 5 is the same as 3 except xdm is used to give you the X 
login with a line like this:

# Run xdm in runlevel 5
# xdm is now a separate service
x:5:respawn:/etc/X11/prefdm -nodaemon

(it may be somewhat different in Suse, but similar).

The default run level is determined by a line like so:

id:3:initdefault:

in your system this is probably 

id:5:initdefault

You will also see some lines which call the startup shell scripts.

There are also generally 2 other lines, which may be something you can 
experiment with getting rid of, as they are generally useless and might be 
causing trouble:

# When our UPS tells us power has failed, assume we have a few minutes
# of power left.  Schedule a shutdown for 2 minutes from now.
# This does, of course, assume you have powerd installed and your
# UPS connected and working correctly.
pf::powerfail:/sbin/shutdown -f -h +2 "Power Failure; System Shutting Down"

# If power was restored before the shutdown kicked in, cancel it.
pr:12345:powerokwait:/sbin/shutdown -c "Power Restored; Shutdown Cancelled"

Obviously if you have no UPS these lines are irrelevant. You might think APM 
would work similarly to a UPS, but it doesn't, and these lines won't affect 
APM behaviour one way or another. Its remotely possible that somehow init is 
being confused into thinking that you should go to the "pf" state however. 
I'd think you would get a console message in that case, but...

Personally I'd revert my BIOS upgrade, unless Suse tech support has any 
decent advice for you.

On Monday 08 April 2002 06:06, you wrote:
> single user: /sbin/getty 38400 tty2 gives me nothing, no error.
> Alt-F2 is not working and has not been ever since the bios upgrade.
>
> Somethinge esle that I observed is that the name of the host is never
> properly given, it is alsways (none):. Is it a feature of the single
> user boot mode?
>
> On Mon, 8 Apr 2002, Tino Keitel wrote:
> > On Mon, Apr 08, 2002 at 04:05:14 +0200, Boyan wrote:
> > > Hello,
> > >
> > > > > INIT:	 version 2.78 booting
> > > > > INIT:	 Entering run level:5
> > > > > Id "1" respawhing too fast: disabled for 5 minutes
> > > > > Id "2" respawhing too fast: disabled for 5 minutes
> > > > > Id "3" respawhing too fast: disabled for 5 minutes
> > > > > Id "4" respawhing too fast: disabled for 5 minutes
> > > > > Id "5" respawhing too fast: disabled for 5 minutes
> > > > > Id "6" respawhing too fast: disabled for 5 minutes
> > > > > INIT:	no more processes left in this run level
> > > >
> > > > Have you enabled CONFIG_VT in you kernel?
> > >
> > > This is what I do not know for sure, since I use the stock suse
> > > kernel, never compiled one of my own.
> > >
> > > The story gets more involved since today I was actually able to
> > > fully run linux albeit as a single (none) user. Logged in with lilo.
> > > Whenever I try from there init 2 or 3 (console with network) the same
> > > errors come. On the other hand starting the services one by one for
> > > the given runlevel I eventually was able to reach level 5 and from
> > > there started xdm and had the usual kde2. Almost usual, because none
> > > of the configuration utilities work. Any click on Yast or Konsole
> > > gives me:
> > > "DCOP communication error (Konsole)
> > > There was an error setting up inter-process communications for
> > > KDE. The message returned by the system was:
> > > Authentication rejected, reason: None of the authentication protocols
> > > specified are supported and host-based authentication failed. Please,
> > > check that the "dcopserver" program is running."
> > >
> > > Now a check gives /opt/kde2/bin/dcopserver running.
> > >
> > > An attemt to umount with a mouse click the windows partition /dev/hda5
> > > (still mounted by the single user, was needed to write the log
> > > messages during the diagnostic process) gives me:
> > > "Could not start process, can't talk to Klauncher"
> > >
> > > What puzzles me (but may be a rule) is that never had I the option to
> > > change consoles with Alt-F2, F3 or Ctr-Alt-Fn the runlevel regardless.
> > > Only Alt-F1 was functional, the single user login. I logged out and in
> > > this time Gnome and again working as usual, but Yast still fails.
> > > Decided to reboot (not knowing what else can be done) and boot
> > > linux normally and crashed as described in the previous letter.
> >
> > Appart from the KDE problems, there is something wrong with your virtual
> > consoles, but I can't really help you with the information you have
> > posted. What happens if you type '/sbin/getty 38400 tty2' in single user
> > mode and switch to VT2 (Alt-F2)?
> >
> > Tino
> >
> > --
> > tino.keitel@innominate.com
> > dipl.-inf.                        Innominate Security Technologies AG
> > software engineer                                   networking people
> > tel: +49.30.6392-3308                      http://www.innominate.com/
> >
> > ----- The Linux ThinkPad mailing list -----
> > The linux-thinkpad mailing list home page is at:
> > http://www.bm-soft.com/~bm/tp_mailing.html
>
> ----- The Linux ThinkPad mailing list -----
> The linux-thinkpad mailing list home page is at:
> http://www.bm-soft.com/~bm/tp_mailing.html

----- The Linux ThinkPad mailing list -----
The linux-thinkpad mailing list home page is at:
http://www.bm-soft.com/~bm/tp_mailing.html