Re: 1.3.94 Ooops For Sale...

B. Galliart (bgallia@luc.edu)
Sat, 27 Apr 1996 03:47:57 -0500 (CDT)


> Simon Shapiro wrote:
> > Few more tidbits;
> >
> > master.debian.org went into a coma less than 12 hours after booting=
1.3.94.
> > What happened? No clue. the only symptoms (differentiating coma f=
rom
> > death):
> >
> > 1. ``Ouch, kerneld wanted to sleep in interrupt'' every second or =
two
> > 2. Login freezes after taking the login name on the console.
> > 3. Existing telnet session totally frozen.
> > 4. ``Cannot load interpreter'' in response to login - sometimes.
> >
> > This was preceeded by the same unholy stream of prophanity as descr=
ibed in
> > my previous posting under this title (``route to %p was born dead''=
).

On Thu, 25 Apr 1996, Linus Torvalds wrote:
>
> Ok, "kerneld" goes disabled in the next kernel, and won't be resuscit=
ated until
> after 2.0 is out unless somebody really starts to look into this thin=
g. I'm not
> using kerneld personally, and for a few reasons I don't think I _will=
_ be using
> it in the near future, so I won't be fixing this.
>
> Bj=F6rn?
>
> Linus

How can the problem with kerneld be recreated?

I'm not noticing a problem with kerneld yet. I'm using both the follow=
ing:

kernel 1.3.94
gcc 2.6.3 (aout)
binutil 2.5.2.6 (aout)
modules 1.3.69f

- and -

kernel 1.3.91
gcc 2.7.0 (ELF)
binutil 2.5.2 (ELF)
modules 1.3.69f

I have been playing with the following modules without a problem (being
automatically loaded/unloaded via kerneld):

floopy.o, loop.o, fat.o, msdos.o, isofs.o, minix.o, ncpfs.o, ftape.o,
ipx.o, lp.o, psaux.o, serial.o, sound.o, slhc.o, slip.o, ppp.o

Both systems have been on for longer than 12 hours and the ELF based
system with the 1.3.91 kernel has an uptime of 8 days.

If your really want to see the kernel freeze up tho, try using umsdos w=
ith
the greasy weasel. :)

- Ben Galliart (bgallia@luc.edu)