Re: hang trying to sync hardware clock ...

[Adam Shand <larry@alaska.net>]

> It is very unlikely, but... a typo in the hwclock script could prevent
> the next message from appearing, yet the script is still completing, in
> which case the problem is in the script(s) being executed after
> "hwclock".  If the above yields nothing then check /etc/rcS.d and look
> at the script that come after Sxxhwclock when you list that dir.
> 
> Whatever the outcome of this experiment is, you should report the
> problem to the debian-boot mailing list.

well it's certainly something with the hwclock.sh script ... i booted into
emergency mode off the rescue cd, and tried to run the hwclock script and
got the same hang running:

/etc/init.d# ./hwclock.sh show

locked the box up completely.  that should not happen.  so i booted again
and moved hwclock.sh to hwclock.sh-broken and it booted up properly.

i'll play a bit more and then report it to someone once i have an idea what
the problem actually is.

> The new ext2 features means you can't mount partitions created by the
> install when running a 2.0.x kernel, not a problem unless you try to
> downgrade to a 2.0.x kernel later on or still have slink running and try
> to mount "potato" partitions.  The size of the HD should not make any
> difference.

i know but when faced with the bizarre start eliminating the unknown and
unfamiliar.

> I don't recall seeing any report of this problem go by on debian-boot,
> but you would have to check on that list (archives or ask) or with the
> bug tracking system (it should be under boot-floppies because "dpkg -S
> /etc/init.d/hwclock" doesn't return any package name).

i'll check it out, thanks for the reply.

adam.