RE: [cobalt-users] at reboot date is incorrect, apparentlyhwclock --hctosys isn't being executed

["Andy Brown" <andy.brown@xxxxxxxxxxxxx>]

<snip>
> on 4-24-02 5:57 AM, Paul Ferwerda at paul@xxxxxxxxxxxx was 
> reported to have made a statement that said this:
> 
> > I searched the archives but didn't see an answer.
-----
> We have the same issue on our Raq3's. We had a cobalt 
> engineer looking into it, they created a script that would 
> try to fix the drift. It never really worked. He had 
> suggested we change the battery. We did that and one of the 
> boxes started behaving. The next comment from him was not so 
> great, he stated there is a bug on the bios on the raq3 that 
> is part of the problem. Sorry not much help. We just make 
> sure we fix the drift ASAP after reboots.


Would something like this work, add in above the exit 0 line in
/etc/sysconfig/network-scripts/ifup-post the following:

/usr/sbin/ntpdate ntp.demon.co.uk
(change ntp.demon.co.uk to a local timeserver for you)

So the end of the file looks like:

# Notify programs that have requested notification
( cd /var/run/netreport || exit
  for i in * ; do
    [ -f $i ] && \
      kill -SIGIO $i >/dev/null 2>&1 || \
        rm -f $i >/dev/null 2>&1
  done
)
/usr/sbin/ntpdate ntp.demon.co.uk
exit 0


Hopefully that should sync your clock up straight after network
interfaces/routing comes up, so pretty early on in terms of webserving,
etc.

Regards,

Andy
andy@xxxxxxxxxxxxxxxxxxxxx
HOWTO's PKGs and info --> http://www.raqpak.com/ <-- for Qube and Raq