checkroot.sh always runs

[Gerald Waugh <gwaugh@frontstreetnetworks.com>]

Hi,

Sometime ago I asked about the issue of checkroot.sh (fsck) running on every 
reboot. It is not too much of an issue on a production system as reboots are 
seldom required. But can be a real nuisance at times (30 minutes on a 250GB 
drive).

Peter Horton suggested;
<quote>
 Try enabling 'sulogin', logging in before the root is remounted r/w,
 loading the time from the RTC (hwclock --hctosys --utc) and then
 checking the affected volume (e2fsck -f /dev/hda???). IIRC it'll then
 check on the next boot but should be okay after that ... 
 PS - rather that enabling 'sulogin' you could always pass
 'init=/bin/bash' to the kernel instead.
</quote>
So, I did the following... (IIRC)
Booted into a shell from the LCD startup boot menu and executed
   mount
   load /boot/vmlinux-2.4.27-r5k-cobalt
   execute root=/dev/hda2 console=ttyS0,115200 init=/bin/bash
   hwclock --hctosys --utc
   fsck /dev/hda2
   reboot
But after numerous test reboots, checkroot.sh (fsck) still checks /dev/hda2 
which is a little over 200GB and takes about 30 minutes to run.

BTW this is an original RaQ (1). Oddly I do have one out of several (4) that 
we are currently testing, and it does not respond in this manner. The LCD 
briefly displays "checkroot.sh"

And thanks to all the people making the Cobalt RaQ/Qubes usefull!!!

Any ideas???

Gerald