I'm using a custom init script, so that I can run Displayserver as well as the Selecter program I mentioned in another thread.

The problem is that I just installed a second hard disk, and my init is obviously not correctly mounting that second hard disk because the extra free space is not showing up in emplode.

I can tell from the bootup messages that the second hard disk is correctly being detected at the hardware level. So I'm pretty sure it's my init.

Here is the init, does anyone have any suggestions as to what I should change?


#!/bin/sh
PATH=.:/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/sbin:/empeg/bin
PS1=empeg:\\w\\$
/bin/mount -n -o nocheck,rw /proc
/bin/mount -n -o nocheck,ro /dev/hda4 /drive0
[ -e /proc/ide/hdb ] && /bin/mount -n -o nocheck,ro /dev/hdb4 /drive1
cd /usr/local/displayserver
./displayserver -noauth
while [ 1 ] ; do
echo
echo ----------------------------
echo - RUNNING SELECTER PROGRAM -
echo ----------------------------
echo Q exits, or press a button...
echo
echo - UP - Same as last setting.
echo - DOWN - Normal compression.
echo - RIGHT - High compression.
echo - LEFT - No compression.
echo ----------------------------
if /drive0/var/selecter | grep 'QUIT' ; then
echo Running bash
bash
fi
echo
echo Running Player...
/empeg/bin/player
done




___________
Tony Fabris

Bobo's helping me on ICQ, and simply adding:

/bin/mount -n -o nocheck,ro /dev/hdc4 /drive1

should make it work. But what I'm wondering is what does this line do:

[ -e /proc/ide/hdb ] && /bin/mount -n -o nocheck,ro /dev/hdb4 /drive1

I got this init from RJlov and only altered the ECHOs, so I've got no idea what it's doing at all.

___________
Tony Fabris

well.. it looks like the line is supposed to be a check to see if the drive exists in /proc first, then it mounts the drive, but he may have a more powerfull /bin/sh installed that can actualy do the [ -e /proc/ide/hdb ] check.

if drive entry in kernel exists, then
mount /dev/hdb4


12gig red mk2 -- 080000125

[ -e /proc/ide/hdb ] && /bin/mount -n -o nocheck,ro /dev/hdb4 /drive1

I got this init from RJlov and only altered the ECHOs, so I've got no idea what it's doing at all.

As you probably already figured out, the first drive is /dev/hda4 and second drive is /dev/hdc4. The [ -e ... ] line basically just checks that whether /proc/ide/hdb exists and only in that case tries to mount the drive. You can safely always try to mount the /dev/hdc4, even though it wouldn't exist, it doesn't do more harm than print an error message.

Kim

It could be a Mark1/Mark2 problem, as well. Difference between HDC and HDB?

So it's safe just to remove that testing line and simply mount the sucker? Do I mount it as hdb4 or hdc 4?

___________
Tony Fabris

On the Mk2 hdc is a link to hdb so the same player settings work for both Mk1 and Mk2. That may be changing in 1.1 but I couldn't tell you that for sure. You could just put in a mount for both c and b and which ever one exists will be mounted.

Brian

-See my empeg here-

Yeah, I just replaced the test line with a straight mount of hdc4 and now all is well. Thanks everyone!

___________
Tony Fabris

Some weeks ago I figured this out also. the init from displayserver now uses :

[ -e /proc/ide/hdb ] && /bin/mount -n -o nocheck,ro /dev/hdc4 /drive1



Frank van Gestel

Will that work, since it's testing for hdb? I thought you have to test for hdc.

___________
Tony Fabris

I think the proc file-system stil uses the kernel-based naming - so hdb is correct

To the best of my recollection:

IIRC, hdb is what is actually there to see on a MkII as it has both
disks on the same cable (IDE channel) as master and slave (hda and hdb).

The MkI uses 2 IDE channels and the second disk is master on the
second channel, ie hdc.

To simplify life, hdc exists on the MkII too, but only as a symlink
to hdb.

So, on a MkII, if you want to check for a second disk, you do need to
check for hdb (which is where it physically exists) - but you mount it
as hdc (to keep things compatible with the MkI)

/Michael