make release / automated install / install.cfg woes

[Allen]

Is there some extra magic to making install.cfg work when rolling a 
release?  I spent a fair amount of time this weekend preparing a custom 
release based off releng_6_0, and it works great but for one tiny 
detail..

After the installation is 'finished' and I reboot the system without 
the CD, it hangs with the bios complaining that it can't find anything 
to boot, which sounds suspiciously like the MBR wasn't written out 
correctly. The disk part of my install.cfg says:

=====
disk=da0
bootManager=boot
partition=all
diskPartitionEditor

da0s1-1=ufs 524288 /
da0s1-2=swap 2097152 none
da0s1-3=ufs 524288 /var
da0s1-4=ufs 524288 /tmp
da0s1-5=ufs 0 /usr 1
diskLabelEditor
=====

The disk itself is setup absolutely correctly, the partitions are made, 
and the instally goes according to task up to and including the package 
installation... it just doesn't appear to Do The Right Thing when it 
comes to actually making the hdd bootable.

Does this look like it should work?  I've fiddled with various changes 
to the "bootManager" and "partition" lines, so far to no avail.  Am I 
missing something here or should I start trying to figure out if the 
test system will boot off *any* hard drive?

On a related note, when I make a change to this file, can I get away 
with something less than a new full "make release?"  Is make rerelase 
or something else suitable when everything else is done and I'm just 
fiddling with this file?

Thanks for any pointers.  When I'm done with this I think I'm going to 
submit a doc patch for the 'make release' instructions.  They are 
woefully inadequate for doing anything other than making a copy of your 
own release (unautomated) installations, and the example install.cfg 
still dates back to 4.x and jkh's laptop.. heh.