[9fans] Looking good...

[9fans] Looking good...

[Steve]

Lucio de Re wrote:
> I managed to test a combination of "mkfs | mkext" to copy part of the
> CDROM to the 280Meg MO disk, and I'd now like to be more selective
> about what I put on there while at the same time using the
> "_conform.map" file to rename files correctly.

What you want to do it boot a cdrom-capable kernel on a plan9 terminal (which
it appears you have already done), and then use mkfs|mkext with a custom
proto file that will copy only the stuff you want.
Look in /lib/proto for the prototype files, in particular /lib/proto/portproto,
copy portproto to /tmp/myproto, edit it to suit your needs,
and use something like the following on your cdrom terminal:

mount -c /srv/il!* /n/kremvax
mount /srv/boot /n/boot
disk/mkfs -a -s /n/boot /tmp/myproto | disk/mkext -u -d /n/kremvax

You don't need to think __conform.map if you mount the cdrom on a plan9
system. That "find" script is for unpacking things under Unix.

[9fans] Looking good...

[Lucio]

>
> You don't need to think __conform.map if you mount the cdrom on a plan9
> system. That "find" script is for unpacking things under Unix.

I guess one needs to read between the lines.  The docs do say that the
software knows how to translate names, I didn't realise that meant that
the bits on the CDROM actually contained the filename in disguise and
that "_conform.map" was just a red herring :-)

Good, now to squeeze the CDROM onto 280Meg...
--
Lucio de Re (lucio@proxima.alt.za)
Disclaimer: I'm working at getting my opinions to agree with me.

[9fans] Looking good...

[Lucio]

I now have a Plan 9 file server (don't laugh, it's a 386DX40 with 8Meg
of RAM and the SONY Magneto-Optical drive - sadly, the 2Gig ST12400N I
nearly got working fell off my desk and is now departed) that works.

I managed to test a combination of "mkfs | mkext" to copy part of the
CDROM to the 280Meg MO disk, and I'd now like to be more selective
about what I put on there while at the same time using the
"_conform.map" file to rename files correctly.

Steve's script to do this needs "find" which is a utility Plan 9 lacks
(I presume for good cause) and I'm at a loss on how to replace its
functionality (I assume I'm not seeing the wood for the trees).
Unfortunately, this puts me in a bit of a quandary: on the one hand I
haven't enough space to use the installation diskette to upload the CD
in its entirety (and get the names translated in the process) and I
don't know how to get the names translated (other than by manually
searching through the destination directory) if I generate the disk by
hand.

I'll not be offended if someone rubs my nose into some obvious
documentation I have not yet read, I thought I'd been through all the
necessary information, but there's always something I overlooked or
seem not to know about.
--
Lucio de Re (lucio@proxima.alt.za)
Disclaimer: I'm working at getting my opinions to agree with me.