AlphaCore 1.0 final release !

[Michal Jaegermann]

On Wed, May 11, 2005 at 11:55:44AM -0400, Steven N. Hirsch wrote:
> 
> Deleting the C slice has fixed it.  The installer is now happy with the 
> drive and sees all the existing partitions.  While this is a reasonable 
> workaround in my case, it may be a problem for those who want to share a 
> drive between Tru64 and Linux.

This is obviously a hacky workaround but once you installed you can
add this "C slice" back.  As you noticed Linux does not care.
Are you sure that a partition type on that was set to "none"?  It
clearly confuses anaconda but maybe "none" would help?

And no, you cannot use "DOS type" partitioning on your boot disk if you
are booting via SRM.  Your firmware would not know what to do with that.
The same principle applies, but the other way around (i.e. you "have to"
instead of "cannot"), if you boot through AlphaBIOS/MILO.  On non-boot
disks a partitioning type does not matter for Linux as long as your
kernel has a support for it.

> I'm not sure whether Tru64 will be upset 
> by the lack of a C slice.

I do not think so (although I am not absolutely sure - it was a while).
This partition is used to provide an access to the whole disk.  That
what Linux is doing reading a writing things like /dev/sda.  You will
loose that capabiity until such partition is added.  On a disklabel you
have eight "slots".  As long as you have one free you can rewrite
disklabel in a desired way without touching anything on file systems.

IIRC old versions of anaconda did not have this problem.  Once Alpha was
off the radar the code in question was clearly "improved".

   Michal