[vox-tech] Re: Partition question
Margo Schulter
mschulter at calweb.com
Fri Aug 13 14:56:04 PDT 2004
Jeff Newmiller <jdnewmil at dcn.davis.ca.us> wrote:
> Kernel bloat is one thing, but far more valuable is the ability to
> have multiple kernels to boot from in case something is broken.
An excellent point! I can appreciate it, since I'm planning to try a
kernel configuration supporting 32-bit vesafb graphics. Thanks for a
most helpful reply, which also gives me an opportunity to clarify the
schemes I was discussing (not as sophisticated as some of your
suggestions!).
[On my first possible solution]:
> Last time I tried something like this, fdisk was unable to insert
> partitions. I would be surprised if this limitation had changed.
To clarify, my current DOS primary partition (hda1) and extended
partition (hda2) with its three logical drives (hda5, hda6, hda7)
take up in all 2G of a 37G disk, with the rest unused. This scenario
would involve simply adding two more primary partitions after the
extended partition to fill the rest of the space with a single native
Linux distro (swap plus root):
hda1 primary 0M - 256M 256M fat16 (DOS C:)
hda2 extended 256M - 2048M 1792M
hda5 logical 256M - 512M 256M fat16 (DOS D:)
hda6 logical 512M - 1024M 512M fat16 (DOS E:)
hda7 logical 1024M - 2048M 1024M fat16 (DOS F:)
................................................................
hda3 primary 2048M - 2816M 768M linux-swap
hda4 primary 2816M - 38166M 35350M ext3 (Linux \)
On my third proposal, not well explained in my post:
> I am puzzled, though... this eliminates drives D:, E:, and
> F:... in what way is this superior to having a logical
> partition that DOS cannot access?
I agree <grin>, and should have explained this scenario more clearly.
Here the extended partition would be enlarged with parted enough to
add two more logical drives hda8 and hda9, providing 768M swap
partitions for two native Linux distros, with primary partitions hda3
and hda4 added to serve as root partitions for these distros.
hda1 primary 0M - 256M 256M fat16 (DOS C:)
hda2 extended 256M - 3584M 3328M
hda5 logical 256M - 512M 256M fat16 (DOS D:)
hda6 logical 512M - 1024M 512M fat16 (DOS E:)
hda7 logical 1024M - 2048M 1024M fat16 (DOS F:)
.....................................................................
hda8 logical 2048M - 2816M 768M linux-swap (SW)
hda9 logical 2816M - 3584M 768M linux-swap (Gentoo)
hda3 primary 3584M - 12800M 9216M ext3 (SW /)
hda4 primary 12800M - 38166M 25366M ext3 (Gentoo /)
This keeps the DOS drives D:, E:, and F: -- if DOS will still
recognize them! Here hda2 stays within the 8G limit for chs as
opposed to lba addressing -- if that was the only problem with my
earlier attempt. It might be a question of trial and error.
> Consider:
> hda1 = C:
> hda2 = swap
> hda3 = logical
> hda5 = Slackware root
> hda6 = Gentoo root
> hda7 = /home or /home/shared
> There is even room in here for a partition before the swap
> partition that could serve as a drive D: for DOS.
This is an interesting scenario, and the shared /home idea is
something that I hadn't considered -- and the symlink that you
discuss. There would be two primary DOS partitions C: and D:,
with one of them (the latter?) enlarged enough to hold all the
stuff now in E: and F: with some room for new files -- say
about 1536M (the current size of E: plus F:), plus any overhead.
for larger clusters.
Thank you for such a considered and helpful reply, and for
reminding me that this multi-boot stuff is not necessarily
an optimal arrangement (to say the least).
Most appreciatively,
Margo
mschulter at calweb.com
More information about the vox-tech
mailing list