Disk Partitioning Error

[rgthomson]

The snapshot yesterday (KDE Live) still creates a partition scheme with overlaps when "install to hard drive" (/usr/bin/liveinst) is selected.

Here's output from uname -a and fdisk -l:

Linux localhost.localdomain 2.6.29-0.258.2.3.rc8.git2.fc11.i586 #1 SMP Tue Mar 24 18:37:23 EDT 2009 i686 i686 i386 GNU/Linux

Disk /dev/sda: 40.0 GB, 40007761920 bytes
255 heads, 63 sectors/track, 4864 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x0008b390

Device Boot Start End Blocks Id System
/dev/sda1 * 1 26 204800 83 Linux
Partition 1 does not end on cylinder boundary.
/dev/sda2 26 4864 38865248+ 8e Linux LVM

Same problem occurred in the previous spins and with the install DVD, and were consistent whether default partitioning was accepted or the partitioning was manually adjusted. Hardware is a VIA motherboard with 1gb ram and a 40gb IDE drive, prior tests with a Dell / sata drive were identical.

Rich

[SlowJet]

213857280 = 26 cyl

The default boot size is 200MB
200*1024*1024 = 209715200

204800*1024 = 209715200

So make it 8225280 bytes x 3 x 8 = 197406720 bytes, to make it on a Cyl boundary and close to 200MB.

Use the calculator for a different size.

SJ

Make it easier.
th 3x8 = 24 cyl (I thoht the 8225280 was 82MB but it is only 8MB)
The ideal size would be some 8225280 x NN Cyl evenly divisible by 4096.
And sure enough 197406720 bytes / 4096 = 48195 (4k blks)

However, the anaconda installer only allows MB to allocate space and I don't see a factor of NN CYL ~ 200 Mb / 4096.
So use fdisk or parted to pre-allocate the the partition on CYL boundaries if this is causing a problem.

[rgthomson]

It doesn't matter what I do, the partition still has START of the following partition the same as the END of the previous partition. On my example, they are both 26. That's not consistent with all other partitioning results I've seen. Rich

[SlowJet]

That's because the first used part of the 26 and the second is starting in the middle of 26 somewhere.
Is it causing a problem? What problem?

SJ

P. S. I have been trying to find a factor for MB on a Cyl boundary but I don't think is is posible.
You would need to use fdisk or parted and start sda2 on Cyl 27.

SJ

[rgthomson]

All other automatic or manual partitioning I've done does not overlap the cylinders. If FC11 is creating a partition with a portion of a cylinder, that would be out of the ordinary.

Here's output from uname and sfdisk from FC11 and FC5. You're right, it appears that the first partition ends before the end of cylinder 25, and partition 2 starts in the middle of cylinder 25.

Perhaps it is no problem after all, but it is strange and fdisk warns about it.

Linux localhost.localdomain 2.6.29-0.258.2.3.rc8.git2.fc11.i586 #1 SMP Tue Mar 24 18:37:23 EDT 2009 i686 i686 i386 GNU/Linux

Disk /dev/sda: 4864 cylinders, 255 heads, 63 sectors/track
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0

Device Boot Start End #cyls #blocks Id System
/dev/sda1 * 0+ 25- 26- 204800 83 Linux
/dev/sda2 25+ 4863 4839- 38865248+ 8e Linux LVM



Linux 2.6.18-1.2257.fc5 #1 Fri Dec 15 16:04:33 EST 2006 i686 i686 i386 GNU/Linux

Disk /dev/hda: 4864 cylinders, 255 heads, 63 sectors/track
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0

Device Boot Start End #cyls #blocks Id System
/dev/hda1 * 0+ 12 13- 104391 83 Linux
/dev/hda2 13 1387 1375 11044687+ 83 Linux
/dev/hda3 1388 2040 653 5245222+ 83 Linux
/dev/hda4 2041 4863 2823 22675747+ 5 Extended
/dev/hda5 2041+ 2301 261- 2096451 82 Linux swap / Solaris
/dev/hda6 2302+ 2432 131- 1052226 83 Linux
/dev/hda7 2433+ 2433 1- 8001 83 Linux

[rgthomson]

https://bugzilla.redhat.com/show_bug.cgi?id=490922

They are aware of it and do nit consider it a problem.