Will Fedora 7/8 X86_64 run on a Dimension 8400?

[drick]

Originally Posted by philip4567

drick was suggesting using the raid option when you're setting up your disks during installation.

I think the other alternative, if you're not bothered about using raid, is to add tthe nodmraid option when you get to the boot options when booting from the DVD. Choose your boot option and then press tab. Then you can add a space and nodmraid after the boot command. Finally, press enter.

The thing is Linux still sees your raid capability even when you disable it in the bios.

just because i'm a glutton for punishment (well that and the other install was really flaky) i decided to try this anyways and get the 64 bit version running.

no dice, it does the exact same thing when set to RAID/AHCI in the BIOS. i'm going to try it again with the BIOS set to no RAID now

[drick]

Originally Posted by jrummy27

I've seen formatting hangs like this in anaconda when a previous attempt to format did not complete (i.e., your attempt #1 hanging may have caused all other attempts to hang). You may want to boot into rescue mode or the live cd and zero out the disks or wipe out the partition tables.

It may also help to see where formatting is hanging. You can get verbose formatting messages with Ctrl+Alt+F2,3,or 4 (can't remember which).

I've seen many instances in which BIOS raid didn't work properly. I've been told the same thing as bradchaus, that it is controlled by software drivers in windows which don't exist in linux.

John

it's ctr+alt+f4 for the verbose output - here is a subset of the error log:

buffer i/o error on device sr0
squashfs error: sb_bread failed reading block 0x12c4
squashfs error: unable to read page
squashfs error: sb_bread failed reading block 0x12c4
squashfs error: unable to read page
squashfs error: sb_bread failed reading block 0x12c4
squashfs error: unable to read page
mke2fs 7641 segfault at xxxxx rip xxxxx rsp xxxxx error 6

i wish there was some way to cut and paste it out for you guys.

[drick]

Originally Posted by jrummy27

In rescue mode you can use fdisk or parted to wipe out the partition tables. For fdisk, you would do something like

[root@dhcp243-44 ~]# fdisk /dev/sda

The number of cylinders for this disk is set to 4863.
There is nothing wrong with that, but this is larger than 1024,
and could in certain setups cause problems with:
1) software that runs at boot time (e.g., old versions of LILO)
2) booting and partitioning software from other OSs
   (e.g., DOS FDISK, OS/2 FDISK)

Command (m for help): p

Disk /dev/sda: 40.0 GB, 40000000000 bytes
255 heads, 63 sectors/track, 4863 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x000910fd

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1   *           1          13      104391   83  Linux
/dev/sda2              14         268     2048287+  82  Linux swap / Solaris
/dev/sda3             269        4863    36909337+  8e  Linux LVM

Command (m for help): d
Partition number (1-4): 1

Command (m for help): d
Partition number (1-4): 2

Command (m for help): d
Selected partition 3

Command (m for help): w

If you want to zero the disk erasing all data on it, you can use dd

dd if=/dev/zero of=/dev/sda bs=1M count=#

.

thanks for this john, i also gave this a try to see if it would help by chance. i blew away the partitions from sda & sdb as suggested, and reran the installer (without the nodmraid option). the dd cmd didn't work, it didn't seem to like the # sign behind count?

that operation completed successfully, but then the installer hung on something else

[jrummy27]

the dd cmd didn't work, it didn't seem to like the # sign behind count?

Sorry, I should have clarified that you should replace # with an actual number (# of Mb on the disk).

mke2fs 7641 segfault at xxxxx rip xxxxx rsp xxxxx error 6

You should probably file a bug on this

[Zotter]

FWIW - I just did a clean install of F8 (32bit) on a Dimension 8400, 450GB SATA drive. Only one drive.

No issues, hassles or problems.

BIOS set for RAID/AHCI auto detect.

[A.Serbinski]

Disabling or enabling RAID in the bios will do NOTHING. The OS doesn't ask the bios if RAID is enabled. The problem is that once a raid set is created, it writes that fact TO THE DISKS, so even if you disable RAID in the bios, it will still be detected and it will still try to activate it. The solution to this is to WIPE the RAID METADATA from the disks. This is done with the dmraid command, ie "/sbin/dmraid -E /dev/sda". It can also be done by completely obliterating all data on the disks, ie "dd if=/dev/zero of=/dev/sda". It can NOT be done with fdisk or mkfs.

If there is only one disk, then regardless of anything, it will be treated as a single disk, therefore it will work. If there are two disks and RAID is enabled in the bios, it will RECREATE THE RAID METADATA --- and this you do NOT want to happen.


Steps;
1) Disable RAID in the bios.
2) Delete RAID metadata from the disks.
3) Install OS, maybe choose proper software raid if you feel like it.

NOTE: bios raid is fake raid. It *looks* like hardware raid, but its all in the driver, therefore it is software raid. It should ALWAYS be avoided, except in those rare circumstances where the disk must be readable by both lin and win. In those circumstances however, it should NOT be your boot disk.