Install failure

Wapocalypse · 11-29-2002, 04:07 PM

Hi.

I'm trying to install Redhat 8, and I've made the CDs and tried to boot them. It starts to load the kernel but gets to something like the following (I can't remember exactly but I'm sure you know what I mean)....

Partition Check:
dev/hda hda1 hda5
dev/hdb

...and freezes. I've tried messing about with BIOS settings, eg. turning off UDMA, and now I've deleted all the partitions on it and am reformatting.

It used to work OK under the original RH7.3 that I downloaded. However, when I tried a newer kernel on that, I had exactly the same problem. I lost the CDs for 7.3 so I got 8.0 instead.

What can I do? Searching on Google I found at least 3 people with this issue, but no solutions.

It's my second hard drive, primary slave, and it did have Linux partitions on it. The first drive is primary master and is made up of several partitions, all NTFS with Windows 2000 on hda1.

Thanks,

Rob Pridham

ikw38 · 11-29-2002, 04:18 PM

Since this a second drive I am unsure if this will work. I have experience such strange things trying to install Vendow$ 2000. The only way I was EVER to get it installed was to do LOW LEVEL or to ZERO out the drive then it went on fine. I am unsure at this point if this is the cause but might worth the time and trouble to do. Something about the Low Level partition the OS does not like? Also in CMOS Turn OFF Plug & Play OS this can cause problems as well.
You can get the tools neede to do this at most MFG sights, if not find a copy of Disk Manger 2000 ver 4 and it will do it.

Wapocalypse · 11-29-2002, 05:04 PM

I don't think a LLF will help, but as a last resort I guess I could try. PnP OS is turned off. I tried rearranging their positions in the system but that got me nowhere.

moses · 11-29-2002, 06:38 PM

Does this happen when you boot from CD? There is some information
output before the partition check. Possibly there is something like:

Code:

Kernel command line: auto BOOT_IMAGE=Linux.2.4.19.4 ro root=305 hdd=ide-scsi ide
bus=66 ide0=dma ide1=dma video=vesa:ywrap,pmipal
ide_setup: hdd=ide-scsi
ide_setup: idebus=66
ide_setup: ide0=dma
ide_setup: ide1=dma
Initializing CPU#0

Does the kernel setup the IDE bus to have DMA (or anything else)
turned on?

Wapocalypse · 11-29-2002, 08:39 PM

I don't know about that exactly, but it indicated that UDMA66 was in use when a) it's switched off in the BIOS and b) the drive in question is not on an 80 pin cable (this is how it must stay, before anyone says 'move it').

Please help or I'm abandoning this whole idea and going straight back to Windows - for all the great things the geeks keep telling me everyday, it doesn't bloody work

And I did a LLF; nothing new.

ikw38 · 11-29-2002, 09:19 PM

Correct me if I am wrong but this sounds to bad HD issue- AKA bad sectors?
It makes no sense to me a hard drive freezes in the middle of an install unless we have Hardware issue.Some eons ago I had one do as you have describe when installing Vendows$ 3.11 and that turned out to be a controller on the MFM hd I had.
The only other secnerio I can offer I have Older Western Digital dives be incompatible with other HD Because it wanted to be the MASTER. But I have not seen this in several years now --3-4 at least. Is the second drive bigger than the first? Other than this I would try this as an experiment
take your Master Drive off and Install Linux on the second acting as a Master. Once installed set it as a slave again, then use a Program such as Boot Magic to setup a boot program for you. This is a little work but the rewards are great.

Linux Rocks...

moses · 11-29-2002, 09:35 PM

Ok, when you boot with the CD, I think it gives you a prompt for a few
seconds before loading Linux (I don't use RH), possibly telling you that
you can hit some key to modify the boot parameters. You need to hit
that key (often if you hit the left shift key, it'll work). At the prompt,
input the following:
ide0=nodma

Wapocalypse · 11-30-2002, 04:13 AM

Thanks. Does the 0 represent the primary bus?

Wapocalypse · 11-30-2002, 04:19 AM

Nope, still no luck. I have found a utility from the manufacturer that will supposedly turn off UDMA66 at a hardware level on the drive in question BUT I need to be in DOS and all my attempts to create a boot disk, with Win98 files on, under Win2K have failed. Miserably.

LATEST:
Did that, the Linux loader now shows it as being UDMA33 which is fine. Still doesn't work though!

lynch · 11-30-2002, 05:50 AM

Go here and download a win98 OEM bootdisk.
You cant make a dos-based bootdisk from a ntfs-based OS.It doesnt have the dos system files you need.That's why the format command in w2k doesnt have an /s switch.
You need to make the ERD set in w2k.
lynch

Wapocalypse · 11-30-2002, 06:02 AM

Already tried the stuff I needed the bootdisk for, it did what it was meant to but Linux still won't boot.

OK. I've now gone back to a very minimal install of 7.3, but what good is that if the next version of the kernel doesn't support my disk?

Here's the important bit of what dmesg produces on the WORKING system:

Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
VP_IDE: IDE controller on PCI bus 00 dev 89
PCI: No IRQ known for interrupt pin A of device 00:11.1. Please try using pci=biosirq.
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
VP_IDE: Unknown VIA SouthBridge, contact Vojtech Pavlik <vojtech@suse.cz>
hda: IC35L060AVVA07-0, ATA DISK drive
hdb: MSI CD-RW MS-8348, ATAPI CD/DVD-ROM drive
hdc: FUJITSU MPF3204AT, ATA DISK drive
hdd: IDE DVD-ROM 16X, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide0: probed IRQ 14 failed, using default.
ide1 at 0x170-0x177,0x376 on irq 15
ide1: probed IRQ 15 failed, using default.
hda: 120103200 sectors (61493 MB) w/1863KiB Cache, CHS=7476/255/63
hdc: 40031712 sectors (20496 MB) w/512KiB Cache, CHS=39714/16/63
ide-floppy driver 0.99.newide
Partition check:
hda: hda1 hda2 < hda5 >
hdc: hdc1 hdc2 hdc3

Now I can't tell you exactly what happens on the new one, but it looks something like this:

hda: IC35L060AVVA07-0, ATA DISK drive
hdb: MSI CD-RW MS-8348, ATAPI CD/DVD-ROM drive
hdc: FUJITSU MPF3204AT, ATA DISK drive
hdd: IDE DVD-ROM 16X, ATAPI CD/DVD-ROM drive
hda: 120103200 sectors (61493 MB) w/1863KiB Cache, CHS=7476/255/63 (UDMA100)
hdc: 40031712 sectors (20496 MB) w/512KiB Cache, CHS=39714/16/63 (UDMA33)
ide-floppy driver 0.99.newide
Partition check:
hda: hda1 hda2 < hda5 >
hdc:

Note the addition of UDMA and the removal of the IRQ part.

Maybe I can stop this happening with a boot parameter? I did hdc=noprobe but then it isn't detected at all.

I don't have the usr/src/.../ide.txt on this install so where can I get that? I'm told it will help.

Amerist · 11-30-2002, 06:13 AM

Have you checked your install CDs to make sure they are without errors? Run mediacheck on disc 1 (you should eventually check all three install discs to make sure they are OK, even if they are retail discs.) Sounds to me like an error in the disc rather than a hard drive thang.

What kind of drives are you using, and what kind of cables are you using for your primary chain? (example UDMA33 on a UDMA33 cable, or UDMA66 on a UDMA66 cable?)

You can't use a UDMA66 drive on a UDMA33 cable
A UDMA33 drive on the same UDMA66 cable as a UDMA66 drive may not work. Some chipsets don't support setting different transfer rates for different devices.

If the discs are OK, and you're not mixing your drives, then may I suggest replacing your IDE cable for your primary IDE channel? I know that sounds a bit unorthodox, but I had a problem installing Mandrake in 97 and I tried everything, except switching the cable... and guess what finally solved the problem. After 2 weeks of pulling my hair out, I found that a single line in the IDE cable had shorted somehow and actually left a visible hole in the cable. I wonder how many yanks and pulls an IDE cable can take before it finally breaks.

Wapocalypse · 11-30-2002, 06:20 AM

I think the discs are OK because as I said at the beginning I had this problem on the latest kernel under 7.3 as well.

The info is just above, but you can now regard it as a UDMA33 drive on a 40 pin cable.

A long time ago a pin connector on the hard drive 'disappeared'

BUT before everyone jumps in crying 'Aha!' it has made NO difference at all and Fujitsu told me it was probably just a ground pin.

Plus, all the other people out there who this doesn't work for (search for 'partition check', and half of the results are this problem, without solutions) probably don't have missing pins.

Amerist · 11-30-2002, 06:29 AM

Which pin was it, exactly. I have a book right here and I can tell you for sure what that pin is used for.

pins 2, 19, 22, 24, 26, 30, & 40 are ground. The rest are important.

Wapocalypse · 11-30-2002, 06:31 AM

Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
VP_IDE: IDE controller on PCI bus 00 dev 89
PCI: No IRQ known for interrupt pin A of device 00:11.1. Please try using pci=biosirq.
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
VP_IDE: VIA vt8233a (rev 00) IDE UDMA133 controller on pic00:11.1
ide0: BM-DMA at 0xb000 - 0xb007, BIOS settings, hda: DMA, hdb: DMA
ide1: BM-DMA at 0xb008 - 0xb00f, BIOS settings, hdc: DMA, hdd: DMA
hda: IC35L060AVVA07-0, ATA DISK drive
hdb: MSI CD-RW MS-8348, ATAPI CD/DVD-ROM drive
hdc: FUJITSU MPF3204AT, ATA DISK drive
hdd: IDE DVD-ROM 16X, ATAPI CD/DVD-ROM drive
hda: 120103200 sectors (61493 MB) w/1863KiB Cache, CHS=7476/255/63 (UDMA100)
hdc: 40031712 sectors (20496 MB) w/512KiB Cache, CHS=39714/16/63 (UDMA33)
ide-floppy driver 0.99.newide
Partition check:
hda: hda1 hda2 < hda5 >
hdc: