SCIENTIFIC-LINUX-USERS Archives

November 2005

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Kevin Hill <[log in to unmask]>
Reply To:
Kevin Hill <[log in to unmask]>
Date:
Tue, 22 Nov 2005 09:18:40 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (45 lines)
Kevin Hill wrote:
> Hi,
> Here's what I spent my weekend doing at home:
> I'm trying to install SL 4.1 (or 4.2 rc2?3, same results) on a system 
> that is already running Fedora Core 1.
> 
> The system has 4 60GB IDE HD's set up as a couple raid 1 arrays for /usr 
> and /var and a big raid 5 array.
> 
> When I run the installer I get errors about DMA timeouts and when it 
> eventually times out and gets to the partitioning screen, its missing 1 
> of the drives, and even weirder, the raid1 devices that were on both 
> that drive and another drive it does see are missing as well.
> 
> I though the problem might be the VIA chipset built in the motherboard, 
> so I switched to a couple ide PCI cards I had lying around.  The drives 
> are currently running off:
> 
> IDE interface: CMD Technology Inc PCI0649 (rev 02)
> Unknown mass storage controller: Promise Technology, Inc. 20267 (rev 02)
> 
> although swapping controllers made no difference in symptoms.
> 
> I thought the problem might be following one of the drives, which it 
> seemed to when swapping things around, however when i tried installing 
> with the suspect drive removed, which shouldn't matter in this setup, I 
> got the same symptoms.
> 
> I tried running the installer with an ide=nodma option, and that 
> resulted in a kernel panic where it usually gets the dma errors.

Update:
Found a live cd based on SL4.1 and it boots and starts the raid arrays 
correctly, so its definitely something the boot kernel or anaconda is 
doing... time to start digging I guess...

-kevin



-- 
Kevin M. Hill       http://home.fnal.gov/~kevinh/
CD/CSS/CSI	         http://www-csi.fnal.gov/
Fermilab	             http://www.fnal.gov/

ATOM RSS1 RSS2