pcs_weak_memory

This is part of The Pile, a partial archive of some open source mailing lists and newsgroups.




Date: Thu, 19 Oct 2000 17:30:39 -0700
To: svlug@svlug.org
Subject: Re: [svlug] Re: Intel D815EAA woes
From: Rick Moen <rick@linuxmafia.com>

begin  Ian MacLure quotation:

> Finally were able to install some Linuxen on the one good box but
> still only from CD. Install Floppy? Fuhgeddaboudit.....

You know, faced with that situation, I would extract the hard drive from
one of those boxes, put it in my K6/233 [1], install some distribution
onto it there, adjust the drivers and lilo for funtionality in the
target box, and tranplant it back.

Then, of course, I'd compile a kernel a few times with make -j, because
I'd suspect....

> After I did some BIOS tinkering ( UDMA100 off, QUICK/SILENT BOOT OFF,
> rearrange boot device priotities, etc )
> it became evident that we had suspect memory....

Ever since Intel helped ease the concept of parity RAM into oblivion
with its Triton motherboard chipsets (thanks loads, Intel!), crappy RAM
has been a lurking menace for PC users.  My rule of thumb is that
bizarre, inexplicable behaviour in a Linux box usually means either bad
RAM, the box having been cracked, or a failing hard drive, in that
order.

[1] With confidence if it was SCSI, and a little trepidation with ATA
subsystems like yours, since ATA manufacturers have often failed to
standardise drive addressing well enough to reliably transport
data-bearing drives without subtle geometry mismatches and sometimes
data corruptions.

===


the rest of The Pile (a partial mailing list archive)

doom@kzsu.stanford.edu