[4234] in RedHat Linux List
Re: redhat-digest Digest V96 #566, errata
daemon@ATHENA.MIT.EDU (Walter B Kulecz PhD)
Wed Nov 13 14:09:35 1996
Date: Wed, 13 Nov 1996 14:04:52 -0600 (CST)
From: Walter B Kulecz PhD <wally@wahine.jsc.nasa.gov>
To: redhat-list@redhat.com
In-Reply-To: <199611131501.KAA23193@redhat.com>
Resent-From: redhat-list@redhat.com
Reply-To: redhat-list@redhat.com
It seems colgate's installation on mixed IDE and SCSI systems is broken.
I have a support ticket out on this and recieved the automated "we can't
help" response after ten days. I asked for an RMA or number to call to
get one which only got a useless automated response from sales.
Since I have access to more hardware than most when a bit of "free time"
became available I decied to try a pure 2940 SCSI install.
I got a perfect clean install by pulling my IDE cables (disabling in CMOS
was not enough) and diconnecting my SCSI drive 0 replacing it with an
external 1G set to ID 0. I used the NEC MBR-7 CDROM slot 0. Choose
"Install Everything " and it worked perfectly, mouse worked in metroX
setup, lilo wrote to MBR of sda and a clean reboot.
I then ftp'd all the most current errata from a local machine I'd stored
them, did "cd /updates" followed by "rpm -Uvh *.rpm" everything except
rdate (which failed with installed is newer) installed fine and
ghostscript gave an error: can't delete /usr/share/ghostscript/3.33
directory is not empty (which is easy enough to remove once I know what is
there is not needed for anything). Next I did the series of commands as
root described in the errata. Finally I ran lilo since one of the rpm's
had the ping bug fix kernel.
Easy enough and no special instructions, doesn't the errata say to do "rpm
-Uvh *.rpm" somewhere? Updating 3.0.3 to the 2.0 kernel required this to
be repeated twice as some things failed from "wrong order dependencies"
but the second one picked up what had failed the first time, colgate
appeared about as easy as can be.
It seems things either work great or you are hosed, not much middle
ground and very few clues available for those hosed. I'd wager having /
on IDE and /usr on SCSI was the problem, but why? and what's the
solution? I have a working colgate now but only at the expense of
breaking my system for everything else (WinNT, Win95, OS/2, etc. I use
the IDE syquest to boot the system I need at the moment)
The lilo and kernel layout look very different from the 3.0.3 3.0
upgrade, I assume this is the effects of "modular everything" which it
seems still has some bugs. I need an answer other than "blow away the
redhat 2.0.18 and untar 2.0.25 with a standard lilo" and copy everything
to my mixed IDE SCSI drive layout.
I've seen others mention this but no fix, metroX leaves the console with
a very poor display on exit, what's the fix? I also had to kill gpm with
"gpm -k" for metroX to start. I'll try the gpm -R and alternate mouse
device fix later.
The new window manager makes me what to update my two full time linux
boxes but until I learn the reason for this problem and its solution I
can't risk it.
--wally.
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
________________________________________________________________________
http://www.redhat.com/RedHat-FAQ http://www.redhat.com/RedHat-Errata
http://www.redhat.com/RedHat-Tips http://www.redhat.com/mailing-lists
------------------------------------------------------------------------
To unsubscribe: mail -s unsubscribe redhat-list-request@redhat.com < /dev/null
