Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 13 May 1999 20:21:26 -0500 (CDT)
From:      Chris Dillon <cdillon@wolves.k12.mo.us>
To:        "Jordan K. Hubbard" <jkh@zippy.cdrom.com>
Cc:        Brian McGovern <bmcgover@cisco.com>, stable@FreeBSD.ORG, jkh@FreeBSD.ORG
Subject:   Re: Fubar w/3.2-BETA 
Message-ID:  <Pine.BSF.4.10.9905131933530.46480-100000@mail.wolves.k12.mo.us>
In-Reply-To: <Pine.BSF.4.10.9905122230540.38803-100000@mail.wolves.k12.mo.us>

next in thread | previous in thread | raw e-mail | index | archive | help
> On Wed, 12 May 1999, Jordan K. Hubbard wrote:
> 
> > This is really weird; I can't reproduce this!
> > 
> > I wonder if it's because my test box uses SCSI.  Damn, time to dust
> > off the IDE system I think. :)

More on the quirk I encountered... I just attempted an install of
3.2-19990512-BETA, the latest available on releng3.freebsd.org.  In a
nutshell, I:

Disabled all disableable (is that a word?) devices EXCEPT: fd0, wdc1
(yes, no wdc0), ppc0, sio0, atkbd0, psm0, sc0.

The PCI devices probed/attached are fxp0-fxp6 (though I only had one
fxp card installed the first time around), ncr0, ncr1.

Attached to ncr0 are three 9.1GB UW SCSI drives.

"Custom" install.

I partition each disk using "Use Entire Disk" in the partition editor,
using a true partition entry (not dangerously dedicated), set bootable
on da0 and da1.

Label them as such:
	da0s1a: 100M - /
	da0s1e: 200M - /var
	da0s1f: 1700M - /usr
	da0s1g: remainder (6675MB) - /cache.0
	da1s1e: 100M - /root.bak
	da1s1f: 1000M - /devel
	da1s1g: 900M - /usr/src
	da1s1h: remainder (6675MB) - /cache.1
	da2s1e: 1000M - /usr/obj
	da2s1f: 1000M - /spare
	da2s1g: remainder (6675MB) - /cache.2

Pick bin, dict, doc, games, info, man distributions.
Media - FTP - fxp6 - 3.0 SNAP Server.
Commit.

"You must have at least one swap partition..."
"Couldn't make filesystems properly. Aborting"
"Commit completed with errors.  Not updating /etc"

Whoops.  Go back into the label editor.

Remove:
	da0s1f: 1700M - /usr
	da1s1g: 900M - /usr/src

Create:
	da0s1b: 300M - swap
        da0s1f: 1400M - /usr
	da1s1b: 300M - swap
	da1s1g: 600M - /usr/src

Commit.

"WARNING! Unable to swap to /dev/da0s1b: Device not configured...."
"Making filesystem for /dev/rda0s1a" (or whatever... It happened
quickly)
"Unable to add /mnt/dev/da1s1b as a swap device: Device not
configured."

Over on the debug console:
	DEBUG: MakeDev: Unknown major/minor for devtype -
	DEBUG: MakeDev: Unknown major/minor for devtype -
	DEBUG: MakeDev: Unknown major/minor for devtype -
	DEBUG: MakeDev: Unknown major/minor for devtype -
	DEBUG: MakeDev: Unknown major/minor for devtype -

All of the filesystems are newfs'd, FTP installation begins, and I
abort it (just wanted to make sure it would get that far).

Just noticed something else.  After aborting the install and playing
around a bit with attempting to choose the FTP site again (started
install a second time just fine after re-choosing site), I finally
went to do something else and "I've got a Signal 11.  Not good!".  
Debug console said something like (I really need to write things down
_before_ I make them go away):

DEBUG: Deleting default route on fxp6
DEBUG: Caught signal 11

After I hit OK, it rebooted.

AND, now that I've gone back into sysinstall, I examine the previously
created partitions via label editor.  It seems the second time around
after making the changes in the label editor, they were never actually
committed.  The supposedly deleted/recreated da0s1f and da1s1g are
still the originals (original size), and da0s1b and da1s1b don't
exist.  Should I have hit (W)rite in the label editor the second time
around before I went to re-commit (that might have solved all the
problems)? If so, should the default behaviour be changed to make
those changes automatically upon final "Commit"?

Anyway, assuming I do things right the first time, everything goes
without a hitch.  I'm only thinking about those people who
occasionally screw up.  :-)


-- Chris Dillon - cdillon@wolves.k12.mo.us - cdillon@inter-linc.net
   FreeBSD: The fastest and most stable server OS on the planet.
   For Intel x86 and Alpha architectures (SPARC under development).
   ( http://www.freebsd.org )

   "One should admire Windows users.  It takes a great deal of
    courage to trust Windows with your data."



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.10.9905131933530.46480-100000>