Date: 21 Mar 1995 17:35:17 EST From: "RLARSO01" <HARRIS.RLARSO01@IC1D.HARRIS.COM> To: questions@FreeBSD.org Cc: "RONALD N. LARSON" <HARRIS.RLARSO01@IC1D.HARRIS.COM> Subject: FreeBSD 2.0 Installation problem/question/concern Message-ID: <ATCSD3L.RLARSO01.3960.1995 0321 17 35 17 35>
next in thread | raw e-mail | index | archive | help
I'm recently purchased the 2.0 1-95 Release CD-ROM from Walnut Creek.
I'm switching from Coherent 4.2.05 to FreeBSD as far as a home use
UN*X platform goes because of more cutting edge stuff, as well as Mark
Williams Co. being nonexistent since 1 Feb. God willing, and with any
kind assistance on your part (fingers crossed), you'll have a very
thankful & fanatically faithful convert on you hands...:-)
Currently I'm at the point of where you do your first reboot off the
hard disk, and install the rest of the goodies off the CD (which I
would VERY much like to achieve ASAP). After all the device status
messages have been pumped out (they SEEM to be correct,everything that
I have on my system registers - I don't know what all the devnames
translate out to yet), I get a "panic: could not mount root" message.
Using the plain 2.0 boot disk for initial setup, before a disk sync
and reboot is done, this is what I get:
changing root device to sd0a....
--- something about adjusting d partition blocks on wd0 -----
panic: could not mount root
Then the sync/reboot goes into effect.
This would be fine if I had set up BSD root in sd0a, but if I've got
this devname convention down right, it should be looking for the
kernel in sd1a - I have three hard drives, 1 325MB IDE drive, 1 327MB
SCSI-2 drive, and 1 1.8GB SCSI-2 drive with an 856MB BSD
partion/"slice" :-) at the beginning of the disk (and I've already
disabled extended translation and blew away the disk down to nothing).
Boot defaults to wd(0,a) still so if I type in hd(2,a)/kernel it boots
up saying it's booting sd(0,a)kernel. If I type in hd(1,a)kernel it
doesn't find anything, same with wd(1,a)kernel. However if I type in
sd(2,a)kernel it does start booting up except at the end it tries to
change the boot device to sd2a (aggggh!) and the issues the same panic
message. Again I think if I've caught a clue about the devnames, it
should be looking for the initial kernel in sd1a because I set things
up in my second SCSI drive.
Hoping that maybe the 2-10-95 snapshot could help me get further, I
rawwrote (is that a valid word?) the boot & cpio floppies and tried
again. I noticed that booting was quicker/smoother (hopefully an
indication of better things on the way!) and some of the boot-up
messages changed, but basically I still get the same panic message.:-(
Oh, I did get to install the cpio disk first...ha ha.
Using the 2.1-950210-SNAP boot/cpio disks for initial setup, I noticed
that Boot now defaulted to wd(2,a) and would find the kernel without
any redirection, but it would still get to the point of saying
"changing /dev/root to sd2a...." (aggggh! again) and then the panic in
the streets message with a sync/reboot to go. Is there anything else I
can do? Should I try again with ext. translation set on? If I do get
this to work with the SNAP disks, will I be able to load the 2.0 CD or
should I get the rest of the snapshot? Is this message getting to be
too long?
I would really appreciate any comments/suggestions/assistance, and I
am willing to devote the time and effort to get this setup going. I'll
post my system configuration at the end of this message. As far as I
know, all of my hardware is supported/compatible. At any rate thaks in
advance, and I'm looking forward to contributing to this effort in
anyway in the future (once I'm past first base of course).
Sincerely,
Ron Larson
rlarso01@ic1d.harris.com
Here's my system specs (thanks for bearing with me!):
Vega 486DX2-66 Motherboard with Haydn Symphony II chipset,AMI BIOS
(11-11-92 date), 32MB RAM (could be 64MB soon), 1MB cache
Diamond Viper VLB 2MB VRAM (Weitek P9000 based)
Adaptec AHA2842VL SCSI adapter (AIC-7770 based) - host adap. #1 with:
- a Quantum LPS340 327MB SCSI-2 drive, ID #0 (DOS data only)
- a Quantum PD1800S 1.8GB SCSI-2 drive, ID #1 (855MB BSD slice, 62MB
OS/2 2.11 HPFS system part., 800MB OS/2 HPFS data part.)
- a Tandberg 4100 2GB tape drive, ID #2
A generic FooSong,Ltd. Taiwan VLB IDE controller with 2 ser., 1 par.,
1 game port with:
- a Western Digital AC2340 325MB IDE drive ( 323MB DOS 6.22/Windows
for Workgroups 3.11 part., 1MB OS/2 2.1 Boot Manager part. which
finds the BSD slice without a glitch, or so I may think)
- a TEAC Combo 3.5/5.25 floppy drive
Soundblaster 16 SCSI-2 (AHA 1522 equiv.?) - host adap. #2 with:
- a Toshiba XM-3401 2x CD-ROM drive, ID #4
Zoom 14.4 v.32bis FaxModem (set to COM4).
As I've indicated, I'm using OS/2 Boot Manager. I have disabled
extended translation so the geometry for my 1.8GB drive is 64 heads,
32 sectors, and 1717 cylinders. I used BSD's FDISK to re-set up my
slices, with slice 1 going to BSD, 2 going to OS/2 system, 3 going to
OS/2 data, 4 unused).
For disklabeling, my a partition is 20MB, my b swap partition is
128MB, my e partition is 250MB, my f partition is 350 something, and g
partition is 105MB, all being '/', swap, '/usr', '/usr/local', and
'/usr/users' respectively (h partition is unused).
I was getting 32k block size messages originally prior to disabling
ext. translation and haven't gotten that since then (plus the
filesystems were getting an initial build). But at the time I also
didn't know what the Adaptec's geometry was (it's 255 heads, 63
sectors). Do you think setup would have worked had I had this data at
the time? Or does BSD really hate ext. translation like Clinton hates
Congress? Is there a problem with my using the 2nd SCSI drive (i.e.
the 3rd hard drive in the system)? Or does God have a special plan for
me?
Once again thanks in advance, cause I could sure use the help.