Subject: Re: IBM RS/6000 - more trouble or just my terminal?
To: None <port-prep@netbsd.org>
From: Keith Bennett <kanezor@gmail.com>
List: port-prep
Date: 01/31/2005 22:56:09
On Mon, 31 Jan 2005 20:27:43 +0100, Jochen Kunz
<jkunz@unixag-kl.fh-kl.de> wrote:
> On Mon, 31 Jan 2005 00:31:55 -0600
> Keith Bennett <kanezor@gmail.com> wrote:
>
> > It said "reset: not found" ... I don't think reset is included on the
> > installation diskette... ;)
> You use reset(1) on the host shell where your terminal emulator runs. If
> you have a hardware terminal use its "screen reset" or the like fuction.
>
> > It seems it didn't recognize cu or minicom as a terminal, but xterm
> > worked.
> cu(1) and minicom(1) are no terminal types. They are terminal programms
> running on an other machine emulating a terminal.
>
> > That'll be something to look into, I think. My terminal is an IBM
> > InfoWindow II 3153 if anyone is interested.
> Aha. A hardware terminal. Either your terminal isn't set up to use the
> proper terminal emulaton ("vt100", "vt220" or "ANSI" should do) or its
> emulaton is buggy. Unfortunately the later is quite common.
None of those worked. The only one that both NetBSD would recognize
and would work on my terminal is "vt52".
> > As for the installation... like I said earlier, I'm new to *BSD (new
> > to any *nix for that matter). I loaded up the sysinst program, and
> > tried installing... selected Full installation, when it asked if I
> > wanted to edit the MBR or use entire disk, I selected the entire disk,
> > it then asked "Yes or no?" with no information regarding it... wtf,
> > eh?
> PReP machines need a MBR partition table. There are two partitions in
> that table: One containing the OS kernel and one conaining the NetBSD
> file systems. The later partition is split up into slices (BSD term for
> partition) by a disklabel(5). Slice A is allways boot, B swap, C entire
> disk / entire BSD part of the disk, D entire disk, E etc. are free for
> any use.
>
> > Any help?
> I posted detaild instalation instructions and an installation log to the
> list some days ago. Search the archive for:
> Subject: Re: netbsd 2.0 on rs6000 43p, what I've been able to do so far,
> and where i got stuck
> Date: Tue, 25 Jan 2005 22:08:04 +0100
> also note my mail with
> Date: Wed, 26 Jan 2005 18:42:17 +0100
I followed those instructions. I even verified that I'm using the
patched kernel from
ftp://ftp.netbsd.org/pub/NetBSD/arch/prep/snapshot/release-2-0_20050119/installation/floppy/,
by adding a "patched" folder to the /prep/installation/ folder of the
ISO, adding the images there, and burning a new CD. Unfrotunately,
when I tried installing from that CD, I noticed that a couple of the
packages were not able to be read, so what I ended up doing is setting
up the drive with my second CD, rebooting, then installing with my
first CD.
I still have the problem of the kernel panicking with "lockmgr: no context".
Here's a log of what I'm doing (I'm writing this out manually, so ugh!):
[bootup using sysinst_com0.fs floppy]
Terminal type? [vt100] vt52
(I)nstall, (S)hell, or (H)alt ? s
# mount -t cd9660 /dev/cd0a /mnt
cd0: no disk label
# cd /mnt/prep/installation/patched
# ls
generic.fs generic_com0.fs sysinst.fs sysinst_com0.fs
# dd if=sysinst_com0.fs of=/dev/sd0
uid 0, pid 19, command dd, on /: file system full
/: write failed, file system is full
dd: /dev/sd0: No space left on device
857+0 records in
856+0 records out
438272 bytes transferred in 0.739 secs (593060 bytes/sec)
# dd if=sysinst_com0.fs of=/dev/sd0c
2858+1 records in
2858+1 records out
1463708 bytes transferred in 5.975 secs (244972 bytes/sec)
# cd /
# fdisk /dev/sd0
fdisk: DIOCGDEFLABEL: Inappropriate ioctl for device
fdisk: DIOCGDINFO: Inappropriate ioctl for device
# fdisk /dev/sd0c
sd0: no disk label
Disk: /dev/sd0c
NetBSD disklabel disk geometry:
cylinders: 11474, heads: 5, sectors/track: 309 (1545 sectors/cylinder)
total sectors: 17774160
BIOS disk geometry:
cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 17774160
Partition table:
0: Linux/MINIX (sharing disk with DRDOS) or Personal RISC boot (sysid 65)
start 0, size 2879 (1 MB, Cyls 0-0/45/45), Active
1: <UNUSED>
2: <UNUSED>
3: <UNUSED>
# fdisk -u -1 /dev/sd0c
sd0: no disk label
Disk: /dev/sd0c
NetBSD disklabel disk geometry:
cylinders: 11474, heads: 5, sectors/track: 309 (1545 sectors/cylinder)
total sectors: 17774160
BIOS disk geometry:
cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 17774160
Do you want to change our idea of what BIOS thinks? [n]
Partition 1:
<UNUSED>
The data for partition 1 is:
<UNUSED>
sysid: [0..255 default: 169]
start: [0..1106cyl default: 2879, 0cyk, 1MB] 1cyl
size: [0..1105cyl default: 17758095, 1105cyl, 8671MB]
We haven't written the MBR back to disk yet. This is your last chance.
Partition table:
0: Linux/MINIX (sharing disk with DRDOS) or Personal RISC boot (sysid 65)
start 0, size 2879 (1 MB, Cyls 0-0/45/45), Active
1: NetBSD (sysid 169)
start 16065, size 17758095 (8671 MB, Cyls 1-1106/99/34)
2: <UNUSED>
3: <UNUSED>
Should we write new partition table? [n] y
# disklabel -i -I /dev/sd0c
partition> P
8 partitions:
# size offset fstype [fsize bsize cpg/sgs]
a: 787950 16065 4.2BSD 1024 8192 43776 # (Cyl. 10*- 520*)
b: 525300 804015 swap # (Cyl. 520*- 860*)
c: 17774160 0 unused 0 0 # (Cyl. 0 - 11504*)
d: 2098110 1329315 4.2BSD 1024 8192 43776 # (Cyl. 860*- 2218*)
d: 2098110 3427425 4.2BSD 1024 8192 43776 # (Cyl. 2218*- 3576*)
d: 12245670 5525535 4.2BSD 1024 8192 43776 # (Cyl. 3576*- 11502*)
partition> R
Rounding [sectors]: c
partition> W
Label disk [n]? y
NetBSD slice at 16065, partition C at 0
Write outside MBR partition? [n]: y
Label written
partition> Q
# reboot
...
rebooting, inserting other disc, will verify sd0 was written correctly
before install
...
Terminal type? [vt100] vt52
(I)nstall, (S)hell, or (H)alt? s
# fdisk /dev/sd0
sd0: no disk label
Disk: /dev/sd0c
NetBSD disklabel disk geometry:
cylinders: 11474, heads: 5, sectors/track: 309 (1545 sectors/cylinder)
total sectors: 17774160
BIOS disk geometry:
cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 17774160
Partition table:
0: Linux/MINIX (sharing disk with DRDOS) or Personal RISC boot (sysid 65)
start 0, size 2879 (1 MB, Cyls 0-0/45/45), Active
1: NetBSD (sysid 169)
start 16065, size 17758095 (8671 MB, Cyls 1-1106/99/34)
2: <UNUSED>
3: <UNUSED>
# disklabel /dev/sd0
... (a lot of text goes beyond upper part of terminal, is there a way
to display it by page?)
total sectors: 17774160
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0 #microseconds
track-to-track seek: 0 #microseconds
drivedata: 0
8 partitions:
a: 787950 16065 4.2BSD 1024 8192 43776 # (Cyl. 10*- 520*)
b: 525300 804015 swap # (Cyl. 520*- 860*)
c: 17774160 0 unused 0 0 # (Cyl. 0 - 11504*)
d: 2098110 1329315 4.2BSD 1024 8192 43776 # (Cyl. 860*- 2218*)
d: 2098110 3427425 4.2BSD 1024 8192 43776 # (Cyl. 2218*- 3576*)
d: 12245670 5525535 4.2BSD 1024 8192 43776 # (Cyl. 3576*- 11502*)
# /sysinst
(a: Install NetBSD to hard disk)
(b: Yes)
(a: sd0)
(a: Full installation)
(a: Edit the MBR partition table)
(a: Megabytes)
(x: Partition table OK)
(a: Set sizes of NetBSD partitions)
Here's the default NetBSD partition sizes:
MB Cylinders Sectors Filesystem
319 (8542) 424 655080 + /
128 170 262650 swap
0 0 0 tmp (mfs)
0 0 0 /usr
0 0 0 /var
0 0 0 /home
MB Cylinders Sectors Filesystem
319 424 655080 + /
128 170 262650 swap
64 85 131325 tmp (mfs)
1024 1358 2098110 /usr
1024 1358 2098110 /var
5979 7926 12245670 /home
(Accept partition sizes. Free space 1 MB, 2 free partitions.)
That brings me to this screen:
Start MB End MB Size MB FS type NewFS Mount Mount point
--------- --------- --------- ---------- ----- ----- -----------
a: 7 391 384 FFSv1 Yes Yes /
b: 392 648 256 swap
c: 0 8677 8678 Whole disk
d: 649 1672 1024 FFSv1 Yes Yes /usr
d: 1673 2697 1024 FFSv1 Yes Yes /var
d: 2698 8676 5979 FFSv1 Yes Yes /home
d: 0 0 0 unused
(Partition sizes ok)
[mydisk]: kanenetbsd
Shall we continue?
(Yes)
Write outside MBR partition? [n]: y
(hit enter to continue)
(newfs'ing all the partitions)
(a: Progress bar (recommended))
(selected CDROM/DVD) -- I didnt pay attention to what letter option it
was, bleh... oh well
(hit enter to continue)
(Installing)
All selected distribution sets unpacked successfully. (hit enter to continue)
(setup timezone)
(no not change cipher)
Here's anothere place where my terminal is rather annoying... it
displays a prompt with "Yes" and "No", yet there's no other text to
say what it's for. I've tried this enough times to know that it's
asking me if I want to change my root password, which of course, I
select "Yes".
(Enter password)
(Re-enter password)
(Use csh as shell)
Installation is complete. (hit enter to continue)
(d: Reboot the computer)
...
When firmware loads up (terminal displays RS/6000 with test sequence),
I eject the boot floppy.
...
Now comes up the NetBSD kernel! Oh yay! Except, it's panicked! Oh no!
Same screen as before.
>> NetBSD/prep BOOT, Revision 1.5
>> (builds@build, Mon Dec 6 00:12:27 UTC 2004)
Boot: Loading in()
3537044+185960=0x38d160
start=0x100000
panic: lockmgr: no context
The operating system has halted.
Please press any key to reboot.
At this point, I turn off the computer and wait for more help. ;)
I'm sure you'll probably not like me setting up the partitions the way
I did; is that what's causing this?