cut-and-paste the 1.3_BETA NOTES file into the standard release notes.

This commit is contained in:
jonathan 1997-12-22 00:40:41 +00:00
parent 141bc49c28
commit 137862e580
3 changed files with 225 additions and 88 deletions

View File

@ -1,13 +1,16 @@
Reminder:
Always back up files before installing or upgrading.
If installing from source, always install and boot a new kernel
before installing a new userland!
If at all possible, you should consult the ``Installation Guide''
document on the NetBSD/pmax web page, at http://www.netbsd.org/ports/pmax.
Installing NetBSD is a relatively complex process, but if you have
the above document in hand it shouldn't be too much trouble.
It has not been possible to reproduce the pmax installation Web page
in fixed-font hardcopy format for the _VER release. You must either visit
the URL above, or request a rendered version (e.g, PostScript).
Please follow the dinstructions at
Please follow the instructions at
http://www.netbsd.org/ports/pmax/pmax-install.html,
@ -16,6 +19,127 @@ to install a miniroot, label a disk, and extract the NetBSD/pmax _VER
installation tar sets.
Install via diskless boot.
--------------------------
The preferred path is to upgrade or install by diskless-booting a
minimal system via NFS, and using that to upgrade or install. The
file
installation/netboot/diskimage.tar.gz
contains a suitable set of files. (it is a tar copy of the contents of
a root filesystem diskimage) . You will need to find an NFS server,
unpack the tarfile, and setup BOOTP/dhcp service for your pmax.
Complete instructions are in the installation notes or the pmax web
page.
Since the system install utility, sysinst, requires a read/write root,
netbooting is only feasible if your NFS server exports the diskless
root read-write. If this is not possible, you should install via
diskimage.
Install via diskimage.
----------------------
If netbooting with a _writable_ NFS root is not possible, the
recommended installation is to unpack and copy a diskimage onto the
raw partition of a disk. The diskimage file is in
installation/diskimage/diskimage.gz
is shipped compressed and is around 8020 kBytes; it uncompresses to
exactly 32Mbytes.
To install the diskimage onto disk rzX on a NetBSD/pmax system, do:
disklabel -W /dev/rrzXc
gunzip -c diskimage.gz | dd of=/dev/rrzX2c bs=10240
Most other NetBSD ports are similar, but use rsdXc instead of rrzXc.
On NetBSD/i386, the `raw disk partition' is the 'd' parttion, so do:
disklabel -W /dev/rsdXd
gunzip -c diskimage.gz | dd of=/dev/rsdXd bs=10240
On NetBSD, be sure to use disklabel -W to enable writing to the label
area of the disk. If you forget this and/or use the `block' device,
the dd command will silently fail.
On MS-DOS, use an unzip utlility, then use rawrite.
Then boot using, e.g,
>> boot -f rz(0,X,0)netbsd # 3100
>> boot 5/rzX/netbsd # 5000/200
>> boot 3/rzX/netbsd # other machines
(NOTE: replace the X with the unit number of your disk:
boot 3/rz2/netbsd to boot drive 2 on a 5000/xxx.)
then continue from ``Once you've booted the diskimage''.
Installing from Ultrix
----------------------
The Ultrix bootloader can boot ECOFF-format NetBSD kernels.
So the technique of dd'ing a disk image into swap, copying
a kernel to the root, and then booting via the "n" argument
should work with Ultrix too. Since Ultrix cannot mount
4.4BSD format FFS filesystems, just dd the diskimage into
your swap partition, copy a netbsd.ecoff kernel from the FTP site,
and proceed from step #5 of the `Upgrade via diskimage' above.
Again, be sure to specify the partition where you dd'ed the NetBSD
diskimage as your root device in step #7!
Once you've booted the diskimage
---------------------------
Once you'e booted a diskimage and pointed the kernel at the approriate
devices to use for root, and dump, NetBSD kernel will prompt you for
a single-user shell;
Enter pathname of shell or RETURN for sh:
At this point, press the RETURN key.
then when you get a standalone root prompt, set your terminal type
TERM=pmax; export TERM;
Next, you need to ensure that the root filesystem is writable so that
sysinst can create temporary files, temporary disklabels, etc.
If you booted via NFS, the diskless root should already be writable.
If you booted from disk, type
mount /dev/rzXY
where X is the disk unit you booted from, and Y is either a or b,
(e.g., use /dev/rz2a for drive 2 with diskimage in the 'a' partition,
and rz2b for diskimage in swap.)
Then, start sysinst. For the ALPHA release, you should start sysinst with
an empty release-version string:
sysinst -r ""
(to stop sysinst automatically looking for base13, etc13, comp13, etc.)
Choose 'install' or 'upgrade'.
When you've finished the install/upgrade, be sure to edit /etc/rc.conf
and set "rc_configured" to YES when you are done. If you're doing an
upgrade, merge your old etc from /etc.old into /etc.
You're then ready to reboot!
Post-intallation notes:
----------------------

View File

@ -1,5 +1,10 @@
Currently NetBSD/pmax supports three different installation methods.
From most convenient to least convenient, they are:
Installing NetBSD/pmax using the sysinst tool and an image of a full
bootable root filesystem is now a relatively painless process. The
diskimage is avaiable via (either a diskimage, dd'able to a raw disk
or tarfile to for NFS .
From most convenient to least convenient, the installation methods
are:
1. Booting as a diskless workstation via Ethernet,
followed by initialization of the local disk and
@ -12,20 +17,29 @@ From most convenient to least convenient, they are:
NetBSD/pmax root filesystem, and moving the disk
to the target.
4. Installation from Ultrix or other OSes by putting a copy
of the diskimage into the existing swap partition and a copy
of the NetBSD kernel into your Ultrix root filesystem.
Before you start, you must choose an installation method. If you have
an Ethernet connection to an NFS server that can provide even ~30M for
a diskless-root filesystem, then insatllation via the net is best.
a diskless-root filesystem, then installation via the net is best.
Next best, if your DECstation is already running Ultrix and has two
disk drives (or one, if you live dangerously), is to copy a diskimage
onto one drive. Finally, you can install by using a second machine as
a helper to prepare a bootable NetBSD/pmax disk.
If your target is going to run diskless, then installation proceeds as for
method 1.
If your target is going to run diskless, then installation proceeds as
for method 1.
This release of NetBSD/pmax uses the new sysinst installation utility.
You should examine the guide on the NetBSD/pmax web site, which has
more complete and more up-to-date instructions and tips than are given in
this document.
more complete and more up-to-date instructions for sysinst. The
following is a brief synopsis which has been successfully followed by
both first-time NetBSD/pmax installers and to upgrade existing
gsystems.
You should familiarize yourself with the console PROM environment
and the hardware configuration. The PROMs on the older Decstation
@ -50,13 +64,10 @@ and on other 5000 series machines,
boot 3/mop
You will also need to know the total size (in sectors) and the
approximate geometry of the disks you are installing onto, so that
you can label your disks for the BSD fast filesystem (FFS). The
system comes with sample disk labels for DEC-supplied SCSI drives.
For third-party drives you will need to get head/sector/cylinder
information. For newer ZBR drives you can safely make this
information up.
approximate geometry of the disks you are installing onto, so that you
can label your disks for the BSD fast filesystem (FFS). For most SCSI
drives (including all SCSI-2 drives), the kernel will correctly detect
the disk geometry. The sysinst tool will suggest these as the default.
If you're installing NetBSD/pmax for the first time it's a very good
@ -65,6 +76,7 @@ installing NetBSD. Changing the size of partitions after you've
installed is difficult. If you do not have a spare bootable disk, it
may be simpler to re-install NetBSD again from scratch.
If you install by copying a disk image, and you want to change the size
of the root partition from the default 32Mbytes, you will need a second
`scratch' disk. You should copy the diskimage onto the `scratch' disk,
@ -78,10 +90,10 @@ partition, which is why you need two disks to tailor the root filesystem
size.
Assuming a classic partition scheme with separate root (`/') and /usr
filesystems, a comfortable size for the NetBSD root filesystem partition
is about 32M. A good initial size for the swap partition is twice the
amount of physical memory in your machine (though, unlike Ultrix, there
are no restrictions on the size of the swap partition that would render
part of your memory unusable). The default swap size is 64Mbytes, which
is adequate for doing a full system build. A full binary installation,
with X11R6.3, takes about 130MB in `/usr'.
filesystems, a comfortable size for the NetBSD root filesystem
partition is about 32M. A good initial size for the swap partition is
twice the amount of physical memory in your machine (though, unlike
Ultrix, there are no restrictions on the size of the swap partition
that would render part of your memory unusable). The default swap
size is 64Mbytes, which is adequate for doing a full system build. A
full binary installation, with X11R6.3, takes about 150MB in `/usr'.

View File

@ -1,91 +1,92 @@
It may be possible to easily upgrade your existing NetBSD/pmax system
using the upgrade program in the miniroot.
NOTE: this has not been tested and should be done with caution,
after backing up your system.
The preferred upgrade path is to set up a diskless-boot host, unpack
and boot boot 1.3 diskimage as for a network installation, and to use
the sysinst tool to upgrade your system. Please see the `Installation'
section for further information.
If you wish to upgrade your system by this method, simply select
the `upgrade' option once the miniroot has booted. The upgrade program
with then guide you through the procedure. The upgrade program will:
* Enable the network based on your system's current
network configuration.
Upgrade via diskimage.
----------------------
* Mount your existing filesystems.
If you cannot netboot, the recommended path is to upgrade by booting a
diskimage from your swap partition. Pmaxes cannot boot out out of
anything but the 'a' partition. However, you *can* boot an upgrade
kernel off your 'a' partition and tell that kernel to use your
'b' partition as its root. The steps to do this
(after you've fetched the diskimage) with a current root of rzX are:
* Extract binary sets from the media of your choice.
1) boot single-user from your current root, rzX.
Be *sure* not to start swapping:
* Fixup your system's existing /etc/fstab, changing the
occurrences of `ufs' to `ffs' and let you edit the
resulting file.
>> boot -f rz(0,X,0)netbsd -s # 3100
>> boot 5/rzX/netbsd -s # 5000/200
>> boot 3/rzX/netbsd -s # others
* Make new device nodes in your root filesystem.
(NOTE: replace the X with the unit number of your disk:
boot 3/rz2/netbsd to boot drive 2 on a 5000/xxx.)
* Copy a new kernel onto your root partition.
NOTE: the existing kernel WILL NOT be backed up; doing
so would be pointless, since older NetBSD kernels will
not work correctly with all NetBSD _VER executables.
2) When you get a single-user prompt, remount the
root filesystem read-write. (You wil need to update the
kernel soon.)
* Install a new boot block.
# mount /
* Check your filesystems for integrity.
Then mount the filesystem with the diskimage, and
uncompress and dd the diskimage into swap (b) partition.
You will also need /usr mounted to run gunzip:
While using the miniroot's upgrade program is the preferred method
of upgrading your system, it is possible to upgrade your system
manually. To do this, follow the following procedure:
# mount /usr
# gunzip -c diskimage.gz | dd bs=10240 of=/dev/rrzXb
* Place _at least_ the `base' binary set in a filesystem
accessible to the target machine. A local filesystem
is preferred, since there may be incompatibilities
between the NetBSD _VER kernel and older route(8)
binaries.
3) Mount swap (b) partition readonly on /mnt:
# mount -r -t ffs /dev/rzXb /mnt
* Back up your pre-existing kernel and copy the _VER
kernel into your root partition.
4) Copy the kernel from the B partition to your root:
cp -p /mnt/netbsd /netbsd-1.3
(this is important; you want the kernel in / and swap
to be a release kernel, or the release binaries will not work.)
* Reboot with the _VER kernel into single-user mode.
5) halt:
# halt
* Check all filesystems:
6) Reboot with an argument of "n", telling the kernel to
ask what device to use as root:
/sbin/fsck -p
>> boot -f rz(0,X,0)netbsd-1.3 n # 3100
>> boot 5/rzX/netbsd-1.3 n # 5000/200
>> boot 3/rzX/netbsd-1.3 n # others
* Mount all local filesystems:
(NOTE: the n after the kernel name is a literal "n",
not the disk unit number or partition.
It is an argument telling the kernel to ask for a root device.
NOTE: replace the X with the unit number of your disk:
boot 3/rz2/netbsd to boot drive 2 on a 5000/xxx.)
/sbin/mount -a -t nonfs
* If you keep /usr or /usr/share on an NFS server, you
will want to mount those filesystems as well. To do
this, you will need to enable the network:
7) The "n" argument tells the kernel to prompt you for
the root device, dump device, and root fileysystem type.
Enter when the kernel asks for
Root device:
sh /etc/netstart
Tell it rzXb, where X is the same disk unit as in step 6.
Here's an example, again assuming drive 2 as in step 6:
NOTE: the route(8) commands may fail due to potential
incompatibilities between route(8) and the NetBSD _VER
kernel. Once you have enabled the network, mount the
NFS filesystems. If you use amd(8), you may or may not
have to mount these filesystems manually. Your mileage
may vary.
KN03-AA V5.2b (PC: 0x80051f1c, SP: 0xffffdeb0)
>> boot 3/rz2/netbsd n
* Make sure you are in the root filesystem and extract
the `base' binary set:
boot device: rz2
root device (default rz2a):
<<< enter `rz2b' >>>
dump device (default rz2b):
<<< enter `none' >>>
file system (default generic):
<<< enter `ffs' >>>
root on rz2b
cd /
tar --unlink -zxvpf /path/to/base13.tar.gz
NOTE: the `--unlink' option is _very_ important!
then continue from the ``Once you've booted the diskimage'' step of
the Installation instructions.
* Sync the filesystems:
sync
* At this point you may extract any other binary sets
you may have placed on local filesystems, or you may
wish to extract additional sets at a later time.
To extract these sets, use the following commands:
cd /
tar --unlink -zxvpf <path to set>
NOTE: you SHOULD NOT extract the `etc' set if upgrading. Instead, you
should extract that set into another area and carefully merge the changes
by hand.