Re: Xenix Filesystem
From: Brian K. White (brian_at_aljex.com)
Date: 07/26/04
- Previous message: Tony Lawrence: "Re: Cut him a break"
- In reply to: Ralph Scharping: "Re: Xenix Filesystem"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Mon, 26 Jul 2004 12:31:54 -0400
Ralph Scharping wrote:
> Stupid question:
>
> how do I mount a new filesystem in SCO Open Server 5.0.5?
>
> I went through the procedure of mkdev xenix and mkdev 0 IDE-1,
> answered all those questions. Now that is finisched and I don't know
> how to mount. Mounting the drive the way I do in Unix does not work.
> The root filesystems of the host system are not mounted from "regular"
> devices, either. They have devices by the name of "root" or "boot"
> or stuff like that, which I gather can only be a link or a pointer or
> something of the sort. I was not able to tell, wether my new device
> is hidden behind a hd01 or rhd01 or what.
>
> The Disk I am trying to mount contains a Xenix filesystem and is
> secondary slave IDE drive. So far I was not able to fdisk the drive
> because I don't now the associated device.
>
> Thanks!
> Ralph
The new, whole-disk device is probably /dev/hd10.
There is probably 1 fdisk partition on that that occpies the whole disk,
that is probably /dev/hd11.
There is probably at least 2 or 3 divvy partitions within that fdisk
partition. Those divvy partitions each have a filesystem on them, except the
recovery and swap areas.
When the disk was running under xenix and was the only disk in the box,
those divvy partitions had device names like /dev/root and /dev/boot,
/dev/swap, /dev/recover, and possibly /dev/u.
Under 5.0.5, you already have devices with those same names which point to
filesystems on the main 5.0.5 disk.
I don't remember if this works or not, but you can try running "divvy
/dev/hd10"
with any luck it will read the existing divvy table and present you with a
list of divvy partitions with generic names based on the device & partition
numbers rather than the convenience names they had under xenix.
so, where under xenix if you had run "divvy /dev/hd00" (under xenix it was
the first disk)
and gotten something like this:
+-------------------+------------+--------+---+-------------+------------+
| Name | Type | New FS | # | First Block | Last Block |
+-------------------+------------+--------+---+-------------+------------+
| boot | EAFS | no | 0 | 0| 15359|
| swap | NON FS | no | 1 | 15360| 1063935|
| root | XENIX | no | 2 | 1063936| 3623935|
| u | XENIX | no | 3 | 3623936| 35824939|
| | NOT USED | no | 4 | -| -|
| | NOT USED | no | 5 | -| -|
| recover | NON FS | no | 6 | 35824940| 35824949|
| hd0a | WHOLE DISK | no | 7 | 0| 35832950|
+-------------------+------------+--------+---+-------------+------------+
(except with all much smaller numbers since this is really a listing from a
36 gig scsi disk on 5.0.6 and I editied where it said "HTFS" to say "XENIX")
if you now run "divvy /dev/hd10" (under 5.0.5 this will be the 2nd disk)
you should get a listing something like above except instead of "boot" and
"swap" you'll have things that start with "hd1.."
I think all you really need to know is what partition # the filesystems
occupy. In the above example, only 2 and 3 would be interesting for
recovering data, transplanting applications, recovering user accounts
,printer configs, emails, uucp system info etc... you may not even have a
"u", or you may have more and 4 & 5 might be used. Generally, just look for
partitions that say "XENIX" and ignore the rest. More accurately would be to
say, ignore 0, 1, 6, 7 and any that say "NOT USED", which allows for the
possibility that someone might have created some other type of filesystem.
Quit out of divvy without letting it do anything! By default I think it may
try to create filestystems and that would be _bad_.
see:
http://osr5doc.sco.com:457/cgi-bin/man/man?divvy+ADM
http://osr5doc.sco.com:457/cgi-bin/man/man?hd+HW
Assuming you have already run "mkdev hd" once and rebooted, then try the
following:
mknod /dev/hd1root b 1 106
mknod /dev/rhd1root c 1 106
mknod /dev/hd1u b 1 107
mknod /dev/rhd1u c 1 107
mkdir /xroot
mkdir /xu
mount -t XENIX /dev/hd1root /xroot
mount -t XENIX /dev/hd1u /xu
In the mknod commands I'm not sure the "1" is the correct major number for
ide devices but it's ok to try it. If it doesn't work, it won't hurt
anything as long as you don't try to write to the created device node by
using mkfs or dd or something. Just trying to mount and having it fail is
ok. If the mount commands below don't fail, then you are ok.
You can find out the correct major number by doing this:
# ls -l /dev/hd10
brw------- 1 sysinfo sysinfo 1, 64 Jan 12 2004 /dev/hd10
If yours looks the same as abobve, then "1" is corect. Otherwise, in my
mknod commands above, replace "1" with whatever came just before the comma
in the ls command above.
So if your ls command looked like this:
# ls -l /dev/hd10
brw------- 1 sysinfo sysinfo 4, 64 Jan 12 2004 /dev/hd10
then use mknod commands like this:
mknod /dev/hd1root b 4 106
-- Brian K. White -- brian@aljex.com -- http://www.aljex.com/bkw/ +++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++. filePro BBx Linux SCO Prosper/FACTS AutoCAD #callahans Satriani
- Previous message: Tony Lawrence: "Re: Cut him a break"
- In reply to: Ralph Scharping: "Re: Xenix Filesystem"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|
