Path: utzoo!attcan!uunet!ncrlnk!ncr-sd!hp-sdd!ucsdhub!isg100!jack!nusdhub!rwhite
From: rwhite@nusdhub.UUCP (Robert C. White Jr.)
Newsgroups: comp.sys.att
Subject: Re: 6386 and Dead Priam Hard Drive
Message-ID: <1253@nusdhub.UUCP>
Date: 17 Nov 88 00:03:13 GMT
References: <8300001@gistdev>
Organization: National University, San Diego
Lines: 28

in article <8300001@gistdev>, jeff@gistdev says:
<> Our problem is this: the disk died.  It looks like an area under "/" went
<> bad.  We can boot to a point where most drivers are loaded and then we get
<> a panic trap.  The VTOC seems to be scrozzed as well. 
<> 
<> As far as we can tell, /usr and /usr2 are ok.  We have some VERY important
<> source code under /usr2 that for some stupid reason (cost of media) wasn't
<> backed up for over a month and a half.  Considerable development went on
<> during that time and now we have a deadline to release the program and we
<> can't get at it.
<> 
<> Is it possible to mount the /usr2 partition on another machine running the
<> same configuration? If so, how do we do it?

If you can get to some machine running the same revision of the software
then use the standard "how to break into a pc running unix" technique.

Make a bootable floppy on some machine (dosn't matter what machine.)
put on a stripped down kernel (e.g. loose the device drivers where
possible) as much of the /dev directory as you can scavange (off an
old backup perhaps) /etc/passwd /etc/group /etc/init ... and
/etc/mount.  /bin/sh /bin/login /bin/cpio and the like.

Having made this "scavengeing diskette" you should be able to mount
your old /usr /and /usr2 and archive them onto another source.


Rob.