25 Jan
2005
25 Jan
'05
11:48 p.m.
Jules Richardson wrote:
Looking at the raw disk image I've saved, I've worked out where the
partition table is within the image (it's at offset 0x280). Most of the
stuff preceding it is zeros, except for a few flags and one place that
seems to be total size of the disk in blocks *including partition table
data*.
If you have a saved image you should be able to try and mount it using a
'loopback' file system on linux. No doubt that's what you tried.
did you try any of the bsd parition programs, liked parted? who knows.
All I remember is that sometimes we had to boot from a floppy and
humm... that ran some sort of fsck? it's been a while. We never
replaced a hard disk or tried to fool with one.
I'd guess I'd be suprised if the file system was anything other than
system 5 ufs. As I recall those systems were pretty vanilla unix wise.
I don't think they had any bsd-isms in them (but I may be wrong).
it might be interesting to hack up a copy of an old sys v fsck and open
the image in read-only mode. that way you could fool around with
various offsets and try to find the super block.
once you know where the partition starts, and assuming you did a block
copy under unix (so the blocks are in the right order), you might be
able to grab an old copy of "sash" or something like that and get it to
read the file system. at least then you could "ls".
just some random notions.
In the end, I think I'd write a program to transform the image into a
new image which had a modern partition map pointing to the super block
and then get linux to mount it. easy for me to say :-)
-brad