9 Aug
2006
9 Aug
'06
10:48 a.m.
Interesting note, a lot of these tapes have hundreds of 0-length files
on them; these seem to be placed between files with actual data in
them. I have no idea if this was a standard practice (maybe
ComputerVision hadn't heard of "tar" :)).
--
What you are seeing are file marks, which will return a zero length file if
you just 'dd' them. I haven't looked at CV tapes in a LONG time, but from
memory they are modified SunOS 3 - 4 distributions. What you should find is
a table of contents at the front, some stand-alone utils, a dump file of the
system, and lots of tarballs for the optional parts of the system.
9 Aug
9 Aug
10:57 a.m.
Yeah -- the tapes I have that are just rebranded Sun distributions (SunOS
and a few others) follow the pattern you mention. However, some of the
other tapes that have pure CV stuff (a lot of the CADD stuff) have hundreds
of files (one of them has over 800), most of which are between 10 and 100k,
all of them separated by these 0-length file marks.
I guess I'll know if I've archived these correctly when I get to the point
where I can install them :).
Thanks,
Josh
Al Kossow writes:
Interesting note, a lot of these tapes have hundreds
of 0-length files
on them; these seem to be placed between files with actual data in
them. I have no idea if this was a standard practice (maybe
ComputerVision hadn't heard of "tar" :)).
--
What you are seeing are file marks, which will return a zero length file if
you just 'dd' them. I haven't looked at CV tapes in a LONG time, but from
memory they are modified SunOS 3 - 4 distributions. What you should find is
a table of contents at the front, some stand-alone utils, a dump file of the
system, and lots of tarballs for the optional parts of the system.
12:40 p.m.
On 8/9/2006 at 10:48 AM Al Kossow wrote:
Interesting note, a lot of these tapes have hundreds of
0-length files
on them; these seem to be placed between files with actual data in
them. I have no idea if this was a standard practice (maybe
ComputerVision hadn't heard of "tar" :)).
Perhaps nothing more than a double filemark or its equivalent? Very comon
in 7- and 9-track tapes.
Cheers,
Chuck
12:51 p.m.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: <cctalk at classiccmp.org>
Sent: Wednesday, August 09, 2006 9:40 PM
Subject: Re: ComputerVision CADDStation tapes now available (sorta)
On 8/9/2006 at 10:48 AM Al Kossow wrote:
The standard for IBM labelled tapes is that there is 1 tapemark between
files, and 2 tapemarks to indicate that the last file has been read.
As there are two tapemarks between files, how can you then detect
end-of-data ? 3 or 4 tapemarks ?
Nico
Interesting note, a lot of these tapes have
hundreds of 0-length files
on them; these seem to be placed between files with actual data in
them. I have no idea if this was a standard practice (maybe
ComputerVision hadn't heard of "tar" :)).
Perhaps nothing more than a double filemark or its equivalent? Very comon
in 7- and 9-track tapes.
Cheers,
Chuck
1:07 p.m.
On 8/9/2006 at 9:51 PM Nico de Jong wrote:
The standard for IBM labelled tapes is that there is 1
tapemark between
files, and 2 tapemarks to indicate that the last file has been read.
As there are two tapemarks between files, how can you then detect
end-of-data ? 3 or 4 tapemarks ?
Depends on the vendor. Some used more than 2 filemarks. From a hardware
(and software) point of view, there's nothing to prevent it. If you think
about it, it makes sense. A corrupted filemark will cause a tape drive to
skip one more record than requested. A double filemark adds a bit of
redundancy.
Particularly in tapes, IBM's practices weren't universal; nor, for that
matter did everyone adhere to ANSI standards. You can probably still find
plenty of old tapes without labels, with strange block sizes, in mixed
densities, etc. 200 bpi 7-track EVEN parity tapes were even used at one
time (There was no way to write a binary zero character). Tape conversion
between vendors used to be a big deal 30+ years ago.
Chuck
6710
days inactive
6710
days old
4 comments
4 participants
participants (4)
-
aek＠bitsavers.org -
cclist＠sydex.com -
derschjo＠msu.edu -
nico＠farumdata.dk