19 May
2005
19 May
'05
10:27 a.m.
Yes, Andy Goldstein (or whoever, but I think Andy wrote the first
version) really did a nice job in VMS BACKUP. I once wrote a program
to read such tapes on a Unix system and needed to learn the
techniques used. Very nice. Indeed an entire /BLOCK_SIZE block of
data within one /GROUP_SIZE could be completely corrupted on tape,
yet BACKUP could correctly restore it.
John
--
Do you still have this (I assume it's different from the various C
versions of "VMSBACKUP"
floating around)
That program does not do a very good job of recovering from archive
errors.
There appears to be very little written down about the variations in
the VMS Backup
format as well.
19 May
19 May
10:38 a.m.
Al Kossow wrote:
There appears to be very little written down about the
variations in
the VMS Backup format as well.
I think the last time the format was documented was in
an appendix to one of the VAX/VMS V3 manuals. Not much
has (apparently) changed since then.
You can get backup to spit out a fair amount of info
by adding /ANALYZE to a BACKUP/LIST command.
And the source listings are available.
So the stuff may not be around in an easy-to-program
format, but the info does exist.
Antonio
--
---------------
Antonio Carlini arcarlini at iee.org
12:46 p.m.
Al,
At 10:27 AM -0700 5/19/05, Al Kossow wrote:
Do you still have this (I assume it's different
from the various C
versions of "VMSBACKUP"
floating around)
Yes, I think I could probably find it or some generation of it with
some work. And yes, it was different from VMSBACKUP that made the
rounds. I never published that bit of work.
That program does not do a very good job of recovering
from archive errors.
Don't know how that program works. As I remember, you need to keep
/GROUP_SIZE buffers laying around at all times in order to perform
the verification and error correction step.
There appears to be very little written down about the
variations in
the VMS Backup
format as well.
Quite true, but...
At 6:38 PM +0100 5/19/05, Antonio Carlini wrote:
I think the last time the format was documented was in
an appendix to one of the VAX/VMS V3 manuals. Not much
has (apparently) changed since then.
You can get backup to spit out a fair amount of info
by adding /ANALYZE to a BACKUP/LIST command.
And the source listings are available.
Indeed that is the way I discovered things. At the time I had a
fiche reader on my desk at work (as well as a machine readable
version of the V3 listings) and used /ANALYZE liberally.
I still have some of the paper documents from V3 that specified the
format. Would you like those scanned?
Just out of curiosity, is this program something you could use Al?
John
20 May
20 May
2:31 p.m.
John A. Dundas III wrote:
Indeed that is the way I discovered things. At the
time I had
a fiche reader on my desk at work (as well as a machine
readable version of the V3 listings) and used /ANALYZE
liberally.
I still have some of the paper documents from V3 that
specified the format. Would you like those scanned?
I didn't see anyone else say yes, so I'll say it. If
you have a VMS V3 docset, it would be nice to see it scanned.
V6 (or thereabouts) onwards is available in PDF and HTML.
V5 (ish) onwards (say 1993+) is available in Bookreader
format and there are programs around that can display
and convert that.
V4 and before exists (AFAIK) only on paper (and maybe
microfiche).
Antonio
--
---------------
Antonio Carlini arcarlini at iee.org
4:42 p.m.
V4 and before exists (AFAIK) only on paper
By the time of V4, most all of the manuals were being maintained
internally in VAX DOCUMENT. It's fairly standard SGML, although
there were some specialized macros used just for DEC-ish look and
feel at the presentation.
'
Those that weren't VAX DOCUMENT were DSR (Runoff).
On their way to paper the VAX DOCUMENT source was turned into
postscript. Some layered products came with postscript release
notes and electronic documentation (usually bookreader but
sometimes flat text and sometimes postscript.)
Tim.
21 May
21 May
2:32 a.m.
shoppa_classiccmp at trailing-edge.com wrote:
All true, but I meant that the V4 (an ealier) docs
do not exist in electronic format outside of DEC.
These days, they may not exist in electronic format
_inside_ either :-)
Antonio
--
---------------
Antonio Carlini arcarlini at iee.org
V4 and before
exists (AFAIK) only on paper
By the time of V4, most all of the manuals were being
maintained internally in VAX DOCUMENT. It's fairly standard
SGML, although there were some specialized macros used just
for DEC-ish look and feel at the presentation.
'
Those that weren't VAX DOCUMENT were DSR (Runoff).
7155
days inactive
7157
days old
5 comments
4 participants
participants (4)
-
a.carlini＠ntlworld.com -
aek＠bitsavers.org -
dundas＠caltech.edu -
shoppa_classiccmp＠trailing-edge.com