From:	AITGW::"moeller%gwdgv1.dnet.gwdg.de@VTVM2.CC.VT.EDU" "GWDGV1::MOELLER" 22-NOV-1991 04:46:40.72
To:	arisia::everhart
CC:	
Subj:	re^2: VMS BACKUP docs ...

Received: by AITGW.DECnet (utk-mail11 v1.5) ; Fri, 22 Nov 91 04:46:03 EST
Received: from vtvm2.cc.vt.edu by aitgw.ge.com (5.65/GE Gateway 1.4)
	id AA28486; Fri, 22 Nov 91 04:45:58 -0500
Message-Id: <9111220945.AA28486@aitgw.ge.com>
Received: from DGOGWDG1.BITNET by VTVM2.CC.VT.EDU (IBM VM SMTP V2R1)
   with BSMTP id 8632; Fri, 22 Nov 91 04:44:25 EST
Received: from dnet.gwdg.de by DGOGWDG1.BITNET (Mailer R2.07) with BSMTP id
 2053; Fri, 22 Nov 91 10:04:55 MEZ
Date:    Fri, 22 Nov 1991 10:04:53 +0100
From: "GWDGV1::MOELLER" <moeller%gwdgv1.dnet.gwdg.de@VTVM2.CC.VT.EDU>
To: arisia::everhart
Subject: re^2: VMS BACKUP docs ...
 
Glenn, your mail headers are still wrong ...
       v------- quotes required here --------------------------v
> From: Glenn Everhart 215 354 7610 Everhart%Arisia.dnet.ge.com
>  <EVERHART@ARISIA.dnet.ge.com>
 
I have to admit that I find making a virtual disk out of a backup/physical
tape rather weird, having not seen a single-tape backup for a long time ...
 
> [...] though I will have to find
> someone with the V3 utilities manual to find out how the CRC is done.
> Backup has two polynomials it uses...autodin and crc16..and I suspect
> I'll be able to isolate the correct one pretty fast.
 
I know why I salvaged said manual ... let me quote (but please don't expect me
to type in all of its appendix):
 
BBH$L_CRC	contains the CRC computed on the entire data block using
		the AUTODIN II CRC algorithm. The CRC field is set to 0
		during the computation. If the flag bit BBH$V_NOCRC is set,
		this field contains 0.
 
BBH$L_FLAGS	contains the following flag bits for the block:
			BBH$V_NOCRC	set if no CRC was computed
					for this block
		[no more bits mentioned - wjm]
 
BBH$W_CHECKSUM	contains a checksum of the block header computed by the
		CRC-16 algorithm.
 
According to the accompanying figure, the offsets ought to be
36. for CRC, 44. for FLAGS, and 254. for CHECKSUM.
 
>  As for turning the SDI files to .H, I suspect unsdl (another of Joe
> Meadows' goodies) seems likely to be helpful.
 
I may not know the exact file types - on the "fiche" one only finds
the SDL source, and lacking the SDL "compiler", UNSDL won't help.
There are no binary BACKUP definitions distributed with VMS.
 
> One day I'll need to see if I can build a client for
> your remote tape that will run on unix, since I've got an 8mm on a vax
> and an 8mm on a sun and would love to be able to do tape-tape
> copies.
 
Nice to hear that *someone* thinks about extending that beast.
Maybe I mentioned in the past that I did port the tape server once -
to our VM/CMS IBM DECnetted via INTERLINK (a rather unusual setup,
I guess) - and found it very slow due to the network protocol
(handshake) overhead. So far, I haven't learned anything about
un*x tape handling.
 
Wolfgang J. Moeller, GWDG, D-3400 Goettingen, F.R.Germany | Disclaimer ...
PSI%(0262)45050352008::MOELLER      Phone: +49 551 201516 | No claim intended!
Internet: moeller@gwdgv1.dnet.gwdg.de   | This space intentionally left blank.
