home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!sun-barr!ames!agate!ucbvax!NSCVAX.PRINCETON.EDU!dragon
- From: dragon@NSCVAX.PRINCETON.EDU (Mighty Firebreather)
- Newsgroups: comp.os.vms
- Subject: RE: VMS/Un*x media exchange compatibilities
- Message-ID: <0095E1A1.5FB6FDA0.7100@nscvax.princeton.edu>
- Date: 25 Jul 92 16:51:13 GMT
- Sender: daemon@ucbvax.BERKELEY.EDU
- Distribution: world
- Organization: The Internet
- Lines: 74
-
-
- Chris Marzec <MARZEC%GUNBRF.BITNET@VTVM2.CC.VT.EDU> writes:
- >
- >We are interested in methods of data exchange between VMS and other systems
- >such as Un*x via magnetic tape media preferably DAT. At present we have been
- >able to distribute to non-VMS sites by using the following protocol for mag
- >tape and TK dirves:
- >
- > 1. rewrite VMS files from "variable length records / CR carriagecontrol"
- > to "fixed-length 80-byte rec / CR carriagecontrol"
- > 2. INITIALIZE tape device with some label
- > 3. MOUNT/FOREIGN/RECORDSIZE=80/BLOCKSIZE=4800 device:
- > =2048 for TK50, TK70 (and DAT) tape
-
- This probably overwrites your label. It certainly will result in
- no HDR1, HDR2, HDR3 labels being written and no trailer labels either! If
- you MOUNT /FOREIGN you do not get an ANSI labeled tape!
-
- > 4. COPY VMS-file device:
- >
- >In the past most all non-VMS sites were able to use this media. We assume that
- >this protocol yields an industry standard ANSI-labeled tape. However, lately
- >we attempted to distribute data via the above protocol except using a WangDAT
- >Model 1300 DDS DAT tape drive. The system used is a VAXstation 3100 running
- >V5.5. The customer was able to read/copy only the first file of the volume
- >after which no more data was accessible. We are curious to get feedback from
- >this protocol, specifically if it can be used for all non-VMS systems and if
- >it is an efficient procedure (time and space).
- >
- >Questions:
- > 1. is "VMS-ANSI-labeled" synonymous with "ANSI-labeled"
-
- Sort of. There are various revisions or levels of the ANSI standard
- and I believe that VMS uses level 3 while level 4 is the current version
- (or something like that).
-
- > 2. the DAT drive spec for blocksize is 36, is this significant because we
- > are able to read the tape inhouse when made using the above protocol
-
- A blocksize of 36????? 36 byte blocks??? I don't believe it! 36K
- maybe?
-
- > 3. does VMS file attributes affect the format of data when written to tape
- > ie. fixed-length vs. stream_LF (is one better than the other)
-
- For years, the one tape format that *any* system can read, has
- been 80 byte *unblocked* fixed length records with *no* labels! With the
- advent of streaming cartridge drives, this may no longer be true. It was
- never very efficient in terms of either speed or tape usage; it was just so
- simple and standard that even the brain damaged could read it.
-
- These days anybody, with the possible exceptions of IBM and Unix
- systems should be able to read an ANSI labeled tape.
-
- To make an ANSI labeled tape:
- $ ALLOCATE MUA0: TAPE
- $ INITIALIZE TAPE label
- $ MOUNT /BLOCKSIZE=xxxx /RECORDSIZE=yyy TAPE:
- $ COPY MYFILE.DAT TAPE:
- $
- If MYFILE.DAT has the fixed length record attribute, you will get
- fixed length records on tape. If the records are 80 bytes long, you will
- get 80 byte fixed length records. If yyy is less than the actual record
- size you will get an error. Blocksize defaults to 2048. You might want to
- use a larger value.
-
- *************************************************************************
- * *
- * Here, there be dragons! *
- * dragon@nscvax.princeton.edu *
- * *
- * Richard B. Gilbert *
- *************************************************************************
-
-