Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!ut-emx!ibmchs!auschs!awdprime!testsys.austin.ibm.com!mbrown
From: mbrown@testsys.austin.ibm.com (Mark Brown)
Newsgroups: comp.unix.aix
Subject: Re: 1/4" tape and record sizes
Message-ID: <8164@awdprime.UUCP>
Date: 4 Jun 91 15:50:54 GMT
References: <1991Jun3.024011.387@odi.com>
Sender: news@awdprime.UUCP
Reply-To: mbrown@testsys.austin.ibm.com (Mark Brown)
Organization: IBM Austin, TX
Lines: 34

benson@odi.com (Benson I. Margulies) writes:
|Path: awdprime!auschs!panews!uunet!uunet!odi!benson
|From: benson@odi.com (Benson I. Margulies)
|Newsgroups: comp.unix.aix
|Subject: 1/4" tape and record sizes
|Message-ID: <1991Jun3.024011.387@odi.com>
|Date: 3 Jun 91 02:40:11 GMT
|Reply-To: benson@odi.com (Benson I. Margulies)
|Organization: Object Design Inc., Burlington, MA
|Lines: 12
|
|My 320 has the usual IBM/tandberg 1/4 inch polymorphously perverse
|tape drive. (anyone know where to get a drive like this that writes
|both 60 and 150 MB for a sun?)
|
|In smit, its configured for 512 byte blocks. I write tapes with backup
|-i (for consumption of installp). tcopy, documented to report on
|record sizes, swears that the tape blocks are 65536 bytes long.
|I think it's lying. Can anyone convince me otherwise?

My tape expert says:
  The record sizes reported on by tcopy are not physical tape record sizes 
but rather the size of the reads and/or writes done by tcopy.  

Sounds to me like there is a documentation problem, or backup isn't
writing things out in the block size you think it is (when in -i).

I'm adding it to my list-o-things.


      DISCLAIMER: Any personal opinions stated here are just that.
Mark Brown       IBM PSP Austin, TX. | Don't you feel more like
(512) 823-3741   VNET: MBROWN@AUSVMQ | you do now
MAIL: mbrown@testsys.austin.ibm.com  | than you did when you started?