Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!caen.engin.umich.edu!stealth
From: stealth@caen.engin.umich.edu (Mike Peltier)
Newsgroups: comp.sys.apollo
Subject: Re: DN1000 vs DN3500 file size
Message-ID: <45459a17.b617@bear.engin.umich.edu>
Date: 27 Aug 89 04:03 GMT
References: <2535@orion.cf.uci.edu> <3187@wasatch.utah.edu>
Reply-To: stealth@caen.engin.umich.edu (Mike Peltier)
Organization: University of Michigan Engineering:  Ann Arbor, Michigan
Lines: 24

In article <2535@orion.cf.uci.edu> roode@hydra.cf.uci.edu (Dana Roode) writes:
>We have a DN10000 running SR10.0 and a DN3500 running SR10.1.  Why does a
>file occupying 992 blocks (according to ls -s) on the DN1000 take only
>247 blocks (also according to ls -s) when copied to the DN3500? 
>
> Dana Roode
> UCI

The manual page for 'ls' indicates that the -s option prints the size
of the file in kilobytes.  A file containing "This is a test." showed
up as " 16 -rw-r--r--" on our DN10k, but as "  1 -rw-r--r--" on a DN3.5k.
I think I recall this having something to do with bigger block sizes
on the 10k -- 16kbyte blocks rather than 1kbyte blocks?  That would be
the natural conclusion given the above information...

-	-	-	-	-	-	-	-	-
Michael V. Peltier         | Computer Aided Engineering Network      
1420 King George Blvd.     | University of Michigan,  Ann Arbor       
Ann Arbor, MI  48104-6924  |    stealth@caen.engin.umich.edu          
-- 
-	-	-	-	-	-	-	-	-
Michael V. Peltier         | Computer Aided Engineering Network      
1420 King George Blvd.     | University of Michigan,  Ann Arbor       
Ann Arbor, MI  48104-6924  |    stealth@caen.engin.umich.edu