BSD vs ULTRIX vs SysV

[D0430 at PUCC.BITNET.UUCP]

In article <2123 at brl-smoke.ARPA>, @CSNET-RELAY.arpa, at tufts.csnet:kjs at tufts.c (kjs) writes:
 
>
>I administer the UN*X side of the academic computing facility here at
>Tufts University.  I am currently running two VAX 780's under 4.2BSD.
>The arrival of 4.3BSD has prompted my management to take a look at alternates
>to the BSD UN*X..........
 
I have an application in which I had to make a lot of kernel modifications
to Ultrix 1.1 on a uvaxII.  The assurance that it was 'identical to 4.2'
was just not so.  DEC has done a lot work in various areas to make it more
robust.  For example: it writes partition tables on disks; the uda driver is
a lot bigger and has much more error correction; it uses 4.3 inode caching.
This latter fact caused two of my students several months of grief as they
tried to install RFS.  They succeeded but the difficulties were caused by these
changes.  I am, however, quite happy with it now.  The modifications to run
on the uvaxII were really cute--lots of brain damage in locore.s but
the drivers and higher level kernel routines are virtually the same for uvaxII
and bigger vaxen.  The source is quite cheap to Universities.  We made no
use of DEC support.  I don't know much about 1.2 or 4.3, but I was pleased
with what I saw in 1.1.