Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cornell!uw-beaver!rice!sun-spots-request
From: hsi!stpstn!aad@uunet.uu.net (Anthony A. Datri)
Newsgroups: comp.sys.sun
Subject: Re: How well do HP and SUN workstations mix
Message-ID: <2765@stpstn.UUCP>
Date: 8 Feb 89 01:59:09 GMT
References: <1216@mmm.UUCP> <34780@bbn.COM>
Sender: usenet@rice.edu
Organization: The Stepstone Corporation, Sandy Hook, CT
Lines: 36
Approved: Sun-Spots@rice.edu
Original-Date: 27 Jan 89 15:15:55 GMT
X-Sun-Spots-Digest: Volume 7, Issue 140, message 8 of 9

[[ Some people have told me that they can't reach this person at either
address he gives.  So the From line for this message is taken straight
from the "Berkeley" From line.  If it doesn't work, nothing will.  --wnl ]]

ames!mailrus!BBN.COM!fkittred@uunet.uu.net (Fletcher Kittredge) writes:
>schultz@mmm.serc.3m.com (John C Schultz) writes:
>>We would like to add an HP 9000/340 or 350 (w its own boot disk and tape
>>drive) to an existing small network of SUN stations - all SUN 3
>>somethings....How close is HP's UNIX to SunOS?...

>2) HP-UX is not close to Sun O/S at all.  However, if you are not a Unix
>hacker, you probably won't notice.  HP-UX has a pretty good compatibilty
>package, with the major, inexplicable exception of lack of Unix domain
>sockets.

We're running HP-UX 6.02.  The NFS seems to work fine with Suns, but it
can't rdump to a sun, nor does it have a version of lpr/lpd that'll print
to a remote machine.  I was once told that these work under 6.2 -- as soon
as I get it installed I'll be able to tell.

>>The only difficulty I can see is that complex software such as the GNU code
>>tends to require different flags to compile/link as compared to SUN systems.

HP-UX is a SysV-type OS, with some berkeleyisms added.  It does have
"more", and a csh that does escape completion.  Most of the things I've
tried to recompile under it worked fine by telling the Makefile that it
was SysV.

>3) Since the HPs are faster, we have our general purpose software live on
>them.

Our 9000/320 has 6 meg of memory, and an examination of the CPU board
leads me to believe that it's a 15Mhz 68020, but it still runs
significantly slower than one of our 3/50's.

Anthony A. Datri @SysAdmin(Stepstone Corporation) aad@stepstone.com stpstn!aad