From: ciech...@cis.pfm-mainz.de (Ingo Ciechowski)
Subject: A/UX future and platform ??
Date: 1995/06/13
Message-ID: <ciechowski-1306951925330001@zeus.cis.pfm-mainz.de>#1/1
X-Deja-AN: 104307623
sender: ne...@news.PFM-Mainz.DE (Owner of the news)
organization: CIS private
newsgroups: comp.unix.aux,comp.sys.mac.oop.macapp3,
comp.sys.mac.programmer.help,comp.sys.mac.system
I need to find out what hardware can be used to install A/UX and whether
or not that OS will be supported in the future.
Any comments please directly to ciech...@cis.pfm-mainz.de .
Thanks,
Ingo
ciech...@cis.pfm-mainz.de
From: j...@aggroup.com (Jon Stevens)
Subject: Re: A/UX future and platform ??
Date: 1995/06/15
Message-ID: <jon-1506951111180001@jon.aggroup.com>#1/1
X-Deja-AN: 104719974
references: <ciechowski-1306951925330001@zeus.cis.pfm-mainz.de>
content-type: text/plain; charset=US-ASCII
organization: AG Group
mime-version: 1.0
newsgroups: comp.unix.aux,comp.sys.mac.oop.macapp3,
comp.sys.mac.programmer.help,comp.sys.mac.system
In article <ciechowski-13...@zeus.cis.pfm-mainz.de>,
ciech...@cis.pfm-mainz.de (Ingo Ciechowski) wrote:
> I need to find out what hardware can be used to install A/UX and whether
> or not that OS will be supported in the future.
>
> Any comments please directly to ciech...@cis.pfm-mainz.de .
>
> Thanks,
>
> Ingo
> ciech...@cis.pfm-mainz.de
according to MacLeak a couple of weeks ago...a/ux is officially dead...and
according to my apple friends...if it is not already dead...it will be
soon...now we need to re-start that thread on making a/ux free for general
kernal tweaking!!! ;)
-jon stevens
From: chu...@plaidworks.com (Chuq Von Rospach)
Subject: Re: A/UX future and platform ??
Date: 1995/06/22
Message-ID: <3sc2tr$pdj@news.znet.net>#1/1
X-Deja-AN: 104991166
references: <ciechowski-1306951925330001@zeus.cis.pfm-mainz.de>
<jon-1506951111180001@jon.aggroup.com>
organization: zNET
newsgroups: comp.unix.aux,comp.sys.mac.oop.macapp3,
comp.sys.mac.programmer.help,comp.sys.mac.system
j...@aggroup.com (Jon Stevens) writes:
>according to MacLeak a couple of weeks ago...a/ux is officially dead...
Um, hasn't this been said here more than once? you don't believe it until
it's in MacWeek?
>now we need to re-start that thread on making a/ux free for general
>kernal tweaking!!! ;)
Until the people who license Unix source code decide it's okay for Unix to
be freely distributed, it's not gonna happen. You're welcome to go talk with
them about it. Apple (frankly) can't even consider it.
--
Chuq Von Rospach -=- Author, Software Gnome, and Internet Guy
Member Science Fiction & Fantasy Writers of America
Plaidworks Consulting: Writing, Editing, and Damage Control
<http://plaidworks.com/chuqui/> * GEnie:chuq * CIS:75141,1242
chu...@plaidworks.com * ch...@abs.apple.com * AOL:chuqui
From: bri...@puma.bevd.blacksburg.va.us (Allen Briggs)
Subject: Re: A/UX future and platform ??
Date: 1995/06/24
Message-ID: <3sh2f4$sgm@solaris.cc.vt.edu>#1/1
X-Deja-AN: 104991075
references: <ciechowski-1306951925330001@zeus.cis.pfm-mainz.de>
<jon-1506951111180001@jon.aggroup.com> <3sc2tr$pdj@news.znet.net>
organization: Home, Blacksburg, Virginia
nntp-posting-user: briggs
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
[ comp.sys.mac.oop.macapp3 removed from Newsgroups: ;-) ]
In article <3sc2tr$p...@news.znet.net>,
Chuq Von Rospach <chu...@plaidworks.com> wrote:
>Um, hasn't this been said here more than once? you don't believe it until
>it's in MacWeek?
There are a lot of things said in MacWeek that remind me of, say, the
National Enquirer, too. I've long-since stopped reading it for anything
but the Marketplace section... ;-)
>Until the people who license Unix source code decide it's okay for Unix to
>be freely distributed, it's not gonna happen. You're welcome to go talk with
>them about it. Apple (frankly) can't even consider it.
It should be possible for Apple to release the non-AT&T parts of
A/UX--on some kind of basis. Granted, it would be a lot of effort
to implement even the non-MacOS parts of A/UX on another system (like
NetBSD, Linux, Mach, whatever), but it would certainly make it easier
than the alternative. Failing that, it would be nice if Apple could
release internal documentation (even if it's just on discontinued
systems) to a couple of key people--with the understanding that code
developed from that documentation would be freely redistributable...
I don't see how this much is illegal on Apple's part, nor how it could
hurt them--especially if the arrangment didn't allow for free
redistribution of the documentation (well, it would take someone a bit
of time to assemble the documentation--from what I understand, it's not
in any centralized place ;-). Really, all I want it a copy of an A/UX
or MacOS "kernel" and device hacker's bookshelves... B-)
-allen
--
Allen Briggs - end killing - allen....@bev.net ** MacBSD == NetBSD/mac68k **
From: aru...@freenet.vcu.edu
Subject: Re: A/UX future and platform ??
Date: 1995/06/25
Message-ID: <3sitlo$i5v@freenet.vcu.edu>#1/1
X-Deja-AN: 105041284
distribution: world
organization: Central Virginia's Free-Net
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
> [2]Send email reply to author: Allen Briggs
> References:
> [3]<ciechowski-13...@zeus.cis.pfm-mainz.de>
> [4]<jon-150695...@jon.aggroup.com>
> [5]<3sc2tr$p...@news.znet.net>
>[ comp.sys.mac.oop.macapp3 removed from Newsgroups: ;-) ]
>
>In article [6]<3sc2tr$p...@news.znet.net>,
>Chuq Von Rospach [7]<chu...@plaidworks.com> wrote:
>>Um, hasn't this been said here more than once? you don't believe it until
>>it's in MacWeek?
>
>There are a lot of things said in MacWeek that remind me of, say, the
>National Enquirer, too. I've long-since stopped reading it for anything
>but the Marketplace section... ;-)
>
>>Until the people who license Unix source code decide it's okay for Unix to
>>be freely distributed, it's not gonna happen. You're welcome to go talk with
>>them about it. Apple (frankly) can't even consider it.
Unix source code IS freely distributable: check out Linux: Unix
for DOS machines.... The entire source tree and ALL utilities
are available in source, and, according to the license (GNU
general public license, by the free software foundation), must
be available in source, and everything derived from the source
must fall under the GPL.... Sadly, It's apple who's standing in
the way...
>It should be possible for Apple to release the non-AT&T parts of
>A/UX--on some kind of basis. Granted, it would be a lot of effort
>to implement even the non-MacOS parts of A/UX on another system (like
>NetBSD, Linux, Mach, whatever), but it would certainly make it easier
>than the alternative. Failing that, it would be nice if Apple could
>release internal documentation (even if it's just on discontinued
>systems) to a couple of key people--with the understanding that code
>developed from that documentation would be freely redistributable...
Of course apple could do this: they just won't: they're afraid
it will involve leaking some hidden, ultra-secret part of the
Mac OS...
>I don't see how this much is illegal on Apple's part, nor how it could
>hurt them--especially if the arrangment didn't allow for free
>redistribution of the documentation (well, it would take someone a bit
>of time to assemble the documentation--from what I understand, it's not
>in any centralized place ;-). Really, all I want it a copy of an A/UX
>or MacOS "kernel" and device hacker's bookshelves... B-)
It's NOT illegal: like I said, Unix is for the most part freely
available: i don't even know how much is AT&T derived. Anyway,
for those who are unaware, there is a project to port Linux to
the Mac/68K, and to the powerPC. On the mac side, apple won't
release any info necessary for implementation except with an
NDA, which wouldn't work with the intended license...
for interested readers:
4.4 Mac
-------
A Mac version is under construction but not yet finished. Main (sole?)
developer is George Andre (dat9...@ludat.lth.se). He has written a FAQ
about Linux/68k-Mac which the interested reader should get first. It is
available as ftp://www.ibg.uu.se/pub/maclinux/FAQ.txt. There is a mailing
list too (see: 6.3 Mailinglists).
This is from a document (i'm not quite sure which) regarding
680x0 ports of Linux for Amigas:
Another curious project for a C/C++ programmer with a lot of
free time would be to implement 80x86 Linux for Softwindows
(need version 2, 486-emulation) or for Dos/Windows x86
cards.... All that would be needed, I should think, would be
device drivers for HD/Keyboard/Monitor/Floppy/Mouse/etc...
well, bye now:
algis rudys
From: ch...@adi.com (Franklin Chen)
Subject: Re: A/UX future and platform ??
Date: 1995/06/26
Message-ID: <chen-2606950011360001@pm035-01.dialip.mich.net>#1/1
X-Deja-AN: 105041285
distribution: world
references: <3sitlo$i5v@freenet.vcu.edu>
organization: Applied Dynamics International
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
In article <3sitlo$i...@freenet.vcu.edu>, aru...@freenet.vcu.edu wrote:
> Unix source code IS freely distributable: check out Linux: Unix
> for DOS machines.... The entire source tree and ALL utilities
> are available in source, and, according to the license (GNU
> general public license, by the free software foundation), must
> be available in source, and everything derived from the source
> must fall under the GPL.... Sadly, It's apple who's standing in
> the way...
No offense intended, but you have missed the plain facts: UNIX is hardly
freely distributable. That's the whole reason for the "unencumbered" BSD
efforts, the whole reason Richard Stallman decided to start GNU to come up
with a UNIX-compatible free system, etc. Linux is not UNIX, but a UNIX
clone.
> It's NOT illegal: like I said, Unix is for the most part freely
> available: i don't even know how much is AT&T derived. Anyway,
> for those who are unaware, there is a project to port Linux to
> the Mac/68K, and to the powerPC. On the mac side, apple won't
> release any info necessary for implementation except with an
> NDA, which wouldn't work with the intended license...
Have you never read this in every book and paper on UNIX(TM) you've
encountered??
"UNIX is a trademark of AT&T Bell Laboratories"
(That was the old days; now it's USL or Novell or whatever--I don't look
at the footnote following every mention of "UNIX" in every piece of
documentation any more.)
--
Franklin Chen ch...@adi.com
Applied Dynamics International
3800 Stone School Road
Ann Arbor, MI 48108-2499
From: aru...@freenet.vcu.edu
Subject: Re: A/UX future and platform ??
Date: 1995/06/27
Message-ID: <3spodo$a2s@freenet.vcu.edu>#1/1
X-Deja-AN: 105122883
distribution: world
organization: Central Virginia's Free-Net
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
> [2]Send email reply to author: Franklin Chen
>In article [4]<3sitlo$i...@freenet.vcu.edu>, aru...@freenet.vcu.edu wrote:
>
>> Unix source code IS freely distributable: check out Linux: Unix
>> for DOS machines.... The entire source tree and ALL utilities
>> are available in source, and, according to the license (GNU
>> general public license, by the free software foundation), must
>> be available in source, and everything derived from the source
>> must fall under the GPL.... Sadly, It's apple who's standing in
>> the way...
>
>No offense intended, but you have missed the plain facts: UNIX is hardly
>freely distributable. That's the whole reason for the "unencumbered" BSD
>efforts, the whole reason Richard Stallman decided to start GNU to come up
>with a UNIX-compatible free system, etc. Linux is not UNIX, but a UNIX
>clone.
OK, so does all Unix have to be from AT&T?? Are SunOS, A/UX,
AIX, HP UX, or Ultrix also "Unix Clones??" The fact is, Linux is
entirely Unix-compatible (whatever that means), it features the
same set of utilities, the same basic kernel functionality,
multiple user support, networking utilities, programming
support, etc etc. In fact, Linux is as Unix as the
afformentioned OS's: the only real difference is that Linux is
free.... I think Unix has been around long enough to be a
reference to a type of operating system, and not just an AT&T
copyright word.
>> It's NOT illegal: like I said, Unix is for the most part freely
>> available: i don't even know how much is AT&T derived. Anyway,
>> for those who are unaware, there is a project to port Linux to
>> the Mac/68K, and to the powerPC. On the mac side, apple won't
>> release any info necessary for implementation except with an
>> NDA, which wouldn't work with the intended license...
>
>Have you never read this in every book and paper on UNIX(TM) you've
>encountered??
>
> "UNIX is a trademark of AT&T Bell Laboratories"
>
>(That was the old days; now it's USL or Novell or whatever--I don't look
>at the footnote following every mention of "UNIX" in every piece of
>documentation any more.)
See above: Unix no longer refers to only The one brand by the
one copyright owner: it refers to the type of operating system.
I would even suggest that the copyrights owner would concede
that, considering that the common "misnomer" if you will, i
that the Internet is mostly (almost entirely) Unix, when only a
small part of the "Unix" machines thereon are by the respective
"Unix" owner: most are SunOS, and many more are HP or AIX, and
even BSD. Relatively few are SVR4, UnixWare, or whatever the
Owner of "Unix" makes....
bye now
algis rudys
--
Algis Rudys | "Let be be finale of seem
aru...@freenet.vcu.edu | The only emperor is the
The Governor's School | emperor of ice-cream"
Richmond, VA | -- Wallace Stevens
From: mit...@colorado.edu (Ken Mitton)
Subject: Re: A/UX future and platform ??
Date: 1995/06/28
Message-ID: <mittonk-2806951238280001@tele-anx0131.colorado.edu>#1/1
X-Deja-AN: 105253324
distribution: world
references: <3spodo$a2s@freenet.vcu.edu>
organization: University of Colorado at Boulder
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
>
> OK, so does all Unix have to be from AT&T?? Are SunOS, A/UX,
> AIX, HP UX, or Ultrix also "Unix Clones??" The fact is, Linux is
> entirely Unix-compatible (whatever that means), it features the
> same set of utilities, the same basic kernel functionality,
> multiple user support, networking utilities, programming
> support, etc etc. In fact, Linux is as Unix as the
> afformentioned OS's: the only real difference is that Linux is
> free.... I think Unix has been around long enough to be a
> reference to a type of operating system, and not just an AT&T
> copyright word.
>
Wrong-- SunOS, A/UX, AIX, HP UX, and Ultrix all use AT&T code, and so are
UNIX. Linux and the HURD (future end product of the GNU project) use no
AT&T code, and are therefore not UNIX, even if you couldn't tell the
difference by looking at them.
--Ken Mitton
mit...@colorado.edu
From: aru...@freenet.vcu.edu
Subject: Re: A/UX future and platform ??
Date: 1995/06/30
Message-ID: <3t0233$2jb@freenet.vcu.edu>#1/1
X-Deja-AN: 105253406
distribution: world
organization: Central Virginia's Free-Net
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
> [2]Send email reply to author: Ken Mitton
> References:
> [3]<3spodo$a...@freenet.vcu.edu>
>>
>> OK, so does all Unix have to be from AT&T?? Are SunOS, A/UX,
>> AIX, HP UX, or Ultrix also "Unix Clones??" The fact is, Linux is
>> entirely Unix-compatible (whatever that means), it features the
>> same set of utilities, the same basic kernel functionality,
>> multiple user support, networking utilities, programming
>> support, etc etc. In fact, Linux is as Unix as the
>> afformentioned OS's: the only real difference is that Linux is
>> free.... I think Unix has been around long enough to be a
>> reference to a type of operating system, and not just an AT&T
>> copyright word.
>>
>
>Wrong-- SunOS, A/UX, AIX, HP UX, and Ultrix all use AT&T code, and so are
>UNIX. Linux and the HURD (future end product of the GNU project) use no
>AT&T code, and are therefore not UNIX, even if you couldn't tell the
>difference by looking at them.
Well, according to a post regarding SVR4 on comp.security.unix,
Linux is very similar to SVR4 w/r to source.... In fact, among
the utilities required for Linux operation is "SystemV-style
init, which control booting, shutdowns, and runlevels....
also, BSD is as "Unix" as SysV. To my understanding, Berkeley
and AT&T developed Unix at about the same time, but AT&T got
the pattent first.
regarding Mario Klebsch's comments:
I have never used any other Unix systems for high-end system
administration. I will be doing so on a linux box. I know it IS
possible to boot Linux into single-user mode. However, you
cannot (to my understanding, leastways)take the system from
multi-to single-user mode without rebooting. However,
notwhithstanding names, The are utilities available on Linux to
check FS's, make and restore backups, take kernel dumps, etc
etc. However, some name and specific function problems come
from the fact that Linux has to support various different FS's,
such as MSDOS, Extended, 2nd Extended, Minix, etc, and thus
needs different versions of mkfs and fsck....
i will say that perchance to someone familiar with "commercial"
Unix or BSD, Linux may be somewhat disorienting. However, (1)
all the tools you need ARE, in fact there, and (2) Linux
functions in just the same way as Unix does, notwhithstanding
what the utilities are called
bye now
algis rudys
--
Algis Rudys | The Governor's School
aru...@freenet.vcu.edu | Richmond, VA
----------------------------------------------
"April is the cruellest month, breeding
Lilacs out of the dead land, ..."
From: m...@rob.cs.tu-bs.de (Mario Klebsch DG1AM)
Subject: Re: A/UX future and platform ??
Date: 1995/06/30
Message-ID: <mkl.804524239@duesentrieb>#1/1
X-Deja-AN: 105410024
distribution: world
references: <3t0233$2jb@freenet.vcu.edu>
organization: TU Braunschweig, Informatik (Bueltenweg), Germany
newsgroups: comp.unix.aux,comp.sys.mac.programmer.help,comp.sys.mac.system
aru...@freenet.vcu.edu writes:
>Well, according to a post regarding SVR4 on comp.security.unix,
>Linux is very similar to SVR4 w/r to source.... In fact, among
>the utilities required for Linux operation is "SystemV-style
>init, which control booting, shutdowns, and runlevels....
About a week ago, I had a look at a friends Linux system. I wanted to
tell him, how to start xdm on startup. He told me about /etc/inittab,
and so I told him to write a script for /etc/init.d and to create
links in /etc/rcS.d and /etc/rc2.d. But unfortunatly, these
directories were missing. There were file in /etc/rc.d instead, that
were organised like the BSD rc files. Than I found a comment in
/etc/inittab that initstate 6 is for X operation. On all SYSV
machines, I had to use, initstate 6 is for rebooting the system.
These are my comments to the Linux SystemV-style. Perhaps this is
already changed, but this really is typical for my experience with
Linux.
>also, BSD is as "Unix" as SysV. To my understanding, Berkeley
>and AT&T developed Unix at about the same time, but AT&T got
>the pattent first.
Surely, they both are UNIX. And as an experienced UNIX operator, I had
contact with both of them.
>...
>i will say that perchance to someone familiar with "commercial"
>Unix or BSD, Linux may be somewhat disorienting. However, (1)
>all the tools you need ARE, in fact there, and (2) Linux
>functions in just the same way as Unix does, notwhithstanding
>what the utilities are called
I often was asked as an experienced sysop, to help people with
Linux. And when I don't know the system, I have to relie on the facts
I know about UNIX. And lots of these facts are not portable to Linux,
when you really are in trouble. It is no help, that there are man
pages, when in single user mode, /usr is not mounted and the man pages
are inaccesable.
As long as you don't know the magic word to mount /usr, you are
lost. This happend to me, when I helped a friend to repartition the
boot disk. I did not touch disk with /usr and /home, I made a backup
of the root partition and then I repartitioned the disk. After loading
the memory version of Linux, I was in trouble. The only thing, that
helped me, was the documentation, that was printed prior to the
installation.
But all I want to say is, this is a simple task. I did this several
times in our Suns running SunOS4 and Solaris. I did these things on
other UNIX systems, too. A really good example of how it can be was
the installation of freeBSD, I did for a friend about 4 month ago.
Perhaps, my bad impression on Linux is based on the fact, I was almost
always called, when people had problems, not when it worked fine. Then
I found, almost all my knowledge on UNIX useless on this system.
I thing, one of the best tests for a system is, how it behaves, when
facing a desaster. Can it still boot to a point, where I can fix it?
Can I fix if with my basic knowledge of the system? What tools can I
use to analyze the situation?
73, Mario
--
Mario Klebsch, DG1AM, M.Kl...@tu-bs.de +49 531 / 391 - 7457
Institut fuer Robotik und Prozessinformatik der TU Braunschweig
Hamburger Strasse 267, 38114 Braunschweig, Germany
From: da...@space.mit.edu (John E. Davis)
Subject: Linux is not Unix.
Date: 1995/07/14
Message-ID: <3u647l$pnj@senator-bedfellow.MIT.EDU>#1/1
X-Deja-AN: 106188707
references: <3t0233$2jb@freenet.vcu.edu> <mkl.804524239@duesentrieb>
organization: Center for Space Research
reply-to: da...@space.mit.edu
newsgroups: comp.os.linux.setup,comp.os.linux.misc
Hi,
The following article originally appeared in comp.unix.aux inder the
subject `A/UX future and platform ??'. The objections about Linux raised in
this article may affect whether or not I will get permission to install a
linux system on our network that consists of about 120 Sun workstations.
On 30 Jun 95 14:57:19 GMT, Mario Klebsch DG1AM <m...@rob.cs.tu-bs.de> wrote:
: aru...@freenet.vcu.edu writes:
: >Well, according to a post regarding SVR4 on comp.security.unix,
: >Linux is very similar to SVR4 w/r to source.... In fact, among
: >the utilities required for Linux operation is "SystemV-style
: >init, which control booting, shutdowns, and runlevels....
:
: About a week ago, I had a look at a friends Linux system. I wanted to
: tell him, how to start xdm on startup. He told me about /etc/inittab,
: and so I told him to write a script for /etc/init.d and to create
: links in /etc/rcS.d and /etc/rc2.d. But unfortunatly, these
: directories were missing. There were file in /etc/rc.d instead, that
: were organised like the BSD rc files. Than I found a comment in
: /etc/inittab that initstate 6 is for X operation. On all SYSV
: machines, I had to use, initstate 6 is for rebooting the system.
:
: These are my comments to the Linux SystemV-style. Perhaps this is
: already changed, but this really is typical for my experience with
: Linux.
:
: >also, BSD is as "Unix" as SysV. To my understanding, Berkeley
: >and AT&T developed Unix at about the same time, but AT&T got
: >the pattent first.
:
: Surely, they both are UNIX. And as an experienced UNIX operator, I had
: contact with both of them.
:
: >...
: >i will say that perchance to someone familiar with "commercial"
: >Unix or BSD, Linux may be somewhat disorienting. However, (1)
: >all the tools you need ARE, in fact there, and (2) Linux
: >functions in just the same way as Unix does, notwhithstanding
: >what the utilities are called
:
: I often was asked as an experienced sysop, to help people with
: Linux. And when I don't know the system, I have to relie on the facts
: I know about UNIX. And lots of these facts are not portable to Linux,
: when you really are in trouble. It is no help, that there are man
: pages, when in single user mode, /usr is not mounted and the man pages
: are inaccesable.
:
: As long as you don't know the magic word to mount /usr, you are
: lost. This happend to me, when I helped a friend to repartition the
: boot disk. I did not touch disk with /usr and /home, I made a backup
: of the root partition and then I repartitioned the disk. After loading
: the memory version of Linux, I was in trouble. The only thing, that
: helped me, was the documentation, that was printed prior to the
: installation.
:
: But all I want to say is, this is a simple task. I did this several
: times in our Suns running SunOS4 and Solaris. I did these things on
: other UNIX systems, too. A really good example of how it can be was
: the installation of freeBSD, I did for a friend about 4 month ago.
:
: Perhaps, my bad impression on Linux is based on the fact, I was almost
: always called, when people had problems, not when it worked fine. Then
: I found, almost all my knowledge on UNIX useless on this system.
:
: I thing, one of the best tests for a system is, how it behaves, when
: facing a desaster. Can it still boot to a point, where I can fix it?
: Can I fix if with my basic knowledge of the system? What tools can I
: use to analyze the situation?
:
: 73, Mario
: --
: Mario Klebsch, DG1AM, M.Kl...@tu-bs.de +49 531 / 391 - 7457
: Institut fuer Robotik und Prozessinformatik der TU Braunschweig
: Hamburger Strasse 267, 38114 Braunschweig, Germany
From: j...@obelix.cica.es (Jonathan Noel Tombs)
Subject: Re: Linux is not Unix.
Date: 1995/07/14
Message-ID: <3u6d0n$gc3@obelix.cica.es>#1/1
X-Deja-AN: 106188629
references: <3t0233$2jb@freenet.vcu.edu> <mkl.804524239@duesentrieb>
<3u647l$pnj@senator-bedfellow.mit.edu>
organization: Centro Informatico Cientifico de Andalucia
newsgroups: comp.os.linux.setup,comp.os.linux.misc
In article <3u647l$p...@senator-bedfellow.mit.edu>,
John E. Davis <da...@space.mit.edu> wrote:
> The following article originally appeared in comp.unix.aux inder the
>subject `A/UX future and platform ??'. The objections about Linux raised in
>this article may affect whether or not I will get permission to install a
>linux system on our network that consists of about 120 Sun workstations.
what are the suns running, if it is 4.1.3+ then linux is fairly compatible,
the only major difference is the rc files are in /etc/rc.d, the use of
/etc/inittab and configuration of serial ports. But bare in mind the hardware
is different and SunOs (at least on sparc) doesn't offer Virtual terminals
video mode selection and a ton of 3rd party hardware so some things just can't
be the same. Also Suns come with a tuned for booting unix prom, wheras your
PC will just have a small boot loader, get the admin to readup on passing
arguments to lilo.
I started on SunOs 3.1, went through the Sun0S4.0 to 5.x, and also I have had
tp managed several releases of AIX (wow is that a different unix!), IRIX, OSF/1
Solaris 2, and worked on Convexes and ultrix vaxes. Now out of all these I
would a say as a system manager.
SunOs to AIX is more different than
SunOS to Solaris 2 is more different than
SunOs to IRIX or OSF or Linux.
If your sysadmins don't want a linux box because it will be difficult to
manage then just wait till they buy any other non sun.
We run mixed Suns/Linux here, they share the same nfs file systems, same
user accounts, same yp passwds. Yes you need to learn a bit to start with
but the bigger differences are from the hardware, not the software.
Jon.
From: Karl_Kl...@cs.cmu.edu
Subject: Re: Linux is not Unix.
Date: 1995/07/14
Message-ID: <vxkohyxs34x.fsf@APHRODITE.NECTAR.CS.CMU.EDU>
X-Deja-AN: 106188717
references: <3t0233$2jb@freenet.vcu.edu> <mkl.804524239@duesentrieb>
organization: Carnegie-Mellon Univ, Nectar Project
newsgroups: comp.os.linux.setup,comp.os.linux.misc
da...@space.mit.edu (John E. Davis) writes:
> The objections about Linux raised in
> this article may affect whether or not I will get permission to install a
> linux system on our network that consists of about 120 Sun workstations.
I certainly hope that one negative report is not going to be the
camel's back-breaking straw. If that's all it takes to make a
show-stopper in your environment...well, drop me a note and I'll give
you extreme details on installation of any number of UNIX(-like) OSes
and we'll see how many possibilities we can shoot down for you, which
would leave your administration with an interesting, and strictly
political, choice.
Of the complaints listed, none are "substantial" to me. As far as the
disc repartition nightmare goes, he clearly just plain screwed
something up. I have been working with Linux less than 3 weeks now;
I've installed it 4 times on various PCs (2 of my own), and have yet
to have a problem of any sort. The limit of my difficulty is that the
first time, on my laptop, took rather longer than I thought it ought
to have -- and that was my fault, not InfoMagic's (I used their March
CD-ROM set), because I failed to read carefully. _Any_ vendor's OS
can be a nightmare to install if you don't read the docs properly.
As for "SysV-style init" and related goo...please give us serious
critiques with which to concern ourselves. The fact that SysV uses
state 6 for reboot is a non-issue. Most people I've known running
SysV take it down with "telinit -S" (I think that's the incantation --
it's been quite a while) followed by the power switch.
If one wants to worry about what's a "real" UNIX system, one could
wonder at the horrors of any number of vendors' bastardizations,
including HP-UX (one of my officemates has an ongoing problem with HP,
and most recently, he has found that HP-UX _will_not_ support a 4Gb
drive), OSF/1 (a personal nightmare -- I hack networking code in
alpha_osf20 for a living), and the raw evilness of Sun's latest
abortion.
In fact...you say that you have a heavy Sun environment, yet you are
getting complaints about the supposed dangers of Linux. Pot, meet
kettle; kettle, meet pot: You're both black.
> : I often was asked as an experienced sysop, to help people with
> : Linux. And when I don't know the system, I have to relie on the facts
> : I know about UNIX. And lots of these facts are not portable to Linux,
> : when you really are in trouble. It is no help, that there are man
> : pages, when in single user mode, /usr is not mounted and the man pages
> : are inaccesable.
I will walk out to the thin part of the branch and say that the author
of these words is not much of a UNIX person, regardless of being an
"experienced sysop."
I have been hacking, in genuine UNIX, everything from the kernel
upward for most of the last 15 years, and I find that _most_ of what I
know translates to Linux. Not all, to be sure, but most. The fact
that I was able to do a raw install of Linux onto a laptop in a single
evening, starting from scratch with the InfoMagic booklet telling me
how to partition my DOS drive, was quite a wonder to me.
> : As long as you don't know the magic word to mount /usr, you are
> : lost. This happend to me, when I helped a friend to repartition the
> : boot disk. I did not touch disk with /usr and /home, I made a backup
> : of the root partition and then I repartitioned the disk. After loading
> : the memory version of Linux, I was in trouble. The only thing, that
> : helped me, was the documentation, that was printed prior to the
> : installation.
Oh, horrors, that one might be required to be holding documentation
when installing a system...
Again, this level of critique has no credibility whatever.
> : Perhaps, my bad impression on Linux is based on the fact, I was almost
> : always called, when people had problems, not when it worked fine. Then
> : I found, almost all my knowledge on UNIX useless on this system.
Let's see: The critic observes that cockpit error had previously
damaged the situation so severely that he had trouble recovering to
sane state.
So call, say, Compaq and tell them that you've just deleted C:'s
partition 1...you'll get much the same reply: Start over from scratch,
if you're that stupid. And stupid it is.
> : I thing, one of the best tests for a system is, how it behaves, when
> : facing a desaster. Can it still boot to a point, where I can fix it?
My Linux boxes can boot when facing disaster just fine. I keep a boot
floppy near each one, on a just-in-case basis, and have really needed
it on one once, because I botched LILO installation. Whoopie -- 5
minutes later, and my system booted on its own again.
> : Can I fix if with my basic knowledge of the system? What tools can I
> : use to analyze the situation?
If you can't find the tools for a Linux system, you haven't looked for them.
From: miq...@drinkel.ow.org (Miquel van Smoorenburg)
Subject: Re: Linux is not Unix.
Date: 1995/07/17
Message-ID: <9507172932.014328@drinkel.ow.org>#1/1
X-Deja-AN: 106333828
references: <3t0233$2jb@freenet.vcu.edu> <mkl.804524239@duesentrieb>
<3u647l$pnj@senator-bedfellow.MIT.EDU>
<vxkohyxs34x.fsf@APHRODITE.NECTAR.CS.CMU.EDU>
organization: Alcohol diseases R&D
x-news-software: W-NEWS Release 4.52 for OWLD
newsgroups: comp.os.linux.setup,comp.os.linux.misc
In article <vxkohyx...@APHRODITE.NECTAR.CS.CMU.EDU>,
<Karl_Kl...@cs.cmu.edu> wrote:
>da...@space.mit.edu (John E. Davis) writes:
>> The objections about Linux raised in
>> this article may affect whether or not I will get permission to install a
>> linux system on our network that consists of about 120 Sun workstations.
>
>As for "SysV-style init" and related goo...please give us serious
[stuff deleted]
>critiques with which to concern ourselves. The fact that SysV uses
>state 6 for reboot is a non-issue. Most people I've known running
>SysV take it down with "telinit -S" (I think that's the incantation --
>it's been quite a while) followed by the power switch.
[more stuff deledted]
Well, later versions of sysvinit (from 2.50 and up) haven't been using
runlevel 6 for X. That's a Slackware'ism. I still hope Patrick Volkerding
will upggrade Slackware to sysvinit 2.50 or up, but as 2.50+ is not
really compatible with 2.4 chances are small.
In 2.50+, runlevel 0 is halt, 1 is single user and 6 is reboot.
From 2.56+, even "shutdown -y -i6" will work (undocumented,
for SysV diehards. Yes, I had to work with SCO at the time).
Debian Linux uses sysvinit with a real /etc/init.d (etc...) directory
structure.
Oh, and init reacts to CTRL-ALT-DEL and does a clean shutdown (or
whatever else you tell it to do in /etc/inittab)
Mike.
--
+ Miquel van Smoorenburg + Cistron Internet Services + Living is a |
| miq...@cistron.nl | Independent Dutch ISP | horizontal |
+ miq...@drinkel.ow.org + http://www.cistron.nl/ + fall +