Subject: Linux-Misc Digest #589
From: Digestifier <Linux-Misc-Request@senator-bedfellow.MIT.EDU>
To: Linux-Misc@senator-bedfellow.MIT.EDU
Reply-To: Linux-Misc@senator-bedfellow.MIT.EDU
Date:     Sun, 23 Jan 94 22:15:58 EST

Linux-Misc Digest #589, Volume #1                Sun, 23 Jan 94 22:15:58 EST

Contents:
  Re: Term == SLIP ? (Del Armstrong)
  Re: Linux logo (Bill Bogstad)
  Using Perfmeter with Linux... (Bob Henry)
  Re: Adaptec VESA-Local Bus SCSI controller experinces/advice?? (Kenneth Jamieson)
  Re: PPP slower than SLIP! (Rene COUGNENC)
  Using a linux machine for a SLIP server? (Andy Burnett)
  Re: LINUX on INTERNET via SLIP (Arjan de Vet)
  Re: dos2unix, unix2dos text conversion programs?? (Eyal Lebedinsky)
  Re: How much disk for Slackware 1.1.1 (Rick Slater)
  SLIPery Term (Iain Bryson)
  Linux (Slackware 1.1.1 0.99.14) + INN 1.4 - Need config.data copy! (Mark J. Bailey [ADMIN])
  Linux Pentium workstations $3589 (Edwin Tisdale)
  Re: Slackware needs a shadow package! (Mike Jagdis)
  Re: LINUX on INTERNET via SLIP (Charles Hedrick)
  Re: Linux vs. NetBSD (rodrigo vanegas)
  Re: Need assistance with AHA-1522A SCSI card (Arnoud Martens)
  Re: Linux as X-Terminal? No! (Benjamin Z. Goldsteen)
  Re: Linux logo (Matt Welsh)

----------------------------------------------------------------------------

From: dela@nova.ceas.rochester.edu (Del Armstrong)
Crossposted-To: comp.security.unix
Subject: Re: Term == SLIP ?
Date: 23 Jan 94 18:34:36 GMT
Reply-To: dela@ceas.rochester.edu

In article <CJzzLu.GFy@rahul.net> dhesi@rahul.net (Rahul Dhesi) writes:
>In <2hp40m$e1u@thermal.ceas.rochester.edu>
>dela@thermal.ceas.rochester.edu (Del Armstrong) writes:
>
> >Like I said before, term is a great tool. But sysadmins of machines
> >whose users run term, should understand that this is another
> >vulnerability that they're going to have to deal with.
> 
> In what way is it worse to let user A run term than to let user A log
> in at all?   In either case, (a) A has access to the machine at the
> user level, and (b) A can let B have access to the machine at the user
> level. 

The difference is that my users know they're not supposed to share
their accounts. When they do share their account anyway, they usually
use some discretion in whom they give the account to.  Also, I can
often detect when my users do share their accounts, allowing me to
talk to them about it.

If one of my users runs term from a poorly administered system (the
original assumption in this thread), an intruder could use their
account on my machine without either the user or myself noticing it.

I consider that a significant difference.

This thread isn't a big deal. All I wanted to point out is that it's
nieve not to recognize term as another variable in the security
equation for your site.


        Del Armstrong

======================================================================
dela@ceas.rochester.edu     rutgers!ur-valhalla!dela     (716)275-5342
        Computing and Networking Group, College of Engineering
                University of Rochester, Rochester, NY

------------------------------

From: bogstad@gauss.cs.jhu.edu (Bill Bogstad)
Subject: Re: Linux logo
Date: 23 Jan 1994 14:22:43 -0500

In article <1994Jan23.072751.8858@cs.cornell.edu>,
Matt Welsh <mdw@cs.cornell.edu> wrote:
>>[ASCII Linux logo at boot up]
>...
>No!!!
>
>...
>Seriously, graphics like this smack of unprofessionalism, and they
>remind me all-too-much of a number of signatures that I've seen
>warlorded across the Net. ...

        From what I have heard just about every major personal computer
operating system has some kind of special graphic or other special feature
to prove that it was written by real "hackers".  I guess we could argue
whether or not any of them are professional.  About the only thing that
would be "unprofessional" about this is that it is too obvious.  You should
have to hold down Shift-ALT-L during the boot process to have it appear and
it should be stored in a part of memory which is immediately overwritten by
the kernel during the rest of the boot process.  It will, of course, already
be compressed because we all use zBoot now...

                                        Bill Bogstad
                                        bogstad@cs.jhu.edu

------------------------------

From: henry@silver.ucs.indiana.edu (Bob Henry)
Subject: Using Perfmeter with Linux...
Date: Sun, 23 Jan 1994 20:23:53 GMT

I use olvwm (yes I know you all think fvwm rules), anways, I would like 
to use perf(meter), but the problem is that it uses 'rstatd' and not /proc.

If anyone has ported 'rsatd', hacked perf(meter), or has any thought 
about this they would be appreciated.

Thanx :)


--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Bob Henry - Student of the Human Condition              Coming at you from:
                                                        Purdue Looniversity 
"Enjoyin' the ride..."                                  Planet Earth (?)
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

------------------------------

From: tron1@shasti.xei.com (Kenneth Jamieson)
Crossposted-To: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi
Subject: Re: Adaptec VESA-Local Bus SCSI controller experinces/advice??
Date: 23 Jan 1994 15:52:45 -0000

In article <CK2Kv9.52v@colorado.edu>,
Drew Eckhardt <drew@kinglear.cs.colorado.edu> wrote:
>In article <1994Jan20.145712.13549@microware.com>,
>Adam Goldberg <adam@microware.com> wrote:
>>This is sort of a general plea:
>>
>>(1) Does anyone have any experience with Adaptec's VESA Local Bus SCSI
>>    controller?  Is there a significant performance gain vs. the
>>    1542x?
>
>Which one?  Adaptec has a 6360 based board (ie, a VESA 1520 board)
>as well as a 7770 based board, the 2842.  The former should allow 
>transfer rates similar to an Adaptec 1540 for many applications,
>at the expense of additional CPU time used.  The later should be 
>faster due to the lower per-command overhead.

        I will say that I am running the 2842 in a DOS/WINDOWS box and 
compared to my 1542B (now running in my linux box) it SCREAMS.



-- 
                     Too tired to think of a funny .signature.
                                Kenneth Jamieson
        tron1@shasti.xei.com / tron1@xei.jvnc.net / tron1@wisdom.bubble.org
                     Via Linux ALPHA 0.99 PL14l - Thanks Linus!

------------------------------

From: rene@renux.frmug.fr.net (Rene COUGNENC)
Crossposted-To: comp.os.linux.help
Subject: Re: PPP slower than SLIP!
Date: 23 Jan 1994 08:24:28 GMT
Reply-To: cougnenc@itesec.ensta.fr (Rene COUGNENC)

Ce brave Michael Horwath ecrit:

> Rene COUGNENC (rene@renux.frmug.fr.net) wrote:
> : PING ftp.ibp.fr (132.227.60.2): 56 data bytes
> : 64 bytes from 132.227.60.2: icmp_seq=7 ttl=179 time=158 ms


> I wish I could get ping times like that on my system (and a friend is
> thinking the same thing right here over my shoulder). 


One thing to have in mind whith modern modem links such as v32bis or more,
is that the connection speed may often not be really 14400, but a
slower fallback speed, depending on the phone line quality, the modem
quality, and many other things.

For example, if I always get a good connection between my good old USR-
Dual standard and another USR modem, I never connect a 'Supra' at more
than 7200 bps...

More than that, some modem protocols may be confused by the data transmitted,
I remember a 'Best 2400' modem a few years ago, unable to do some zmodem
transfers whithout hanging on some files...
--
 linux linux linux linux -[ cougnenc@renux.frmug.fr.net ]- linux linux linux 

------------------------------

From: burnett@baldrick.cecer.army.mil (Andy Burnett)
Subject: Using a linux machine for a SLIP server?
Date: 23 Jan 1994 22:21:03 GMT

  I'm trying to set up my linux machine so that I can dial into it and
start up SLIP from it.  When I dial in, logging in with my slip account
name, the linux machine ifconfig's sl0 and adds a route for it.  On my
local machine I do that same.  I can ping my remote linux box, but if I
try and telnet to it, the telnet hangs, and I get this in my remote's
syslog:
        telnetd[PID]: ttloop: peer died: try again

  Can a linux machine be used as a slip server?  Is there something that
needs to be set up that would get rid of these ttloop errors?

                                Andy Burnett
--
Andy Burnett                --  burnett@baldrick.cecer.army.mil
US Army Corps of Engineers  --  Construction Engineering Research Lab

------------------------------

From: devet@adv.win.tue.nl (Arjan de Vet)
Subject: Re: LINUX on INTERNET via SLIP
Date: 23 Jan 1994 22:25:46 +0100

In article <1994Jan23.164301.4004@bhhome.ci.net>,
Bill Heiser <bill@bhhome.ci.net> wrote:

>What level of LINUX, and exactly what networking files do I need to upgrade
>to get SLIP to work reliably?  Is it possible at all?

Linux 0.99.14 works OK for me.

>Then there will be the whole issue of making LINUX secure enough to leave
>on the net without crackers trashing it :-)

Use Wietse Venema's tcp-wrapper to allow, deny and log network services. I
only have a minimum amount of network services available (smtp, ident,
ftp, gopher, and www for all sites, finger and telnet for some `trusted'
sites).

Arjan

--
Arjan de Vet                             <Arjan.de.Vet@adv.win.tue.nl> (home)
Eindhoven University of Technology, the Netherlands <devet@win.tue.nl> (work)

------------------------------

Crossposted-To: comp.os.linux.help
From: eyal@fir.canberra.edu.au (Eyal Lebedinsky)
Subject: Re: dos2unix, unix2dos text conversion programs??
Date: 23 Jan 94 22:19:57 GMT

Hello Everyone,

The original article included programs for dos->unix and vv cr-lf
conversion. These are OK. However, in my experience, it is nicer to have
a pair of programs which are reversible. In other words, there is a
guarantee that running dos->unix then unix->dos will end up with
identical results.

This is not the case with these programs. The problem is that the
dos->unix simply removes the '\r' chars. To be correct it should ONLY
remove those that preced the '\n'. I managed in this way to recover
files that accidentally converted unix->dos (by some comms utilities).
Note however that some implementations of the unix->dos programs have a
problem where if they find a "\r\n" sequence then they do NOT add
another '\r'; this kills the chance of recovery. DOn't know why anubody
would do such a thing, but I found this in a number of situations.

Just when you though there is nothing to this unix<->dos coversions :-)


--
Regards
        Eyal Lebedinsky         eyal@ise.canberra.edu.au

------------------------------

From: slater@nrlssc.navy.mil (Rick Slater)
Subject: Re: How much disk for Slackware 1.1.1
Date: 23 Jan 1994 18:46:44 GMT

J Eric Bracken (bracken@accord.ece.cmu.edu) wrote:

: I can't seem to find how much disk space Slackware 1.1.1 will take up
: when installed.  Does anybody know, or can you point me to a relevant
: file?

I suppose that the least helpful answer is "it depends".  But of course
it does depend upon what disk sets you choose to install.  I have in-
stalled nearly everything but emacs and the network stuff and am
pushing 80 megs of disk space.  This includes X and Xview -- two systems
which require a lot of disk.

A minimal installation would probably consist of only the three A disks
and the four AP disks.  That would occupy fewer than 20 megabytes of
disk.  If you left the AP disks out, you would need around 10 megs of
space, but you couldn't do much with what you had. -- Rick

P.S.  I'm *very* *very* pleased with the Slackware installation,
as compared to an SLS installation that I did more than a year ago.
The Slackware material has all run correctly right "out of the box"
with only a few tiny glitches.  It's a very professionally organized
installation system, and is a real asset to the Linux community.

------------------------------

From: iain@ECE.Concordia.CA (Iain Bryson)
Subject: SLIPery Term
Date: Sun, 23 Jan 1994 22:52:05 GMT

Hi.

I'd like to know how far towards SLIP I can get a term link 
to go.

Does SLIP need root acces simply for it's serial port 
control, (or not even for that) or do you need root
to do the IP stuff?

IF the IP can be done without root, could SLIP be
modified to work through a term link?
Is this being worked on?  Where could the sources
be acquired to see if such modifications are 
possible/plausible (if nobody else is doing this, and, of
course, it is at all possible).

Thank you.
--
 _________________________________________  ______________________
/ Iain Bryson, Montreal, Quebec, Canada.  \/ Things I do:         \
| iain@ece.concordia.ca - for the time    ||C/++, perl, UNIX, X,  | 
|                         being           ||Watch "Strange Brew". |
\_________________________________________/\______________________/
 ______________________________  _________________________________
/ DOS was created by the devil.\/ Happily running Linux 0.99pl14! \
\______________________________/\_________________________________/ 

------------------------------

From: root@perot.mtsu.edu (Mark J. Bailey [ADMIN])
Crossposted-To: comp.os.linux.help,news.software.b,news.software.nntp
Subject: Linux (Slackware 1.1.1 0.99.14) + INN 1.4 - Need config.data copy!
Date: 23 Jan 1994 23:19:49 GMT

Hi,

I have Just loaded Linux (Slackware 1.1.1 Dsitribution with Kernel 0.99.14)
on a 486DX33 EISA system.  I want to load INN 1.4 on it as that is what we
use for our main campus newsserver.  I did all the configuration on INN 
the first time around and was hoping that someone reading this that has 
already installed INN 1.4 on Linux could maybe email me a copy of your 
config.data file for INN so that I could save some time in answering system
dependent config.data entries.  I know (of course) that I will have to 
change SITE dependent values...but SYSTEM (ie, Linux) dependent values should
be (for the most part) similar to all Linux installs.

Please email replies!  Many thanks!

Mark

--
                                                   ^
CSCI LAN/Lab/Systems Director                      |          Mark J. Bailey
Department of Computer Science               NFS - * - YP     root@mtsu.edu
Middle Tennessee State University                 / \         615-898-2397
Murfreesboro, Tennessee  37132  USA             TCP DECnet    615-898-2144 lab

------------------------------

From: edwin@maui.cs.ucla.edu (Edwin Tisdale)
Subject: Linux Pentium workstations $3589
Date: Sun, 23 Jan 94 23:50:28 GMT

A Linux Pentium workstation is just a little faster than (and about one
third of the list price of) a comparably equipped Super SPARC Model 41.
The minimum recommended system includes:

*       Intel 586DX-60 MHz Pentium CPU with heat sink and fan
*       30 MHz PCI-586 mother board with 256 kB cache
*       16 MB 70 ns system memory
*       540 MB 12 ms IDE hard disk drive
*       1.4 MB 3.5"  floppy disk drive
*       1.2 MB 5.25" floppy disk drive
*       PCI 1280 x 1024 1 MB Super VGA graphics card
*       15" 1024 x 768 0.28 mm dot pitch non-interlaced color monitor
*       Mid-size tower case and power supply
*       101 enhanced soft touch keyboard
*       3 button serial high resolution mouse
*       Linux operating system, X, C++, LaTeX, etc.
*       two year warrantee

If you are interested, you should contact Moujan Ahouraian at (800) 486-2020

                20/20 Technologies
                1786 Westwood Blvd.
                W. Los Angeles, CA 90024
                TEL: (310) 441-8855
                FAX: (310) 441-8869

Enjoy, Bob Tisdale.

------------------------------

From: jaggy@purplet.demon.co.uk (Mike Jagdis)
Subject: Re: Slackware needs a shadow package!
Date: Sat, 22 Jan 1994 22:51:00 +0000

* In message <1994Jan22.165357.19502@rpp386>, John F. Haugh

JI> If you can send me the source to in.ftpd I can made the mods needed to
JI> get it to run correctly with Shadow.  I did the same to the
JI> BSD RENO version of ftpd and it only took a few moments.

Yes, the changes required for shadow support *are* trivial. In fact I went 
through the Linux net daemons (ftpd, popd, pcnfsd, bwnfsd etc.) a few months 
ago and add shadow support to them. And sent the patches to the maintainer. 
Who merged them in.

  I've been using shadow for *months* without major problems - including 
network services, xdm etc. The problem with Linux on the Internet is that 
people here have been burnt by some pretty ropey bundling in "distributions" 
and many people still listen to the chinese whispers instead of *thinking* 
for themselves. Hence the common cries of, "I've heard that...". Sad.

                                Mike  
 

------------------------------

From: hedrick@geneva.rutgers.edu (Charles Hedrick)
Subject: Re: LINUX on INTERNET via SLIP
Date: 24 Jan 94 00:29:13 GMT

devet@adv.win.tue.nl (Arjan de Vet) writes:

>Linux 0.99.14 works OK for me.

You're lucky.  There are serious problems on the write side of TCP in
0.99pl14, which caused FTP's to hang for me most of the time, and X
session startups some of the time.  These problems should not occur on
Ethernet, because they're due to using the small MTU's that you
probably want to use on SLIP.

Pl14t seems to be in pretty good shape.  There are still a few
problems, but they're getting increasingly minor.  Basically every
edit since about pl14m has dealt with some cause of TCP connections
hanging.  I can think of a situation that could cause a connection to
hang in pl14t, but I think it's unlikely to happen in practice.  (It
has to do with the fact that the code doesn't probe zero windows.
I've coded a fix, tested it, and submitted it to the proper
authorities.)  I have some hope that that's the last thing that would
cause connections to hang.  The next few edits are likely to deal with
some performance issues, and possibly (I don't know much about this
work) a cleanup of the finite state machine handling.  There are a
number of people going over this code with a fine-toothed comb.

------------------------------

Crossposted-To: comp.unix.advocacy
From: rv@cs.brown.edu (rodrigo vanegas)
Subject: Re: Linux vs. NetBSD
Date: Mon, 24 Jan 1994 00:42:54 GMT

In article <2hv1h9$gt9@olivaw.apanix.apana.org.au>, hart@apanix.apana.org.au (Leigh Hart) writes:

> This [386BSD] is what I consider a very workable and stable network system. 
> Linux, however, still has many problems with its networking code.
> The reason being is the networking code in Linux was written from
> scratch, and it's experiencing the problems that BSD has had/fixed
> over the last 10 years.

But BSD source is free isn't it?  Why don't the linux network code
developers simply copy it out?


rodrigo vanegas
rv@cs.brown.edu

Followups to comp.os.linux.misc

------------------------------

From: arnoud@ijssel.hacktic.nl (Arnoud Martens)
Subject: Re: Need assistance with AHA-1522A SCSI card
Date: Sun, 23 Jan 1994 21:34:52 GMT
Reply-To: arnoudm@ijssel.hacktic.nl

Preston Beach writes in newsgroup comp.os.linux.misc:
> Dear Netland,
>   I have a desire to run Linux SLS on my machine. Everything 'seems' to
> be going alright except for one small problem.  Linux does not recognize
> my SCSI card.  My card does work under DOS but that is not what I want.
> If anyone has any idea what I can do short of buying another SCSI card,
> to get my Adaptec AHA-1522A card to work, please let me know.  I would
> prefer not to have to get another card but the option is there.
> I have looked through alot of docs about hardware usable with Linux and
> some of the Adaptec cards are listed, unfortunately, not this one.
> I read in one doc that Adaptec 1522 IS supported but when I boot it says
> there is no SCSI card.  If someone knows what configuration the card
> needs to be (jumpers) let me know.

Your problem is not the card it is the Linux kernel version you
are trying to install. The AHA152x driver was included in the
standard release since pl14 so you either have to install a
newer distribution or get/compile a newer kernel.

If you want more info on compiling, getting a newer kernel, mail
me.

Gtx:

-- 
~ Arnoud Martens                Delft, the Netherlands
                                +31(0)15-563621
                                arnoudm@ijssel.hacktic.nl


------------------------------

Crossposted-To: comp.windows.x.i386unix
From: ben@rex.uokhsc.edu (Benjamin Z. Goldsteen)
Subject: Re: Linux as X-Terminal? No!
Date: Mon, 24 Jan 1994 01:08:55 GMT
Reply-To: benjamin-goldsteen@uokhsc.edu

byron@cc.gatech.edu (Byron A Jeff) writes:

[lots deleted]

>Ben,

>I have no intention to subjecting my users (mostly students - you know
>the kind that stare at the screen for multiple hours at a time. ;-) The 14"
>mono monitors suck! I'm trying to find something inbetween the 14" and
>the big 19" and 21" greyscales (at $1000 apiece).

>OK, now help me out. Where did you see these *GOOD* greyscale monitors?
>That $150-200 price tag is in the range I'm looking for. What size were they?
>In fact I'd like to drop in high resolution 17" color monitors. I can't justify
>the cost though. I'll ask again:

>Does anyone know of a SVGA compatible 16"-19" greyscale monitor (new) that
>cost less than $400 or so?

     What I was saying was that a good 14" mono monitor was ~$200.  I
am afraid I don't keep up on prices anymore...they just change to fast. 
Somebody told me Computer Discount Warehouse (you can get the number
from 1-800-555-1212) has good prices if you don't need support.  I'm
beginning to look for local places myself especially for hard drives
and monitors.  In the past, I have seen price differences of like 40%
or more but the difference is becoming like <10% for some stores on
some items (for things like monitors you can claim it needs to be
aligned again and for hard drives the replacement is quick).

     Sorry if I my tone sounded like I was critizing you.  I guess I
was a little edgy at the time and it seemed after reading your message
and the follow-up that the goal was a cheapy monitor.  I realize
everybody is trying to do the best they can with what funds they have.

>>     Also from what I hear throw in a few extra bucks and get a SMC
>>Ethernet card over the NE2000 (somebody gave me the best SMC card but I
>>don't have the name right now)..

>Thanks for the suggestion. What's the cheapest price anyone has seen for an 
>SMC?

     The card I was recommended is the SMC Elite-16 (WD8013..they may
have rev'ed another version of it like WD8013EP [Elite Plus] or
WD8013EPC (Elite Plus Combo)...does anybody know the difference between
the Elite and the Ultra?).  I think I saw numbers around $110 for the
Elite.

     I was also recommended against the 3Com 3c509 in case you come
across that.  Supposedly it uses a lot of CPU time (which is like
"free" under DOS but not UNIX...)

Good Luck
-- 
Benjamin Z. Goldsteen

------------------------------

From: mdw@cs.cornell.edu (Matt Welsh)
Subject: Re: Linux logo
Date: Mon, 24 Jan 1994 01:44:19 GMT

In article <2htjfk$g7k@ifi.uio.no> Kjetil T. Homme <kjetilho@ifi.uio.no> writes:
>I see your point, but it is my experience that most OS include some
>frills during boot-up, like Sun's fancy ray-traced colour logo and SGI
>Indy's "ta-da-da---DA!" fanfare. 

Sun has a raytraced logo. 
SGI has a fanfare.

Linux has a giant ASCII barphic. 

What does this say about us? 

Oh, and by the way, Yggdrasil's CD-ROM reportedly does a little sound
bite at boot time, at least when booting from the installation floppy.
But I don't belive that it's in the kernel, it's just a quip fired off
from /etc/rc.

mdw
-- 
"Oh, Golgi, woe is me, you can't even see the sea."

------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Misc-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux.misc) via:

    Internet: Linux-Misc@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    sunsite.unc.edu				pub/Linux

End of Linux-Misc Digest
******************************
