Linux-Misc Digest #21, Volume #26                Fri, 13 Oct 00 14:13:02 EDT

Contents:
  Problems installing OpenSSL from source on Red Hat 6.0 (Stephen Cornell)
  Re: Persistent password problem (Bill Unruh)
  Re: How will I restore back LILO (Bill Unruh)
  Re: Maximum swap partition size? ([EMAIL PROTECTED])
  Re: End-User Alternative to Windows ([EMAIL PROTECTED])
  AVI for LINUX ("Zoran P.")
  Re: Accessing certain sites fails SOLVED. (Dances With Crows)
  ppp compression options ([EMAIL PROTECTED])
  Problems with Netscape on Linux ([EMAIL PROTECTED])
  Re: help audio on linux (Lew Pitcher)
  Re: help with compiling new kernel for ibm laptop (Paul Kimoto)
  Re: End-User Alternative to Windows (The Ghost In The Machine)
  Re: Financial software ~ Quickbooks Pro ~ Linux? (Jean-David Beyer)

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

From: Stephen Cornell <[EMAIL PROTECTED]>
Subject: Problems installing OpenSSL from source on Red Hat 6.0
Date: 13 Oct 2000 17:07:58 +0100

I've downloaded Openssl 0.9.6, and I've been trying to build it with
shared libraries (./config shared).  `make' and `make test' proceed
without a problem, but when I run `make install' I get the following
error (after quite a few bits have been installed):

. 
. 
. 
tools
installing tools...
make[1]: Entering directory `/usr/local/src/openssl-0.9.6/tools'
make[1]: Leaving directory `/usr/local/src/openssl-0.9.6/tools'
installing libcrypto.so
/usr/bin/ranlib: /usr/local/ssl/lib/libcrypto.so: File format not recognized
installing libcrypto.so.0
/usr/bin/ranlib: /usr/local/ssl/lib/libcrypto.so.0: File format not recognized
installing libcrypto.so.0.9.6
/usr/bin/ranlib: /usr/local/ssl/lib/libcrypto.so.0.9.6: File format not recognized
installing libssl.so
/usr/bin/ranlib: /usr/local/ssl/lib/libssl.so: File format not recognized
installing libssl.so.0
/usr/bin/ranlib: /usr/local/ssl/lib/libssl.so.0: File format not recognized
installing libssl.so.0.9.6
/usr/bin/ranlib: /usr/local/ssl/lib/libssl.so.0.9.6: File format not recognized 

I don't know what it's trying to do, but it fails to install libssl.a
in the correct place.  Shouldn't it be running ranlib on libssl.a
instead? 

torus:openssl-0.9.6 (21)% file /usr/local/ssl/lib/libssl.so.0.9.6
/usr/local/ssl/lib/libssl.so.0.9.6: ELF 32-bit LSB shared object, Intel 80386, version 
1, not stripped
torus:openssl-0.9.6 (22)% file /usr/local/src/openssl-0.9.6/libssl.a 
/usr/local/src/openssl-0.9.6/libssl.a: current ar archive

I _think_ I need to build shared versions libraries in order to build
KDE2 with SSL enabled.  

Any advice would be much appreciated.
--
Stephen Cornell          [EMAIL PROTECTED]         Tel/fax +44-1223-336644
University of Cambridge, Zoology Department, Downing Street, CAMBRIDGE CB2 3EJ

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

From: [EMAIL PROTECTED] (Bill Unruh)
Crossposted-To: comp.os.linux.setup
Subject: Re: Persistent password problem
Date: 13 Oct 2000 16:10:21 GMT

In <8s694u$[EMAIL PROTECTED]> [EMAIL PROTECTED] (Ron Ho) writes:

]help. i'm confused.

]i have a linux redhat 6.2 (2.2.14-5.0smp) machine. in my /etc i have two
]files for passwords and shadows:

]chubba-<28> ls -al /etc/pass*
]-rw-r--r--    1 root     root          576 Oct 12 22:27 /etc/passwd
]-rw-r--r--    1 root     root          576 Oct 12 22:27 /etc/passwd-
]chubba-<29> ls -al /etc/shad*
]-r--------    1 root     root          545 Oct 12 22:27 /etc/shadow
]-r--------    1 root     root          545 Oct 12 22:27 /etc/shadow-

]first, what are these secondary files? "passwd-" and "shadow-"?

]second, why is my entry the only one that differs between the two?
](entries below truncated for brevity and my peace of mind)

]       [root]# diff /etc/shad*
]       21c21
]       < ronho:$1$Krxwhgpn$Q(snip):11243:0:99999:7:-1:-1:134537756
]       ---
]       > ronho:$1$Ylp4SnO4$2(snip):11243:0:99999:7:-1:-1:134537756
]       
]how did i do this?

]third, and most confounding:

]a few days ago, i changed my password, by running "passwd" as myself.
]today i noticed that my old password still worked. let's say my old
]password was "foo," and my new password is "bar." so both "foo" and
]"bar" work for me. (HUH?)

]if i run passwd again, and change "bar" to "gack," then now i can log in
]using "foo" and "gack" (but not bar). somehow, the system allows me to
]use my old password even if my new password is changed.

]same thing happens if i run "passwd ronho" as root, and change "gack"
]to "oik." now "foo" and "oik" work.

]same thing happens if i:
]       - run pwunconv
]       - delete the "ronho" entry from the passwd file
]       - as root, run "rm -f /etc/shad*"
]       - run pwconv to restore the shadow passwords
]       - run useradd to recreate my account
]       - as root, do "passwd ronho" and set it to "baa"
]now "foo" and "baa" work. holy smokes.

]BUT...

]here's the kicker. my home directory is nfs-mounted. NONE OF THIS
]FOOLISHNESS WORKS IF THE HOME DIRECTORY IS NOT MOUNTED. i.e. if as root,
]i "umount -a," then login, i cannot use "foo"! i can only use "baa" (or
]whatever the latest correct password is).

]but as soon as i mount the home directory, the problem of the old
]persistent password reappears.

]what am i doing? i think it should be something simple, but i can't
]figure it out.

]thanks in advance!

]yours in frustration,
]ron

The shadow- should be a backup made when you changed your password. However
something strange is going on here. you should not be able to use the
old password as well.  If what you describe is actually correct, this
would indicate a severe bug somewhere. 
Where is your home directory being nfs mounted from?



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

From: [EMAIL PROTECTED] (Bill Unruh)
Crossposted-To: alt.os.linux.mandrake
Subject: Re: How will I restore back LILO
Date: 13 Oct 2000 16:14:15 GMT

In <[EMAIL PROTECTED]> YY Lee <[EMAIL PROTECTED]> writes:

]I'm sending my dual boot new Toshiba Tectra to Toshiba because the
]CD-ROM no longer works.  Me didn't tell them I left a CD in the drive
]for almost a month while mobilizing around.  The /dev/hda1 has Windows

Well, you have told them now. However, I do not see how that should
matter anyway.

]98 in the >2.1 GB while rest of 8.4 GB has Mandrake 7.1 using LILO as OS
]loader.  I'm gonna whack MBR so not to lose the warranty before shipping

What warrenty? Are you telling me that your warrenty card states that
the machine is only allowed to run Windows or the warrenty is void?
Computer hardware warrenties should not depend on the software run on
the machine!

]with MS-DOS 'fdisk /mbr' but not deleting its Linux contents.  How will
]I restore the LILO when I'll have it back?


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

From: [EMAIL PROTECTED]
Subject: Re: Maximum swap partition size?
Date: Fri, 13 Oct 2000 16:09:28 GMT

In article <[EMAIL PROTECTED]>,
<snip>
>
> I frequently see the recommendation to use about twice as much swap
space
> as you have main memory.
<more snip>

i also recall seeing that a few years (probably 4/5) ago when i started
using linux;  then i had a p75 box w/16MB of ram (those were the days!
:)

it does make sence to me (at that time anyway) that the swap be 2x
physical mem - this would allow almost everything to be swapped out.
the following shows my understanding ad yes yes yes, this is not quite
what happens, but it'll do as an example :)

presume swaping algorithm used ignores all the mem mgnt algorithms such
as page aging etc and presuming pages are only removed when the swap
space is full and then pages are removed judging by the length of time
they've been in swap

also presume that we have swap space same as phsy mem.

now, weve used all phys mem running app0.  we load app1.  presume all
the phys mem paged out.  ok, we load app2, and some pages belowing to
app1 have to be swapped out - we rm pages in swap beloning to app0 to
make room for pages of app1.

we use app0 again, but some pages are not in mem or swap. need to grab
then from disk again.

if the swap was 2x phys mem, then you;d have to have a lot of stuff
going to cause the above situation


these days however, unless youre running a your box as a big
server/heavy processing/many users, then i find that having 2x physical
mem a little too much since today we can get 512mb ram quite easily -
yes, yes, apps have grown and require more space but still... if i've
got it totally wrong, plese someone shoot me down - i'll be happy to be
enlightened :)


these days, i've still got a p200 w/64mb of ram and only a 48mb swap
partition - i dont run much on my box these days, pretty much home use,
but even when i'm running things like enlightenment, gimp, and
kernel/compiler builds at the same time, i havent seen the swap space
usage exceed 10mb (via `free`)

well done the kernel-mem-management folk i say! :)

my $0.02
p


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux,comp.os.linux.advocacy
Subject: Re: End-User Alternative to Windows
Date: Fri, 13 Oct 2000 11:36:24 -0500

On Fri, 13 Oct 2000 13:57:06 GMT, "Nicholas Knight" <[EMAIL PROTECTED]>
wrote:

>if you don't want to call DOS an operating system, you can't call linux an
>operating system, because they're more similar than you seem to be able to
>accept

Let him cry about how DOS isn't an OS all he wants.  Everyone
in the world knows better, and the only one being hurt by his
tantrum is himself.



-- 
Stephen Whitis
Email replies should go to...
scw120198 (at) whitis.com

The address in the header is not valid.

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

From: "Zoran P." <[EMAIL PROTECTED]>
Subject: AVI for LINUX
Date: Fri, 13 Oct 2000 18:33:33 +0000

Hi there,


who knows a AVI-Player for SuSE-Linux ??

xdvplay don't work...


Thanx for help !
-Zoran-

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

From: [EMAIL PROTECTED] (Dances With Crows)
Crossposted-To: comp.os.linux.networking
Subject: Re: Accessing certain sites fails SOLVED.
Date: 13 Oct 2000 16:44:20 GMT
Reply-To: [EMAIL PROTECTED]

On Fri, 13 Oct 2000 09:01:45 -0500, Clifford Kite wrote:
>Dances With Crows <[EMAIL PROTECTED]> wrote:
>
>> Attempting to access http://freshmeat.net/ from my Linux box fails.  At
>> the moment, I connect with PPP, and I can access just about any website
>> except that one.  Netscape, kfm, and Lynx all stall out while attempting
>> to get at the site.  Thinking it was a browser problem, I did this:
>
>This kind of problem is usually caused by mismatched MTUs Do not set
>the MTU or MRU for the PPP interface.  If the ISP is stupid enough
>to request a MTU for itself that's smaller than the 1500 default then
>use the pppd default-mru.

Got a message from Bill Staehle saying essentially the same thing.  A
couple of weeks back, my ISP was having fits, and the ppp documentation
I found suggested reducing the MRU.  I did so at about 12:30am one
night, and it didn't help.  The next day, everything worked appropriately 
(the problem was on their end) , except freshmeat.net, and I completely
forgot that I'd reduced the MRU.

The odd thing is that immediately after getting Bill's message, I edited
/etc/ppp/options to restore the MRU to its normal size, disconnected the
modem, then reconnected, and the problem was still there.  Repeating the
disconnect/reconnect made everything work.  Oh well, thanks!

-- 
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin /  Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com     /   Workin' in a code mine, whoops!
=============================/    I hit a seg fault....

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

From: [EMAIL PROTECTED]
Subject: ppp compression options
Date: Fri, 13 Oct 2000 16:39:02 GMT

in my /etc/ppp/options file i have a line that goes like this:

novj novjccomp nodeflate nopcomp nobsdcomp noaccomp

I added this after the advice from someone on one of the linux news
groups.  the purpose of this line was to speed up data rates.  The idea
is that since most isp's already compress the data before sending it,
then it redundant to compress it again.  And he was right, but only for
normal web browsing.  I am very pleased with my surfin' speeds from my
little 56K.

This is not true, however, for large downloads.  Whenever that
anti-compression line is active at dial up, my large downloads always
die before completion.  But if i comment this line in /etc/ppp/options
before dial up, then large downloads are fine and surfin is slow.

Does anyone know how to get around this catch 22?  maybe I have one too
many compression types turned off?

thanx much
-jason


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED]
Subject: Problems with Netscape on Linux
Date: Fri, 13 Oct 2000 16:40:44 GMT

System: RH 6.1, Netscape 4.75

I'm experiencing a couple of problems with Netscape and was wondering if
anyone else has had similar problems, and most importantly, has been
able to find a solution (other then the obvious one of simply not using
it).
The address book is very flaky and useless. I've imported a list of
addresses from my UNIX box at work and ever since, any attempts to enter
an addressees name alias fails. If I'm lucky, netscape will squawk about
"multiple entries" or in most cases fails to find the entry altogether
(while it is clearly in the address book and is not duplicated).
To make matters worse, if I open the address book to manually select an
entry by selecting a name and choosing the "to" button, Netscape simply
blows up. Gone, kaput, bus error, say goodnight! I can't find any other
form of diagnostic information to help narrow down the problem.
The composer is another issue, but I'll save this one for another
thread.
Appreciate any help on this,
Karl


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED] (Lew Pitcher)
Crossposted-To: comp.os.linux.questions
Subject: Re: help audio on linux
Reply-To: [EMAIL PROTECTED]
Date: Fri, 13 Oct 2000 16:53:57 GMT

On Fri, 13 Oct 2000 11:39:58 -0400, Xingzhi Zhang
<[EMAIL PROTECTED]> wrote:

>Dear Linux Fellows:
>
>I know this kind of question muste be asked million times:
>
>I have due booting system for my PC.
>
>I can play music under Windows through CD.
>
>But If I use the CD player in Redhat 6.1 I can not hear the music, the
>player shows every thing is fine, I also can hear it if I plug the speaker
>directly to the CD player, But I can not hear through sound card.
>
>I checked the sound card, it is connect to CD player. and it works under
>windows.
>
>Any idea ?

What volume is the CD feed set to (what does Xmixer show)?


Lew Pitcher
Information Technology Consultant
Toronto Dominion Bank Financial Group

([EMAIL PROTECTED])


(Opinions expressed are my own, not my employer's.)

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

From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: help with compiling new kernel for ibm laptop
Date: 13 Oct 2000 13:30:34 -0500
Reply-To: [EMAIL PROTECTED]

In article <8s7bgl$[EMAIL PROTECTED]>, Hung Ngoc Lai wrote:
> 1) download the kernel-2.2.17.tar.gz from http://www.kernel.org
> 2) download the patch for 2.2.17 from http://www.kernel.org

If you get the linux-2.2.17.tar.* file, that _is_ the source code for
2.2.17.  The patch-2.2.17.* file is the patch that takes 2.2.16 to 2.2.17.

> Everything seems to be working;  however, my NIC does NOT work.  
> I try to start the CARDCTL program.  The program tells me that the
> /proc/devices does not have pcmcia.

The Linux 2.2.* source code, i.e., the program released by Linus Torvalds
and Alan Cox, does not contain PCMCIA support.  That comes from David Hinds
at http://pcmcia-cs.sourceforge.net/.

-- 
Paul Kimoto
This message was originally posted on Usenet in plain text.  Any images, 
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.

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

From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy
Subject: Re: End-User Alternative to Windows
Date: Fri, 13 Oct 2000 18:05:50 GMT

In comp.os.linux.advocacy, Nicholas Knight
<[EMAIL PROTECTED]>
 wrote
on Fri, 13 Oct 2000 13:57:06 GMT
<SwEF5.2944$[EMAIL PROTECTED]>:
><snip>
>>Not quite.  DOS has never been an operating system;  it is just a program
>loader
>>and file system.
>>Calling DOS an OS because it loads first is like calling lilo an operating
>>system.
>
>DOS does the same damn thing linux does to boot..
>the BIOS loads whatever is in the MBR, and in turn the boot record on the
>DOS partition is loaded, which loads the DOS kernel, which loads command.com
>the DOS kernel includes support for things like hard drives and
>serial/LPT/keyboard ports
>the linux kernel boots the same way except it generaly consults a
>configuration file to know what to do after the kernel itself has booted
>
>if you don't want to call DOS an operating system, you can't call linux an
>operating system, because they're more similar than you seem to be able to
>accept

It may behoove us to hash out the term "operating system", admittedly.
>From one rather naive point of view, an operating system has
two characters.

(1) It operates.
(2) It is a system -- a series of parts which work together.

:-)

Both Linux and DOS satisfy these requirements.  To be sure, so does
an old manual typewriter -- which is basically a system of multijointed
levers attached to a series of keys on one side, and strike-heads on
the other, which hit a cylinder (platen?) through a ribbon.  (Usually
a piece of paper is between the ribbon and the cylinder.)

Some IBM units used a type ball -- the Selectric system was loved by
secretaries, as I recall.  Does this make them a non-typewriter?
Not at all, in most respects.  In many modern typewriters (which
are actually small computers hooked up to an impact printer), the
system is mostly electronic and include such things as spelling checkers.

Consider also an old daisy-wheel printer, which differs from a typewriter
in that the striking head does not have the letter imprinted on it;
instead, a rotatable wheel of plastic is inserted between the striking
head and the ribbon; the rotatable wheel has a large number of flexible
fingers which end with the font impression.  There are also old band
printers, where a large number of hammers strike at a high-speed rotating
band with the font letters on it, and dot-matrix printers, of which
there are two variants; one has a row of hammers across the entire page,
and another -- the consumer variety -- a small movable head.

All of these are systems, now more or less obsolete, for striking
a character through a ribbon onto a piece of paper.

Slightly more realistic examples of an "operating system", as we
know it today, would involve a piece of hardware -- the computer
and its peripherals -- and a promise to the user thereof that, if
one calls this routine, this happens.  The most-used name for that
is "API" (application procedure interface, IIRC), but a better name
might be "CPCI" (component procedure call interface); if I'm not
mistaken, CPCI is from an old military spec detailing how software
components interrelate.  This might be abbreviable to PCI, except
that someone's already taken that acronym.  Feh.

One might also consider such terms as "trap service vector" or TSV
(IIRC, a trap is a method by which a program can generate a special type
of exception, which allows the microprocessor to switch to privileged
mode and the OS to take full control) and "interrupt service vector"
(although it's not clear that microprocessors other than the ix86 use
the interrupt vectors for double-duty).

Note that Windows provides a hook routine, according to Andrew Schullman
(_Unauthorized Windows 95_) that allows a program to hook into
these interrupts.  I forget the name, though.

Another requirement might be that an OS inserts a barrier between the
hardware and the program attempting to get something accomplished
through the CPCI.  In short, can the program do "end runs" around
the OS in order to get what it wants done, without damaging itself,
the OS, or anything else?

Clearly, in the case of DOS,  the answer is yes -- and routinely, as well,
although one might ask if, say, Soundblaster's development library is
to be considered part of the CPCI.  But one can wiggle the ports
directly with ease -- to everyone's embarrassment, on occasion.
Ditto for video ram -- which may be one reason why it's now locked at
A000:0000 with such peculiar paging registers.  If the BIOS had had
a method by which to indicate the location of the video ram and
certain characteristics thereof, as opposed to "read a pixel, write
a pixel, clear an area", things might have been vastly different.
(But wishes are legion in this area; one never knows; might as well
wish that Motorola had gotten the 68000 out earlier with more
publicity, as well. :-) )

In the case of Unix (and Linux), the answer is no, a program cannot
get around the OS, at least not without extraordinary measures available
only to the superuser.  (Linux does support I/O to ports, but only
to root-enabled programs.)  While it can be done, it's much much harder.

Windows is somewhere in between, and rather complicated -- mostly because
Win95 sits on top of, straddles, or engulfs DOS (I'm not quite sure
which, but DOS knows Win95 is there, according to _Unauthorized Windows 95_)
but implements protected memory mode for its programs.  My understanding
is that Win98 and WinMe are similar, although I've not used either.

Windows NT, by contrast, has a kernel of its own.  While it may be
vaguely DOS-like in some respects (I don't know), it's clearly multitasking
(unknown how well, but more so than Win95, clearly), and allows for
the loading of DLLs, unlike DOS, which has all it can do to load EXEs.
Win2K is a Windows NT derivative, with many improvements -- at least,
such is my understanding -- but with essentially identical characteristics
to Windows NT.

Programs in Windows, as far as I know anyway, are far more likely to
use Windows-provided services to generate sounds (the most common
"violation" of DOS), rather than toggling the ports directly.
Ditto for video.  In fact, NT has protections similar to Linux in
that regard; user-level programs cannot access ports directly, if
I'm not mistaken.

To sum up:

[1] DOS is basically a glorified interrupt servicer and
    program loader.  It does not do memory or port protection.

[2] DOS+Windows is a full-fledged multitasking OS of its own, albeit not a
    very stable one.  Protections may not always be available, either.

[3] Windows NT is a full-fledged multitasking OS.  Protections
    are always available.

[4] Linux is a full-fledged multitasking OS.

-- 
[EMAIL PROTECTED] -- insert random misquote here

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

From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: Financial software ~ Quickbooks Pro ~ Linux?
Date: Fri, 13 Oct 2000 14:08:13 -0400

Frank Miles wrote:

> In article <[EMAIL PROTECTED]>,
> Jean-David Beyer  <[EMAIL PROTECTED]> wrote:
> >Frank Miles wrote:
> >
> >> In article <8s4i1f$ffi$[EMAIL PROTECTED]>, -ljl-  <[EMAIL PROTECTED]> wrote:
> >> >In article <[EMAIL PROTECTED]>,
> >> >  Jean-David Beyer <[EMAIL PROTECTED]> wrote:
> >> >> Brett Randall wrote:
> >> >>
> >> >> > Hi all
> >> >> >
> >> >> > I am looking to totally replace Windows with Linux. I have been
> >> >> > working with, administrating, programming and hacking up linux for
> >> >> > around 4 years now, but not once have I found a reasonable financial
> >> >> > accounting package for small-to-medium businesses.
> >>
> >> gnucash is pretty good, quite stable (not the development version), but
> >> probably not yet ready for business applications.
> >
> >Which version? I tried one a couple of months ago, presumably not a
> >developmental version (gnucash-1.3.7-1.i386.rpm), and never got it to run right.
> >It also did a lousy job (useless, IMO) of importing from Quicken. Problems
> >included double counting of things.
>
> The current version (1.4.6 or 1.4.7 ?) is considerably better.  Among
> the improvements IIRC was improving Quicken imports.  I never ran Quicken,
> so this information is based on their ChangeLog and my limited experience
> doing imports using the QIF format from CBB ... your mileage may vary.
>
> It's clear that gnucash is undergoing active development -- there's a fair
> amount of activity on their mailing lists.  The developers seem to have an
> ambitious plan for further development, in particular features more applicable
> to running businesses than personal finance.  IMHO it's quite usable now for
> personal finance, even if some features aren't as far along as I'd like.
>
>         -frank
> --

Is there any attempt to make it do federal and state income taxes? It used to take
me 4 long days to do my taxes manually, and just one intolerably long afternoon with
TurboTax. I see no indication that GnuCash will do tax returns. Does it have a
facility to produce a Quicken Master File so I can run TurboTax on it? Or does it
now do income taxes?

--
 .~.   Jean-David Beyer           Registered Linux User 85642.
 /V\                              Registered Machine    73926.
/( )\  Shrewsbury, New Jersey
^^-^^  2:05pm up 3 days, 19:44, 3 users, load average: 2.06, 2.09, 2.01




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


** 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: [EMAIL PROTECTED]

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

    Internet: [EMAIL PROTECTED]

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

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

Reply via email to