Linux-Networking Digest #746, Volume #10          Sun, 4 Apr 99 21:13:45 EDT

Contents:
  Re: Samba: Sharing a remote printer (Carles Arjona)
  Got 'dem DHCP Server Blues! (Michael)
  Re: can't rsh properly! (Carles Arjona)
  Compiling tulip.c (Matt Pokress)
  Re: I'm having a security problem (Job Eisses)
  Re: Don't wanna run 'diald', so what else??? (Scott Alfter)
  Re: Using Linux instead of NT Server in home environment.... ("Anthony W. Youngman")
  Security Question (Tom Tang)
  Re: 4mb Ram 386 router (Rod Roark)
  Re: Compiling tulip.c (John Strange)
  Re: pppd woes (the eternal saga) (S.)
  Re: pppd woes (the eternal saga) (Stuart Park)
  ppp 2.3.7 giving errors (Jack Bowling)
  Re: recursive catching of www-pages? (Luke Cyca)
  Re: RedHat 5.2 PPP hell (Peter Flynn)
  Re: [Help] PPP worked but now fails [modem q] (Peter Flynn)
  Re: PPP problems and Redhat 5.2 (Peter Flynn)
  Multiple ppp setups (Peter Flynn)
  Re: pppd woes (the eternal saga) (Bill Unruh)
  netscape ldap 'Unknown error' (Andrew Congdon)
  Re: pppd woes (the eternal saga) (Mike Graham)
  Re: MS-LINUX ("Spaasky")

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

From: Carles Arjona <[EMAIL PROTECTED]>
Subject: Re: Samba: Sharing a remote printer
Date: Sun, 04 Apr 1999 19:23:53 +0200

[EMAIL PROTECTED] wrote:
> 
> Is it possible to define a remote printer on Linux and then
> share that printer?

Yes it is. Just for testing purposes, I've tried this:

  Linux-2  ---> Linux-1 (samba) --->  WfW3.11 + HP LaseJet 4L

So I could print from Linux-2 with a printer attached to a Windows 3.11
box via Linux-1. (Linux-1 and Linux-2 were RedHat 5.2 boxes).

  W95 ---> Linux-1 (samba) --->  WfW3.11 + HP LaseJet 4L

worked also.

Regards.

Carles Arjona    [EMAIL PROTECTED]
-
Please, just remove the second NOSPAM for replies from newsgroups.
(Yes, NOSPAM is my real username)
[Don't send me e-mail copies of usenet postings, please]

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

From: Michael <[EMAIL PROTECTED]>
Subject: Got 'dem DHCP Server Blues!
Date: Sun, 04 Apr 1999 22:09:24 GMT

Hi-
Set up dhcp server on Redhat 5.2.This is a small lab of 8 machines,
mixed unix/linux/wintel.
The wintel machines obtain leases - no problem.
My SuSE linux box tries, but doesn't obtain a lease.
When the Suse client broadcasts, the server spews.....
 " DHCPDISCOVER from <macaddress> via eth0 "
 " DHCPOFFER on 11.11.11.71 to <macaddress> via eth0 "
 " DHCPDISCOVER from <macaddress> via eth0 "
 " DHCPOFFER on 11.11.11.71 to <macaddress> via eth0 "
........continuously.
So obviously the linux client is broadcasting a DHCP request, the server

is responding, but, for some reason, the lease offer is not being
noticed, or being rejected by the client. I can't find any log either.
Now, this client hooks up via DHCP without a problem at work (which is
running QIP DHCP) so I know the client CAN accept an address.
dhcpd.conf :
subnet 11.0.0.0 netmask 255.0.0.0 {
range 11.11.71 11.11.11.81;
   default-lease-time 600;
   max-lease-time 7200;
   option subnet-mask 255.0.0.0;
   option broadcast-address 11.255.255.255;
   option domain-name "eleven11.net";
}

Any ideas???
Thanks,
Michael




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

From: Carles Arjona <[EMAIL PROTECTED]>
Subject: Re: can't rsh properly!
Date: Sun, 04 Apr 1999 15:26:36 +0200

T. Sutherland <[EMAIL PROTECTED]> wrote:
>
> I have a very simple network setup.  I'm connecting my Pentium to a
> 486DX33.  Both are running RH 5.2.
<snip>
> From the 486, typing rsh nivdome [command] results in permission
> denied.  Typing rsh nivdome from the 486 pops up a password request
> that I have to answer before I can access the pentium.
>
> I'm lost.  Any suggestions?

Yes,

1- You might try:

On nivdome:
  echo  keldome >> /root/.rhosts

On keldome:
   rsh -l root nivdome ls -l /

2- Add a signature to your postings saying that .nospam must be removed.
Most of people don't like to look for something extrange into your
e-mail address in order to help you.

Regards.

Carles Arjona    [EMAIL PROTECTED]
-
Please, just remove the second NOSPAM for replies from newsgroups.
(Yes, NOSPAM is my real username)
[Don't send me e-mail copies of usenet postings, please]

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

From: Matt Pokress <[EMAIL PROTECTED]>
Subject: Compiling tulip.c
Date: Sun, 04 Apr 1999 18:28:53 -0400

I bought a Netgear FA-310-TX.  The Netgear instructions on their site
indicate that card revs higher than D1 (mine is a D2) need to use the
included tulip driver, not the standard tulip module.

The disk actually has tulip.c, so all I have is the source code.  How do
I compile this, and when I am done where will the new driver reside?


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

From: Job Eisses <[EMAIL PROTECTED]>
Subject: Re: I'm having a security problem
Date: Mon, 05 Apr 1999 00:29:19 +0200

Christopher Swanson wrote:
> 
> Hello All!
> 
> I just installed a Red Hat 5.1 system and accidentally ( but
> intentionally ) didn't install anonymous ftp. Unfortunately, this choice
> didn't install any ftp service. So, I took the steps to get an in.ftpd
> executable and restarted inetd. Now, I can connect with a login prompt
> but the system reports :
> "530 User cswanson access denied".
> I went through the shadow password conversion and the hosts.allow file.
> 
> To me this is too much security than I need.
> Does anybody have a hint on what I need to do?

Check the login shell - see "man in.ftpd" why        -job

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

From: [EMAIL PROTECTED] (Scott Alfter)
Crossposted-To: alt.os.linux,comp.os.linux.misc
Subject: Re: Don't wanna run 'diald', so what else???
Date: 4 Apr 1999 15:43:08 -0700

In article <[EMAIL PROTECTED]>,
Andre van Dijk <[EMAIL PROTECTED]> wrote:
>On Sat, 03 Apr 1999 04:48:58 GMT, David M. Cook wrote:
>>echo stuff to that file to control the connection (if you set permissions on
>>it correctly.)  
>
>There's also a X frontend for [diald.ctl]. Try dctrl

You can also use some cgi-bin scripts to control diald.  This has the
advantage of enabling you to control diald from any web browser.  I've set
IE5's homepage on my Win98 box to a page generated by my Linux box that
can send most commands to diald...force the link up or down, open a link,
close it, etc.  When I'm logged in from my Apple II (like I am now),
something like "lynx -dump http://localhost/cgi-bin/ppp-up" will do the same
thing.

The only catch is that diald's named pipe needs to be accessible to cgi-bin
scripts.  This might be a security hole on a multi-user system.  On a home
system with only one user (or very few users), it isn't that big a deal.

  _/_
 / v \
(IIGS(  Scott Alfter (salfter at (bitte keine Spam) delphi dot com)
 \_^_/  http://people.delphi.com/salfter


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

From: "Anthony W. Youngman" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Using Linux instead of NT Server in home environment....
Date: Sun, 4 Apr 1999 23:01:55 +0100
Reply-To: "Anthony W. Youngman" <[EMAIL PROTECTED]>

In article <7dmf1a$2u80l$[EMAIL PROTECTED]>, Stuart Fox
<[EMAIL PROTECTED]> writes
>
>>> Also, time for a few facts
>>>
>>> 1.  NO operating system is bug free
>>> 2. Both Linux camps and MS spend considerable time locating and fixing
>bugs
>>> 3. A properly configured NT box will not Blue Screen, and will be as
>stable
>>> as a well configured Linux box.

You're assuming the user has the experience to configure it properly. In
which case linux is well within his grasp ...
>>
>>Hmmm.     Not sure this is really the case, NT does Blue screen sometime
>>without explanation.     One thing I do know for sure is that Linux is a
>hell
>>of a lot easier to get working again if it fails to boot.    Scramble an NT
>>installation to much and its reinstall time.     A key Linux quality is the
>>ability to recover a system if sometthing goes wrong.    This can be very
>>difficult with NT.
>>
>
>I have some twenty or thirty NT servers under my direct or indirect
>influence, and they do not blue screen.  The occasions when they do, it is
>usually because a third party driver is poorly written (e.g. some of the
>Compaq NIC drivers).  In my experience, most NT blue screens are caused by
>hardware or hardware related faults (or letting some asshole who doesn't
>know what they're doing at your machine).  Recovering an NT box isn't that
>hard, especially given there's such a wide ranging knowledge base available.
>I have never seen a NT box blue screen without a good reason.
>
When I upgraded my NT system to SP4 (okay I think I did something
stupid) it blue-screened on boot. I couldn't recover and ended up using
fdisk deleting my C: drive in order to be able to do a re-install.

When I cocked up my upgrade from SuSE 5.2 to 6.0 it was pretty easy to
recover. The system at least managed to boot and let me get at the tools
I needed. By the way, linux had just done its equivalent of "del
\winnt\system32". I don't think NT would have recovered from that!

As for "linux is for techies", I spend too much time teaching my friends
"the fundamentals of computing" and hauling them out of holes of their
own making. I don't mind provided they want to learn. If you don't
understand what NT is doing "under the bonnet" then you are going to
abuse and break it. Linux just forces you to "wise up" rather quicker.

Like with cars, even if you don't do any maintenance yourself, a car
driven by someone who understands how the engine and transmission etc
work is almost certainly going to be more reliable than one driven by
someone without a clue. The knowledgeable person won't push the car
beyond its limits and will have problems nipped in the bud. The clueless
guy won't realise anything is wrong until he gets stranded by the side
of the motorway with a dead vehicle.
-- 
Anthony W. Youngman - wol at thewolery dot demon dot co dot uk
Trousers with a single hole in their waistband are topologically equivalent
to a doughnut. These sugarcoated trousers have yet to catch on at fast-food
outlets! (SuperStrings by F. David Peat)

If replying by e-mail please mail wol. Anything else may get missed amongst
the spam.


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

From: [EMAIL PROTECTED] (Tom Tang)
Subject: Security Question
Date: 4 Apr 1999 23:02:50 GMT


   Hello All :

     I am configuring a RH 5.2 system to act as a dual-homed
   system and have a few questions.  If IP-forwarding is 
   turned off, is source routing also disabled ?  Also,
   if all services in /etc/inetd.conf are commented out,
   are there still any remote exploits possible ?  Or
   should I just not start the inetd daemon ?  Thanks.

-- 

                                          Tom Tang   
                                          [EMAIL PROTECTED]  
                                          http://wwwcsif.cs.ucdavis.edu/~tangj 

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

From: Rod Roark <[EMAIL PROTECTED]>
Subject: Re: 4mb Ram 386 router
Date: 4 Apr 1999 23:11:53 GMT

Carles Arjona <[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] wrote:
>> 
>> On Sun, 21 Mar 1999 15:41:32 GMT, [EMAIL PROTECTED] (mike dombrowski)
>> wrote:
>> 
>> >I'm currently using a 386sx with 4mb ram to route between my two
>> >networks.
>> >
>> I think an older version of slakware should do the trick,

Perhaps even a newer version.  See
http://www.sunsetsystems.com/tip-lowmem.html for a possible solution.

-- Rod
======================================================================
Sunset Systems                           Preconfigured Linux Computers
http://www.sunsetsystems.com/                      and Custom Software
======================================================================


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

From: [EMAIL PROTECTED] (John Strange)
Subject: Re: Compiling tulip.c
Date: 4 Apr 1999 23:16:22 GMT

You realy need to tell us what distribution (Slackware, Redhat,..)
and what release (3.6, 5.2,...) so we can give more correct answers.

Try this as root

updatedb             # runs for about a minute, logs file locations.
locate tulip.c       # now you know where to put it.
cd /usr/src/linux    # all compiles for kernel and modules are here
make modules         # compiles any modules which have changed
make modules_install


Matt Pokress ([EMAIL PROTECTED]) wrote:
: I bought a Netgear FA-310-TX.  The Netgear instructions on their site
: indicate that card revs higher than D1 (mine is a D2) need to use the
: included tulip driver, not the standard tulip module.

: The disk actually has tulip.c, so all I have is the source code.  How do
: I compile this, and when I am done where will the new driver reside?


--
While Alcatel may claim ownership of all my ideas (on or off the job),
Alcatel does not claim any responsibility for them. Warranty expired when u
opened this article and I will not be responsible for its contents or use.

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

From: [EMAIL PROTECTED] (S.)
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions,comp.os.linux
Subject: Re: pppd woes (the eternal saga)
Date: Thu, 01 Apr 1999 18:28:00 -0900

I was able to finally and thankfully get pppd to work. Thanks every for
your help.

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

From: Stuart Park <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions,comp.os.linux
Subject: Re: pppd woes (the eternal saga)
Date: Sun, 04 Apr 1999 23:04:33 +0000

Bill Unruh wrote:
> >>Well, I've configured the proper files, I've contacted my ISP to find out
> >>whether it uses PAP or CHAP, set that stuff up, and I am still running
> >>into trouble.
> 
> >  Use the scripts.  Look in the directory /usr/doc/ppp-*.*.*/scripts and
> >you'll find a few demo scripts.  Copy ppp-on, ppp-off, and ppp-on-dialer
> 
> The problem is that those are old scripts which assume that
> authentication proceeds via logon. However many ISPs nowadays do not use
> login authentication. They use PAP or CHAP, and those scripts do not do
> much for that situation.So if you are lucky they work, if not you are
> again left high and dry.

Yes, I went through that situation..
i.e. an ISP that originally allowed login authentication but
then switched to mainly PAP.  So I had a good lengthy read through
the 'man' page for "pppd".  It describes all you have to do,
although is a bit cryptic.  Basically it involves setting up
a /etc/ppp directory and also a /etc/ppp/peers directory and
a chat script and a script file to start it all up.

If anyone wants me to, I'll post my configuration files.
( minus phone numbers and passwords of course  :) )


-- 
"You are not to watch this show.. it is immature toilet humour"  - South
Park

Stuart Park
E-Mail: [EMAIL PROTECTED]          Melbourne, Australia

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

From: [EMAIL PROTECTED] (Jack Bowling)
Subject: ppp 2.3.7 giving errors
Date: Sun, 04 Apr 1999 20:55:05 GMT

Just compiled and installed the latest ppp 2.3.7 package to see if it 
would get rid of the stalling that has been occurring on my IP 
connections. Unfortunately, it appears that there is some default 
setting that has either changed or I have not yet set properly because
Iget the following on establishing a ppp connection:

Apr  4 13:37:47 localhost pppd[385]: Protocol-Reject for unsupported 
protocol 0x5072
Apr  4 13:37:47 localhost pppd[385]: Protocol-Reject for unsupported 
protocol 0x7263
Apr  4 13:37:47 localhost pppd[385]: Protocol-Reject for unsupported 
protocol 0x5072
(repeated for life of connection - message given about 4/sec)
(killed by user)
Apr  4 13:37:48 localhost pppd[385]: Terminating on signal 15.
Apr  4 13:37:49 localhost pppd[385]: Connection terminated.
Apr  4 13:37:49 localhost pppd[385]: Connect time 0.4 minutes.
Apr  4 13:37:49 localhost pppd[385]: Sent 95599 bytes, received 102151
bytes.
Apr  4 13:37:49 localhost pppd[385]: Hangup (SIGHUP)
Apr  4 13:37:50 localhost pppd[385]: Exit.
Apr  4 13:39:18 localhost kernel: PPP: ppp line discipline 
successfully unregistered 

The docs for 2.3.7 state that default asyncmap is now set to 0. Could 
this be the problem? Any help greatly appreciated. Running RH 5.2 with
KDE.

=============
Jack Bowling
[EMAIL PROTECTED]

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

From: [EMAIL PROTECTED] (Luke Cyca)
Subject: Re: recursive catching of www-pages?
Date: Sun, 04 Apr 1999 17:50:42 -0600

In article <[EMAIL PROTECTED]>, Jens Hauser
<[EMAIL PROTECTED]> wrote:

> Hi!
> 
> Is there a tool which is able to download a specific
> www-page and all of its links below (recursive) ?
> 
> Jens

For the mac there's Anarchie (www.stairways.com)

-Luke

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

From: Peter Flynn <[EMAIL PROTECTED]>
Subject: Re: RedHat 5.2 PPP hell
Date: Mon, 05 Apr 1999 01:17:17 +0100
Reply-To: [EMAIL PROTECTED]

Basil Blume wrote:
> 
> I used the linuxconf program that comes with the Red Hat distribution to
> configure my ppp and it works great.  Read the section on adding modem/ppp
> conections in the "The Complete Red Hat Linux 5.2 Installation Guide."

I used linuxconf and it made a complete and utter mess of it. The
chat-ppp0
it created made all sorts of unwarranted assumptions about what my modem
would be replying with, and how my ISP would respond, and editing the
"expect" "send" pairs in linuxconf only messed it up even more.

Moral: use linuxconf to set it up, but then watch /var/log/messages
like a hawk and edit chat-ppp0 by hand to make it work.

///Peter
--
<?xml version="1.0"?><?xml-stylesheet href="sig.css" type="text/css"?>
<Sig><Name>Peter Flynn</Name><Email>[EMAIL PROTECTED]</Email></Sig>

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

From: Peter Flynn <[EMAIL PROTECTED]>
Crossposted-To: 
comp.os.linux.setup,comp.os.linux.help,comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: [Help] PPP worked but now fails [modem q]
Date: Mon, 05 Apr 1999 01:23:55 +0100
Reply-To: [EMAIL PROTECTED]

Matt wrote:

[...]
> Apr  3 08:15:27 compuserve chat[200]: ATDT08450801000^M^M
> Apr  3 08:15:27 compuserve chat[200]: CONNECT -- got it
> Apr  3 08:15:27 compuserve chat[200]: send (^M)
> Apr  3 08:15:27 compuserve chat[200]: expect (Host Name:)
> Apr  3 08:15:27 compuserve chat[200]:  49333/ARQ^M
> Apr  3 08:16:12 compuserve chat[200]: ^\@`^N^\^\GG^\G^N^\@

You need an additional line in chat-ppp0 right after the one with 
CONNECT in it, saying:

'49333/ARQ'  ''

This is what I meant when I posted 2 mins ago about the autoconfig
messing things up. The chat scripting doesn't seem to allow for
wildcards, if-switching or anything useful at all (or if it does, 
it's not documented anywhere I can find). It's lunatic in 1999
having to rely on "expect" "send" pairs instead of the programmable 
dialog you get on otherwise inferior operating systems beginning 
with a "W" :-)

///Peter
--
<?xml version="1.0"?><?xml-stylesheet href="sig.css" type="text/css"?>
<Sig><Name>Peter Flynn</Name><Email>[EMAIL PROTECTED]</Email></Sig>

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

From: Peter Flynn <[EMAIL PROTECTED]>
Subject: Re: PPP problems and Redhat 5.2
Date: Mon, 05 Apr 1999 01:28:44 +0100
Reply-To: [EMAIL PROTECTED]

[EMAIL PROTECTED] wrote:
> 
> Something MUST be wrong in RH 5.2 because I've spent many nights trying to
[...]
> BTW, I'm using the same exact setup procedures in 5.2 that I've been using
> under 5.1, but PPP under 5.2 has never worked (LCP errors, IPCP errors, etc.)
> and sometimes when it connects, none of my network apps like netscape, ftp,

I think someone screwed with the autogenerated scripting controls. Check
your
chat-ppp0 against /var/log/messages and add extra "expect" "send" pairs
to
handle the stuff it messes up over.

The whole chat stuff seems to have been written by someone who had heard
login dialog scripts described, but had never actually seen or used one.

///Peter
--
<?xml version="1.0"?><?xml-stylesheet href="sig.css" type="text/css"?>
<Sig><Name>Peter Flynn</Name><Email>[EMAIL PROTECTED]</Email></Sig>

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

From: Peter Flynn <[EMAIL PROTECTED]>
Subject: Multiple ppp setups
Date: Mon, 05 Apr 1999 00:55:16 +0100
Reply-To: [EMAIL PROTECTED]

I've just upgraded RH 5.1 --> 5.2 and now I want to be able to 
dial out to more than one ISP (using ppp) from this lapto, but
it insists on trying to use ppp0 setupd when I try to establish 
the ppp1 link.

The machine has a modem and an Ethernet card, and I use both
(but only one at a time, via Usernet :-) so my configuration
has settings for 

a. eth0 as Adaptor 1 (using a static IP address, hostname, and 
   set of DNS IP addresses and a gateway for use when the 
   machine is connected to the office LAN, works fine).

b. ppp0 when I work from home, dial into the office LAN as a
   login PPP connection into the back of my Sun, so I use the 
   same static IP addres, etc etc as (a) above. This also works 
   fine.

I added a ppp1 for dial-up to a local ISP, but found that there
is (i) nowhere to specify DHCP for this phone number, (ii) also
nowhere to put the DNS addresses which are to be used on this
connection, and (iii) when I do try to connect, /var/log/messages 
shows it dialing and connecting the modems OK, but then "Using 
interface ppp0" even though it was explicitly started by 
clicking ppp1 in Usernet. So of course it fails to make a PPP
link, and dies thinking "Serial line is looped back".

1. How do I force Usernet ppp1 to use my ppp1 setup instead of ppp0?

2. Where do I tell it to use dynamic IP addressing for ppp1?

3. Where do I put the DNS IP addresses for ppp1? (And for that
   matter, where do I tell it to use another gateway?)

I know making multiple IP setups is a nightmare under W*nd*ws 9x
but I'd assumed that Linux would be able to handle it better.

///Peter
--
<?xml version="1.0"?><?xml-stylesheet href="sig.css" type="text/css"?>
<Sig><Name>Peter Flynn</Name><Email>[EMAIL PROTECTED]</Email></Sig>

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

From: [EMAIL PROTECTED] (Bill Unruh)
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions,comp.os.linux
Subject: Re: pppd woes (the eternal saga)
Date: 4 Apr 1999 20:57:23 GMT

In <yAJN2.186$[EMAIL PROTECTED]> [EMAIL PROTECTED] (Mike Graham) 
writes:

>On 4 Apr 1999 02:20:31 GMT, Bill Unruh wrote:

>>The problem is that those are old scripts which assume that
>>authentication proceeds via logon. However many ISPs nowadays do not use
>>login authentication. They use PAP or CHAP, and those scripts do not do
>>much for that situation.So if you are lucky they work, if not you are
>>again left high and dry.

>  Well, maybe I'm just lucky that my ISP doesn't use PAP or CHAP, or maybe
>if they did then I would have had a working connection a lot sooner.  All I
>know is that this works, and it's virtually the only thing that does.  I had
>to set KPPP for login scripting as well; PAP only worked some of the time,
>the login scripting works *every* time.  There's something to be said for
>that.

Yes, you are lucky. I have seen a number of ISPs where it seems you
canlogin, but you cannot. You are left in a limbo if you try. They want
PAP CHAP right after the CONNECT message. In fact Linux with mgetty and
AutoPPP works exactly like that. If youtry to login to a Linux box using
AutoPPP there is a very ood chance you will be out of luck.

The ISPs right now are in transition. Win supports PAP CHAP as the
default, and most ISPs are switching over. Sometimes they have an
intermediate scheme where both work, but then they suddenly pull the
plug on login authentication, and leave people in the lurch.



>  As I've griped before, it annoys me that there are several ways to do
>everything, instead of one right-and-proper-and-works way.  It seems that

I'm sorry, what is the "right and proper" way? Every ISP uses a
different authentication scheme. I have seen one who demanded both login
authentication and then after that PAP authentication. Until all the
ISPs in the world get together and decide on one way of authenticating
(not likely to happen soon. Microsoft for example is threatening to
introduce still another completely incompatible scheme), there simply is
no "right and proper" way.

My page was an attempt to give a logical and consistant way of stepping
through all the options to see which one worked. If you have a shortcut
which worked for your ISP, that is great. However, I assure you your
scheme will NOT work for all ISPs.

>every HOWTO and .doc you read has a different way of going about whatever it
>is you're trying to do.  Very frustrating for the newbie.


I agree. And the problem is that most of those HOWTOs were written a
couple of years ago when login authentication was almost universal. That
has changed. Now many use PAP or CHAP05 or CHAP 80 (or even CHAP81--
Microsofts latest stupidity). And often they do not know themselves what
they use. Probably someone did once, but they have left the
organisation. Or they simply follow some directions they got from MS (or
from mgetty) on how to set up their system, and have little idea what
they are actually doing. So the linux user, who cannot simply download
or be given some incomprehensible package to log onto the ISP as the
Win9x user can, has to figure out what the ISP is doing in order to log on.

Most of the scripts and the HOWTOs make assumptions about what the ISP
wants (or tell the user to find out by asking the ISP what they want-- a
silly suggestion since most ISPs do not know), and for ISPs for whom the
assumptions are right, the scripts are great and work well. For the
others, they do not work, and intense frustration results.

That frustration grows when they are told in newsgroups to use script X
or Y or Z, and when they try it fails for their ISP.


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

From: Andrew Congdon <[EMAIL PROTECTED]>
Subject: netscape ldap 'Unknown error'
Date: Mon, 05 Apr 1999 09:59:44 +1000

I'm getting the following error when using the Netscape 4.51 on
Linux 2.2.x when doing an addressbook match using LDAP (running
on another Linux box):

Failed to search '[directory service name]' due to LDAP error
'Unknown error' (0xFFFFFFFF)

It doesn't seem to occur on other Unix versions nor with Windoze.
It does occasionally work when doing address matching in email (I
never noticed it work in 4.50) but I haven't noticed any logic to
when it works.

Anyone else got it working?

--
Andrew

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

From: [EMAIL PROTECTED] (Mike Graham)
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions,comp.os.linux
Subject: Re: pppd woes (the eternal saga)
Reply-To: [EMAIL PROTECTED]
Date: Sun, 04 Apr 1999 23:41:41 GMT

On 4 Apr 1999 20:57:23 GMT, Bill Unruh wrote:

>I'm sorry, what is the "right and proper" way? Every ISP uses a
>different authentication scheme.

  I don't mean specifically logging into an ISP, I mean just about anything
that you want to set up in linux seems to have a bunch of methods depending
on how old the HOWTO is.  I can't tell you how many times I've struggled to
get something set up only to be told "Oh, I don't use *those* files, I use
these over here; those never worked for me."  I'd rather there was one
proper way to set up each thing.  It's a character flaw of mine.  8-)

>Most of the scripts and the HOWTOs make assumptions about what the ISP
>wants (or tell the user to find out by asking the ISP what they want-- a
>silly suggestion since most ISPs do not know), and for ISPs for whom the
>assumptions are right, the scripts are great and work well. For the
>others, they do not work, and intense frustration results.

  I'm not only talking about ISP stuff, though.  Directory standards are all
over the map.  Your XFree86 stuff might be in /usr/lib/XF86/XF86R6/X11/...
or somewhere totally different.  Your news might be in /var/spool/news or in
/var/spool/slrnpull depending on what package you use.  A bit of
standardization would go down well over here.

>That frustration grows when they are told in newsgroups to use script X
>or Y or Z, and when they try it fails for their ISP.

  That's why I try to be careful to say 'this is what worked for me' rather
than 'this will work for you'.  Besides, it takes all of 10 minutes to try
it.

-- 
Mike Graham, [EMAIL PROTECTED]
Caledon, Ontario, Canada (just NW of Toronto).

Raiser of animals.  Weldor of metals.  Driver of off-road vehicles.
Writer of FAQs.  Keeper of the faith, and all around okay guy.

<http://www.beeline.ca/personal/mike>

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

From: "Spaasky" <[EMAIL PROTECTED]>
Crossposted-To: 
alt.linux,alt.linux.sux,alt.os.linux,alt.os.linux.caldera,comp.os.linux.hardware,comp.protocols.smb
Subject: Re: MS-LINUX
Date: Mon, 5 Apr 1999 16:36:15 -0700


d s f o x @ c o g s c i . u c s d . e d u (David Fox) wrote in message ...
:[EMAIL PROTECTED] (Greg Fischer) writes:
:
:> What a hoot!
:> thanks for the laugh. I don't know if its the work of one sharply witted
:> person or a group effort, but it was funny...
:> April Fools
:
:Its not all that funny.  In the current PC Week Dvorak says MS-Linux
:is fairly likely.

No, he said that he would be "impressed" if MS made linux the core of the
next generation OS. This does sound like an April Fool's joke.

:--
:David Fox           http://hci.ucsd.edu/dsf             xoF divaD
:UCSD HCI Lab                                         baL ICH DSCU



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


** 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.networking) 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-Networking Digest
******************************

Reply via email to