Linux-Networking Digest #787, Volume #11 Mon, 5 Jul 99 10:14:08 EDT
Contents:
Setting up printer under Samba ("Mark Six")
Re: Newbie problems with ethernet (David Carns)
Re: Non-typical firewall IP interfaces numbers (Robin Putzar)
Routing question ("Carl Filpo")
Re: Non-typical firewall IP interfaces numbers (Robin Putzar)
IP Masquerading ([EMAIL PROTECTED])
Re: how to connect 2 Linux PC's directly? (Lew Pitcher)
PPP PAP NT RAS Authentication problem under RedHat 6.0 ("Peter Young")
PCMCIA Token Ring Problems (Gabriele Bilz)
Routing Question ("Carl Filpo")
Re: d-link card (Rod Smith)
Re: PPP connect: ping & traceroute but no data?? (G Crowder)
Re: DNS setup problem ("Michael Faurot")
Re: Using two LNE100TX NICs, adapter module fails? (Rod Smith)
Re: IPChains Problem ("Carl Filpo")
----------------------------------------------------------------------------
From: "Mark Six" <[EMAIL PROTECTED]>
Subject: Setting up printer under Samba
Date: Mon, 5 Jul 1999 20:16:29 +0800
Dear guys,
I got a HP Jet Direct Print Server with lpd support which no driver support
under Redhat 6.0. I tried to setup the printer as remote printer using
printtools, it is ok to print the text and postscript test page. But when I
load the printcap from samba and make it shareable, and also install the
correct printer driver from HP on windows 95/98 workstations, but cannot
print from the workstations.
Anything I did wrong? I think if there is correct printer driver on windows
95/98 workstations, and using Linux to print the raw data to the printer, it
should work, isn't it? Because I am unable to attache the printer directly
to the lpt port of the Linux machine, but is it a problem?
thanks.
=============
Mark Six
------------------------------
From: David Carns <[EMAIL PROTECTED]>
Subject: Re: Newbie problems with ethernet
Date: Mon, 05 Jul 1999 07:21:00 -0500
I don't know if this will solve all of your problems, but this should
help you diagnose what's happening with the modules for your network
card.
Once again, try loading the module for you card. (ie, modprobe rtl8139).
Then type "lsmod" (list modules) and see if the module was really
loaded. If the module is listed, then type "ifconfig" and see if eth0
comes up. If it does, then you should be able to run netcfg can
configure the hardware for IP, Broadcast, etc.
If the module does not get loaded, try loading "modprobe ne." I'm
guessing your card might also be a NE2000 compatible card. Then lsmod
again to see if that module loaded...if yes, then run ifconfig or netcfg
again and I think you'll have some success.
Finally, if none of this helps, make sure your network card is not
conflicting with other hardware. (ie, check you irq / io settings).
You may need to reset your hardware configuration.
"Chanodin L. Rahl" wrote:
>
> Where to start...
>
> I've been attempting to get my cable connection working for the last half-
> week; ever since I finished installing Mandrake 6.0 on my computer. [Specs
> as follows:
> Intel Celeron 333
> 64 MBSDRAM
> CL VooDoo Banshee 16MB
> Acer BH6 Mainboard
> ]
>
> The network card that I had did not work. So I purchased a new one, and at
> the store specifically asked for one that would work under linux.
>
> I purchased an Ovislink RealTek RTL8129/8139 Fast Ethernet Adapter.
> Under protocol for this it says CSMA/CD. Under standard it says IEEEF802.3
> and IEEFE802.3U
>
> Under drivers support including the relevent part is SCO Unix: 8029R,
> 8129TX
>
> I was told on an IRC channel to type modprobe rtl8139 and then run netcfg.
> I typed modprobe rtl8139. Nothing appeared to happen [a new command line
> came up], but I took that for normal for that command, started x and under
> konsole typed netcfg.
>
> The information I entered was based entirely off what I derived from
> winipcfg. The only one question I in netcfg was about gateway. Are gateway
> and default gateway the same, ie when there is a blank line beside either
> one asking for information do I enter the same IP for both?
>
> I remember reading somewhere on linux.com that this ethernet card is not
> eth0, but apt0. If this is one of the trouble causing factors, please
> notify me.
>
> I run netcfg, supposedly entering the correct information. However when I
> save and exit, under konsole, something along the lines of cannot detect
> eth0 is displayed. I cannot connect to the net still.
>
> I've followed what the LDP says, and what the linux.com guide says. I've
> been using linux for less then a week, and what I really would appreciate
> is someone to explain the entire process of setting up internet access to
> me, as opposed to saying "consult the net3 howto". I have done that already
> so many many times.
>
> Following is the information displayed under winipcfg:
> Host name: chanodin.cgocable.com
> DNS Servers: 192.168.0.1
> Node Type: broadcast
> NetBIOS Scope Id: [blank]
> IP Routing enabled: [blank] wins proxy enabled: [blank]
> NetBIOS Resolution uses DNS: checkmark
>
> Adapter address: 00-4F-4E-00-10-98
> IP Address: 24.226.19.89
> Subnet mask: 255.255.248.0
> Default gateway: 24.226.16.1
> DHCP server: 24.9.2.70
> primary wins server: [blank]
> secondary wns server: [blank]
------------------------------
From: Robin Putzar <[EMAIL PROTECTED]>
Subject: Re: Non-typical firewall IP interfaces numbers
Date: Mon, 05 Jul 1999 13:31:00 +0200
Reply-To: [EMAIL PROTECTED]
I have a similar problem, maybe much simpler.
I have a SparcStation under Linux that I want to connect to the internet.
The problem is, that it has a twisted-pair network card, whereas the whole
other network is BNC. Now I want to use a PC as bridge (and firewall). Both
machines have their IP address.
Here�s my current configuration:
gateway
130.149.35.1
|
+- other hosts
| 130.149.35.x
|
linux box A
130.149.35.44, eth0
130.149.35.44, eth1
|
linux box B
130.149.35.82, eth0
Now the symptoms:
- ping from linux box A to gateway is ok.
- ping from linux box A to linux box B and vice versa is ok.
- ping from linux box B to gateway reaches the gateway. The gateway replies
and the reply reaches linux box A. It is not forwarded to linux box B,
however.
Any suggestions?
Robin Putzar
------------------------------
Reply-To: "Carl Filpo" <[EMAIL PROTECTED]>
From: "Carl Filpo" <[EMAIL PROTECTED]>
Subject: Routing question
Date: Mon, 5 Jul 1999 20:38:39 +0800
I have the following setup:
- 1 Static IP number assigned to me when I dial-up (139.130.x.x Netmask
= 255.255.255.192)
- A range of real ip numbers assigned for my LAN
(y.y.y.228 - y.y.y.231
Netmask = 255.255.255.252
Network = y.y.y.228
Broadcast = y.y.y.231) - This gives me 2 hosts to
assign real IPs to
- A LAN with private ips 192.168.0.1 - 192.168.0.10
I have a Debian 2.1 box kernel 2.0.36 running ppp and ipmasquerading
successfully through the ppp
interface set to 139.130.x.x (above).
I would like to run a mail server so users can have: [EMAIL PROTECTED] but
the isp won't allow me to
set the static dialup ip (139.130.x.x) to map to "mydomain.com" in the dns -
this is the dial-up ip only - They
tell me I must use one of the real ip's - y.y.y.229 or y.y.y.230.
I would like to know how i can set up routing or forwarding or whatever it
takes to get machines on the LAN
to maintain the ip masquerading functionality whilst still be able to have
outside people send email to
mydomain.com mapped to y.y.y.229.
Currently:
It seems I have to get the ppp interface on 139.130.x.x to route traffic
through to the network y.y.y.228.
I have 2 NICs in the debian server at the moment - one set to 192.168.0.30
Netmask=255.255.255.0
on a LAN segment with other 192.168.0 PCs.
Another set to y.y.y.229 Netmask = 255.255.255.252 on another segment with
another potential host y.y.y.230
------------------------------
| Linux |
| | y.y.y.230
| y.y.y.229 (eth0) |-------||------- ## Y.Y.Y.228 ##
| |
| 192.168.0.30 (eth1) |----------------## 192.168.0.0 ##
| |
| x.x.x.x (ppp0) |---------------- ## INTERNET ##
| |
================================
I would like to set y.y.y.229 to mydomain.com in the dns but
at the moment I can't see y.y.y.229 from the Internet. This is my problem.
--
Carl Filpo
Computer Network Consultant
=================================================
Carl Filpo BSc(Curtin)
Computer Network Consultant
C&M Computer Solutions
26 Russell St
MORLEY WA 6062
Email: [EMAIL PROTECTED]
Phone: +61 8 9375 1144
Fax: +61 8 9375 1141
Mobile: 0407 396 316
=================================================
------------------------------
From: Robin Putzar <[EMAIL PROTECTED]>
Subject: Re: Non-typical firewall IP interfaces numbers
Date: Mon, 05 Jul 1999 14:46:44 +0200
Reply-To: [EMAIL PROTECTED]
Frank Waarsenburg wrote:
> Since eth0 and eth1 have the same IP number, box A is not able to route between
> them. So, the reply from the gateway will not pass box A. You can create a
> small subnet at the side of box B and tell the gateway to use eth0 at box A for
> forwarding to box B.
As a first try, I have changed the IP number of eth1 in box A to 192.168.35.44
(with netmask 0.0.0.0, broadcast 255.255.255.255). Now I can ping both interfaces
of box A from box B. But it�s still not possible to ping the gateway. (The
packages are still lost on box A.)
I do not have direct access to the gateway. Is there a way to tell box A it should
forward all packages for box B to box B?
Thanks for the answer.
Robin Putzar
------------------------------
From: [EMAIL PROTECTED]
Subject: IP Masquerading
Date: Mon, 05 Jul 1999 10:53:19 GMT
Hi,
I'm trying to set up a firewall, to sit between
our Class C network and the internet.
I want all our current IPs to stay the same, but I
also want to be able to allow access to some of
them from the internet. I've been trying all
kinds of ipfwadm commands, and am able to use,
say, telnet from our network to the internet.
However, I need the firewall to masquerade our IPs
for incoming connections.
Can this be done?
Please CC any replies to my email address.
Many Thanks!
Chris Malcolm
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: how to connect 2 Linux PC's directly?
Reply-To: [EMAIL PROTECTED]
Date: Mon, 05 Jul 1999 12:50:26 GMT
You have a couple of choices (based on your diagram)...
1) Install a second NIC in your "PC A" machine, and cable the
"PC B" machine's NIC to it (use 10BaseT xover or 10Base2), or
2) Install a hub between the NIC in your "PC A" machine, and your
DSL Digital modem. Cable all three devices ("PC A", "PC B", and
"DSL Modem) to the hub, using 10BaseT straight-through cable.
- Both solutions are about equal in price.
- A hub would permit you to expand your 10BaseT network without
expensive rewiring. However, traffic on your internal network
would be detectable (via a sniffer) outside of your network.
- Cabling "PC B" directly to "PC A" via a 2nd NIC would isolate
your internal network from the outside network. However, if you
wish to expand your internal network, you'll have to add either
a hub (if you use 10BaseT) or more cable (if you use 10Base2).
On 4 Jul 1999 19:33:50 -0700, [EMAIL PROTECTED] wrote:
>hello network experts.
>
>simple problem I hope.
>
>I'd like to connect my 2 PC's at home together while one of them remain
>connected to the net. Do I need to buy some kind of hub
>concentrator device? (I allready have one connected using DSL via
>the phone company, and I just brought a new PC).
>
>diagram below.
>
>I'd like to keep PC/A connected to network. But PC B I do not
>need to be connected to outside network (so I can give it any
>IP address I want), I just need it to be connected to PC/A to be able
>to ftp files between the 2.
>
>(I can offcourse connect the two PC's directly, and I did that,
>and that worked (I needed to remove the router gateway entry from
>PC/A every time though) but if I do that, then I would not be able to
>use PC/A on the network at the same time, without having to plug the correct
>cable in, and set the router for the correct gateway every time.
>
>I though may be I can add a second Ethernet card to PC/A
>and use that to connect PC/B into.
>
>this is my current configuration
>
>
>
> +-------------+ +------------------+
> | PC A | | PC B |
> | | | |
> +-Ethernet----+ +-- Ethernet card--+
> Card |
> | |
> | 10baseT cable TO A <---+
> | how??
> +-------------+
> | DSL digital |
> | modem |
> +-------------+
> |
> | phone line
> |
> |
>
>
>thanks for any advice.
>Kirck
>
Lew Pitcher
System Consultant, Integration Solutions Architecture
Toronto Dominion Bank
([EMAIL PROTECTED])
(Opinions expressed are my own, not my employer's.)
------------------------------
From: "Peter Young" <[EMAIL PROTECTED]>
Subject: PPP PAP NT RAS Authentication problem under RedHat 6.0
Date: Mon, 5 Jul 1999 22:54:47 +1000
Hi,
I've just started using a new ISP but I'm having problems getting them to
authenticate. The modem dials and connects fine but after a bit of data
movement the connection drops. The username and password are correct. I've
used netcfg in X to configure the dialup interface and have selected the PAP
Authentication option. I;ve even installed des-lib and ppp-2.3.8.
I have tried using a terminal dialup to see what's happening but it's all
just characters every where.
Can anyone help. My flat mate is getting rather annoyed! btw. it does
connect under WinBlows 2000 but comes up with MS Compression.
Thanks,
Peter . . . .
------------------------------
From: Gabriele Bilz <[EMAIL PROTECTED]>
Subject: PCMCIA Token Ring Problems
Date: Mon, 05 Jul 1999 15:00:20 +0100
Hi Everybody,
I have installed SUSE Linux 6.1. on my laptop (IBM770) and I'm trying
now to reach the network via the PCMCIA IBM Token Ring 16/4 PC-Card.
Unfortunately not very successfully ........
If I boot my laptop with the card connected to the ring, the boot stops
during the ./network start tr0 process. It seems to be because
"ifconfig" sends back a zero Hardwareadress (as far as I understand
...).
Booting my laptop without card, the computer boots without errors
(except the message insmod tr0 failed). If I insert my card after that,
the card ist recognized and the module ibmtr_cs.o is loaded. "Ifconfig"
does not give back anything, seems to be in a loop. In "HOW_TO - Pages"
I read something about changing high memory window, which I did.
The corresponding line in /etc/pcmcia/config.opts looks now:
include memory 0xc0000-0xfffff, memory 0x60000000 - 0x60ffffff
But reaching network is impossible.....
Is there anybody who can help me (if it is possible to understand my
problem ;-) ....) ??
Bye,
Gabi
------------------------------
Reply-To: "Carl Filpo" <[EMAIL PROTECTED]>
From: "Carl Filpo" <[EMAIL PROTECTED]>
Subject: Routing Question
Date: Mon, 5 Jul 1999 21:10:33 +0800
I have the following setup:
- 1 Static IP number assigned to me when I dial-up (139.130.x.x Netmask
= 255.255.255.192)
- A range of real ip numbers assigned for my LAN
(y.y.y.228 - y.y.y.231
Netmask = 255.255.255.252
Network = y.y.y.228
Broadcast = y.y.y.231) - This gives me 2 hosts to
assign real IPs to
- A LAN with private ips 192.168.0.1 - 192.168.0.10
I have a Debian 2.1 box kernel 2.0.36 running ppp and ipmasquerading
successfully through the ppp
interface set to 139.130.x.x (above).
I would like to run a mail server so users can have: [EMAIL PROTECTED] but
the isp won't allow me to
set the static dialup ip (139.130.x.x) to map to "mydomain.com" in the dns -
this is the dial-up ip only - They
tell me I must use one of the real ip's - y.y.y.229 or y.y.y.230.
I would like to know how i can set up routing or forwarding or whatever it
takes to get machines on the LAN
to maintain the ip masquerading functionality whilst still be able to have
outside people send email to
mydomain.com mapped to y.y.y.229.
Currently:
It seems I have to get the ppp interface on 139.130.x.x to route traffic
through to the network y.y.y.228.
I have 2 NICs in the debian server at the moment - one set to 192.168.0.30
Netmask=255.255.255.0
on a LAN segment with other 192.168.0 PCs.
Another set to y.y.y.229 Netmask = 255.255.255.252 on another segment with
another potential host y.y.y.230
------------------------------
| Linux |
| | y.y.y.230
| y.y.y.229 (eth0) |-------||------- ## Y.Y.Y.228 ##
| |
| 192.168.0.30 (eth1) |----------------## 192.168.0.0 ##
| |
| x.x.x.x (ppp0) |---------------- ## INTERNET ##
| |
================================
I would like to set y.y.y.229 to mydomain.com in the dns but
at the moment I can't see y.y.y.229 from the Internet. This is my problem.
--
Carl Filpo
Computer Network Consultant
=================================================
Carl Filpo BSc(Curtin)
Computer Network Consultant
C&M Computer Solutions
26 Russell St
MORLEY WA 6062
Email: [EMAIL PROTECTED]
Phone: +61 8 9375 1144
Fax: +61 8 9375 1141
Mobile: 0407 396 316
=================================================
------------------------------
From: [EMAIL PROTECTED] (Rod Smith)
Subject: Re: d-link card
Date: Mon, 05 Jul 1999 13:09:55 GMT
Reply-To: [EMAIL PROTECTED]
[Posted and mailed]
In article <[EMAIL PROTECTED]>,
"David C. DiNucci" <[EMAIL PROTECTED]> writes:
> Thanks for the suggestions. There are 3 PCI slots on the machine, and I
> now see that one of them is unusable, because a clock chip is in the
> way.
I used to see a lot of this on 486 motherboards, but not on Socket 7
motherboards. Some boards can still fit into such slots if they're shaped
to do it.
> This, in itself, makes me think that maybe my machine is a piece
> of junk, and that this may be the main problem. The BIOS has no setting
> referring to latency (or bus-master slots). I've tried the board in the
> two possible slots, and I've tried via-rhine drivers dated as new as
> 4/30/99 (v1.02), though I bought my board about a year ago.
>
> I guess it's time to get a new machine.
You could also try replacing only the motherboard. That'll be much
cheaper, if you're happy with your other major components (hard drives,
video card, and so on).
> I suppose there could be some
> conflict on either io addresses or IRQs (I still haven't figured out
> fully how to determine or avoid such conflicts)
Try "cat /proc/interrupts" to see what interrupts are in use on the
system. When you reboot with the D-Link DFE-530TX NIC in place, check to
see if it's trying to use one of the interrupts Linux used for something
else when the NIC wasn't in place. If it does, try mucking with your
motherboard's IRQ-assignment parameters from the CMOS settings. Sometimes
you can assign a specific board a specific interrupt. Other times you can
select from two or more priority lists of IRQ assignment.
Another thought: You might look for a BIOS upgrade for your motherboard.
It's conceivable that the BIOS has a bug that's causing the problems, and
a BIOS upgrade might fix it. This is a potentially dangerous thing to
muck with, though; if you make a mistake or if there's a problem, a BIOS
upgrade can result in a computer that won't boot until the BIOS chip is
physically replaced.
--
Rod Smith
[EMAIL PROTECTED]
http://www.channel1.com/users/rodsmith
NOTE: Remove the "uce" word from my address to mail me
Author of _Special Edition Using WordPerfect for Linux_, from Que;
see http://www.channel1.com/users/rodsmith/books.html
------------------------------
From: [EMAIL PROTECTED] (G Crowder)
Subject: Re: PPP connect: ping & traceroute but no data??
Date: Mon, 05 Jul 1999 14:02:45 GMT
>> However, when I telnet, ftp, or use a browser I get all the 'connected
>> to host' messages but no data. For ftp and telnet this means no
>> 'username' prompt while netscape leaves me at 'waiting for data'.
>Let me guess, you just upgraded to redhat 6.0, right? I just did that
>(upgrade from RH5.2) and am having the same symptoms. Note that in
>/var/log/messages I see:
>Jul 3 15:54:08 localhost modprobe: can't locate module ppp-compress-21
I found I needed to add
insmod ppp_deflate
where previously I only had insmod ppp, insmod slhc and insmod
bsd_comp
I didn't need any alias lines.
(This is also after an RH6 upgrade).
Graham C
------------------------------
From: "Michael Faurot" <[EMAIL PROTECTED]>
Subject: Re: DNS setup problem
Date: 30 Jun 1999 05:32:03 GMT
[EMAIL PROTECTED] wrote:
: Thank you for that Michael I just checked my Win95 settings and yes it
: shouldn't be 203.17.244.11 it should be 203.17.224.11
: I just wasted a whole weekend and countless phone calls to my ISP for
: one little typo.
: I guess it can happen to anyone :-)
Of course, now you're an expert at configuring your systems to access
your ISP's DNS servers. :)
: As an aside why was I getting responses from two other IP's when I
: tried pinging 203.17.244.11
I see what you mean . . .
=============================================================================
/misc/tmp> ping 203.17.244.11
PING 203.17.244.11 (203.17.244.11): 56 data bytes
64 bytes from 203.63.121.61: icmp_seq=0 ttl=235 time=4275.3 ms
64 bytes from 203.17.244.10: icmp_seq=0 ttl=234 time=4295.2 ms (DUP!)
64 bytes from 203.63.121.61: icmp_seq=1 ttl=235 time=4260.5 ms
64 bytes from 203.17.244.10: icmp_seq=1 ttl=234 time=4275.0 ms (DUP!)
64 bytes from 203.63.121.61: icmp_seq=2 ttl=235 time=4400.5 ms
64 bytes from 203.17.244.10: icmp_seq=2 ttl=234 time=4410.8 ms (DUP!)
--- 203.17.244.11 ping statistics ---
7 packets transmitted, 3 packets received, +3 duplicates, 57% packet
loss
round-trip min/avg/max = 4260.5/4319.5/4410.8 ms
=============================================================================
I've no real idea on what causes this or why. Frankly I've never seen
this before. However, the man page for ping seems to indicate that
if duplicate packets are present it's a sign of problems. Fortunately
they're not your problems.
--
==============================================================================
Michael | mfaurot | BASIC is the Computer Science equivalent of `Scientific
Faurot | atww.org | Creationism'.
------------------------------
From: [EMAIL PROTECTED] (Rod Smith)
Crossposted-To: comp.os.linux.hardware
Subject: Re: Using two LNE100TX NICs, adapter module fails?
Date: Mon, 05 Jul 1999 13:14:16 GMT
Reply-To: [EMAIL PROTECTED]
[Posted and mailed]
In article <D0Qf3.824$[EMAIL PROTECTED]>,
"mnip" <[EMAIL PROTECTED]> writes:
> I've used Linksys LNE100TX NICs for ages, and now that I'm starting to use
> Linux, I'm having trouble getting my machine with two of these cards to
> start its ethN modules. The cards don't have any IRQ conflicts, they work in
> other OS', and this is a BX6r2 motherboard. The Linux distro is RH6, no
> patches yet. One NIC is a static IP on my LAN, the other is for a pure DHCP
> cablemodem (but I still plug in the static IP leased every morning with NT4,
> at least until I get these cards working under linux).
>
> Is it still a problem to use two identical NICs in the same machine? I saw
> mention once that it was, but that was a while ago, with older kernels.
Try recompiling your kernel, and place the Samba driver IN THE KERNEL (not
as a module). That's reputed to solve some problems like the one you're
reporting.
--
Rod Smith
[EMAIL PROTECTED]
http://www.channel1.com/users/rodsmith
NOTE: Remove the "uce" word from my address to mail me
Author of _Special Edition Using WordPerfect for Linux_, from Que;
see http://www.channel1.com/users/rodsmith/books.html
------------------------------
Reply-To: "Carl Filpo" <[EMAIL PROTECTED]>
From: "Carl Filpo" <[EMAIL PROTECTED]>
Subject: Re: IPChains Problem
Date: Mon, 5 Jul 1999 21:24:47 +0800
I think the whole idea of masquerading is to get private ip's to disguise
connections on the internet as though coming from real ip's.
With only one NIC you can't disguise one for another - doesn't make sense.
You must have 2 NICs - even if one is a ppp device setup for a modem
connection
but it still counts as a NIC.
Bullit <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> hello !
> =====
>
> well im using linux as an router with masquerading and it works well.
> but now i want to know some things :
>
> 1. ) it is possible to masquerade on an pc with only one net-interface
> ( i.e. ethernet card) or must i have two net-interface have !?
>
> 2. ) what happens when i masquerade only on on interface ???
>
> 3.) how can i use masquerade with the new ipchains ?
>
>
> thanks for all answers
>
> bullit
>
------------------------------
** 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
******************************
