Linux-Networking Digest #339, Volume #12 Mon, 23 Aug 99 21:13:36 EDT
Contents:
Re: Routing blues..... ("Paul Vienneau")
Re: can't get any NIC's to work with RH6 (Roy Grimm)
Re: realtek problem (dmalcolm)
RH 6 network error (Nick Rout)
Re: sqlinit: DBROOT must be set !!! ("Percy Cheng")
pop/imap over forwarded/masqued connection (Jeff Lloyd)
Technical Writers and Reviewers (Ryan Finnesey)
How does linuxconf conn and disconn to ppp0? ("MikeSch")
when ppp0 connects, hard drive goes crazy (Mike)
Problem establishing PPP connection (Luis D Casillas)
Working GRE patched Kernel for RH6.0 ("Jeff")
Re: NFS security - how to turn some of it off? (David C.)
IP Masq slow with Cablemodem (Rahjer Dean)
Servers not accessible from network ("Trudel,Guy")
Servers not accessible from network ("Trudel,Guy")
Servers not accessible from network ("Trudel,Guy")
lengthy ARP discovery ([EMAIL PROTECTED])
3c905c - Tornado, Please Help ("Tom Holmes")
Re: Cracks for Linux? ([EMAIL PROTECTED])
Re: Is 3com's Office Connect ISDN modem overkill? ([EMAIL PROTECTED])
Re: Cracks for Linux? that's unfortunate ([EMAIL PROTECTED])
Firewall setup and network routing (Jim Smith)
----------------------------------------------------------------------------
From: "Paul Vienneau" <[EMAIL PROTECTED]>
Subject: Re: Routing blues.....
Date: Mon, 23 Aug 1999 18:57:25 -0400
Able to ping the router...
Pinging 192.168.1.1 with 32 bytes of data:
Reply from 192.168.1.1: bytes=32 time<10ms TTL=255
Reply from 192.168.1.1: bytes=32 time<10ms TTL=255
Reply from 192.168.1.1: bytes=32 time<10ms TTL=255
Reply from 192.168.1.1: bytes=32 time<10ms TTL=255
Any thoughts?
Thanks.
-paul
Allen Wong wrote in message <[EMAIL PROTECTED]>...
>Paul,
>
> I don't like the output of your "tracert". Are you able to ping
>your router from your NT workstations? I believe the tracert should at
>least have this:
>
>traceroute to 192.168.0.1, 30 hops max, 40 byte packets
> 1 <hostname of router> (192.168.1.1) 1.411 ms 0.602 ms 0.584 ms
>
>before it gets stuck.
>
>Allen
>--
>Linux: If you're not careful, you might actually learn something.
------------------------------
From: Roy Grimm <[EMAIL PROTECTED]>
Subject: Re: can't get any NIC's to work with RH6
Date: Mon, 23 Aug 1999 16:25:09 -0500
Bradford wrote:
>
> I've tried intel, smc, 3com I always get an error when RH boots. HELP!!
What error do you get? My Red Hat 6 worked with my 3Com just fine right
out of the box. Perhaps your network setup information is not correct.
If the problem stays no matter what card you have, it's probably not the
card.
Roy
--
I'm not a good artist but I am good at drawing a blank.
------------------------------
From: dmalcolm <[EMAIL PROTECTED]>
Subject: Re: realtek problem
Date: Mon, 23 Aug 1999 15:28:48 -0500
Gaudibri wrote:
> I just installed Red Hat 5.2 on my computer with an old ISA Ethernet realtek
> 8019 inside.
> The problem is that Linux always finds it busy and does not want to activate
> it. The only thing that is working is the loopback device.
>
> To check the base address and irq of the card, i plugged it in an another
> computer powered (:-)) by Windows 98 but even with these parameters, i get
> again and again the same words on the screen : eth0 is busy, busy, busy.
>
> If anyone could help me....
Wish I could help more, but it really sounds like either an IRQ conflict or
driver problem. Could also be an address conflict.
Dan
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Nick Rout)
Crossposted-To: redhat.networking.general
Subject: RH 6 network error
Date: Tue, 24 Aug 1999 10:49:34 +1200
I have done an ftp install on a 486 box that I'm going to use as a
gateway. it has no cdrom so I did an ftp install over my lan. So I know
the ethernet card works. I kept the install very small to fit on the hard
drive. Trouble is, now its up and running but I can't access the lan! I
ping a win95 machine and it just sits there. If I ctrl-c it tells me it
sent x packets but none came back. If I do an ifconfig it shows lo and
eth0 as up, and heres the strange thing - the ping packets seem to have
gone through lo not eth0. If I run ping again then the exact number of
packets sent are added to the count in lo when I do an ifconfig. The
routing table looks ok - three entries (sorry those machines are at home
and I am at work so I can't remember EXACTLY what it says). If I ping
somewhere outside my network eg 200.200.200.200 it reports back "network
unreachable" which seems right.
------------------------------
From: "Percy Cheng" <[EMAIL PROTECTED]>
Crossposted-To:
comp.infosystems.www.servers.unix,comp.os.linux,comp.os.linux.help,japan.www.server.apache
Subject: Re: sqlinit: DBROOT must be set !!!
Date: Mon, 23 Aug 1999 00:19:54 +0800
But if I want to start the httpd manually, or through rc.config to start the
httpd automatically, how do I modified the configuration?? Thanks a lot...
asrmj <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
Look for the initialization script /sbin/init.d/apache: There you will
see how DBROOT is set. You can use this script to start/stop the server;
for help about this, just execute: /sbin/init.d/apache.
asrmj
"Pletschette Andr�" wrote:
>
> When starting Apache, with httpd I get:
>
> sqlinit: DBROOT must be set !!!
>
> Does anybody know what this means and can anybody help me?
>
> Thank's
> ______________________
> Pletschette Andr�
> http://www.grosbous.lu
------------------------------
From: Jeff Lloyd <[EMAIL PROTECTED]>
Subject: pop/imap over forwarded/masqued connection
Date: 23 Aug 1999 18:42:16 -0400
Hi,
I have a stock redhat 6.0 linux box with the 2.2 kernel
forwarding/masquerading for both a linux and W95 box. Under both setups
everything seems to work fine ( the web, ftp, telnet, etc. ) but email
clients. The clients ( I have tested Netscape pop/imap on both linux and
W95 and eudora pop on W95 ) simply cannot retrieve email. Eudora gives up
waiting for "UIDL Information" and Netscape hangs with the message
"Sending Login information".
My setup is simple and right out of the HOWTO. I just have the basic
commands setup to get masquerading working, I haven't tried the more secure
setup.
Eg: This is from my script to start masquerading:
echo "1" > /proc/sys/net/ipv4/ip_forward
echo "1" > /proc/sys/net/ipv4/ip_dynaddr
/sbin/ipchains -M -S 7200 10 160
/sbin/ipchains -P forward DENY
/sbin/ipchains -A forward -s 192.168.0.0/24 -j MASQ
and I have the kernel modules loaded as well. Now everything but mail
retrieval works using this. I haven't tried to send mail yet.
Also, pop/imap mail retrieval works fine from the firewall machine itself.
Thanks for any help,
--
Jeff Lloyd
jlloyd at
algorithmics dot com
------------------------------
From: Ryan Finnesey <[EMAIL PROTECTED]>
Subject: Technical Writers and Reviewers
Date: Mon, 23 Aug 1999 19:26:10 +0100
Reply-To: [EMAIL PROTECTED]
Hi all
I am looking for some Technical Writers and Reviewers to help out with a
new web site that I am working on called Certification Central it will
be hosted at certcentral.com . I need to have various articles written
on subjects such has TCP/IP DNS WINS NDS SNMP NFS FTP etc. Virtually
any topic you will run into when you are going for your certification if
you would like to help out and have done some Technical Writing in the
past or would like to start please let me know.
Please send me an e-mail at [EMAIL PROTECTED]
Thank You
Ryan,
------------------------------
From: "MikeSch" <[EMAIL PROTECTED]>
Subject: How does linuxconf conn and disconn to ppp0?
Date: Mon, 23 Aug 1999 00:19:10 -0400
As clueless as I am, I have managed to connect ppp0 to my isp and with
ipchains everything on my win98 machine just works like a champ. I would
like to be able to connect and disconnect the linux box ppp0 without
trekking to the basement. So far, the only way I can connect and disconnect
is by hitting the respective buttons in linuxconf. Does anyone know what
linuxconf is doing for me in the background that I might duplicate over a
telnet session from my office? (Any ideas for a batch session that can be
run from the win98 client to conn and dis the linux box's ppp0?) This is
just the first step. Some day I will have a dedicated line and all my
win98/nt machines will connect through it. (and I won't have to disconnect)
Until then, I have to keep my ISP bill down!!!!!
Any help or pointers greatly appreciated.
I am still elated that the ipchains thing works so well!!!!
Mike Schrauder
please email
[EMAIL PROTECTED]
------------------------------
Subject: when ppp0 connects, hard drive goes crazy
From: [EMAIL PROTECTED] (Mike)
Date: Tue, 24 Aug 1999 00:17:56 GMT
What could cause this? After it happens, the machine is for all intents and
purposes locked up, because the hard drive never quits accessing.
------------------------------
From: Luis D Casillas <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.help
Subject: Problem establishing PPP connection
Date: 23 Aug 1999 17:01:48 -0700
Hi!
I'm having problems establishing a PPP copnnection with my new ISP. I
have setup my chat script to wait for the login prompt, send my
username, waint for the password prompt, and then send the password
and pass the line over to pppd.
However, after chat sends the username, the PPP host pauses for a few
seconds, and sends out a line like this:
*** Host Crurently Unavailable ***
after which it just resends the "login:" prompt.
The funny part is that Win95 connects allright. I guess you can
guess that my ISP is running NT. :-).
I looked at the PPP howto and the PPP FAQ, and no help there. I've
set up PPP before for like 4 different ISPs, and never had any
problems. The setup that is failing right now basically just changed
in the phone no, username and password in the chat script.
The ISPs tech support people told me, aprat from the "we don't support
Linux" and and "Linux is is full of bugs" (!) lines, that they don't
use PAP, or CHAP. So that one isn't it.
Well, in short, Win95 connects, while with Linux I can't ever get to
the PPP handshake. (Yeah, I tried looking at stuff dialing my ISP
with minicom, and I never got to see the "garbage" that PPP servrs
send out when they're trying to connect.)
Plese send followups both to the newsgroup and my email.
--Luis
------------------------------
Reply-To: "Jeff" <[EMAIL PROTECTED]>
From: "Jeff" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.questions,comp.os.linux.setup
Subject: Working GRE patched Kernel for RH6.0
Date: Mon, 23 Aug 1999 23:49:39 GMT
Hi folks;
I am trying to patch my kernel on a working firewall/masq'ed 500 Mb HD.
I have downed the source and the headers, patch, make and the like. I
started the compile and then ran out of space, (#$%^&*). I freed up some
space and started with mrproper.......I get an error and quite frankly, have
given up. Project time is almost up and I need a working kernel. Nothing
special but IDE and a few modules for two nics...(already have these) Does
anyone have a working kernel that is patched with no 'EXTRA' dependencies. I
will do a make modules (if this is possible with a kernel compiled on
another somewhat 'standard' build). A zipped kernel sent by email would be
greatly appreciated.
Flames >> /dev/null; comments questions welcome.
Jeff
------------------------------
From: [EMAIL PROTECTED] (David C.)
Subject: Re: NFS security - how to turn some of it off?
Date: 23 Aug 1999 19:25:24 -0400
"YouDontKnowWho" <[EMAIL PROTECTED]> writes:
> David C. wrote in message ...
>> Here's the scenario:
>>
>> My /etc/exports file consists entirely of:
>>
>> /home (rw)
>>
>> This should allow everybody to mount the volume, right? Wrong.
>>
>> Most hosts, when they try to mount this volume, get "permission
>> denied" or similar errors. Whenever they attempt to mount the
>> volume, lines like these are appended to /var/log/messages:
>>
>> <date> <host> mountd[###]: mount request from unknown host ##.##.##.##
>>
>> I have found that if I add a hosts's name to /etc/hosts, then it will
>> be able to mount the volume, but not otherwise. It doesn't matter
>> what OS the host is running.
>>
>> How do I turn off this kind of authentication? I tried adding
>> entries to /etc/hosts.allow, but it didn't do any good. I really
>> don't want to download my company's entire DNS table to be my
>> /etc/hosts.
>
> Check into:
>
> 1. =public
This has nothing to do with permissions. It defines the "public root"
directory that WebNFS uses.
Nevertheless, I tried it. It didn't help.
> 2. Adding a domain with a wildcard to the export (192.168.*)
I tried that. It didn't do anything. I tried 192.168.* and
192.168.0.0/255.255.0.0.
> See the man pages for exports.
I did. It didn't help. Perhaps you didn't read my message close
enough. I am exporting the file with no host-names. This is supposed
to grant access to everybody. And it does, as long as the host is named
in my /etc/hosts file. Adding a workstation to /etc/hosts instantly
grants it access.
I thought it might be a pam thing, but other pam-restricted facilities
(like rlogin) work fine. And I can't find an entry in /etc/pam.d that
might correspond to any nfs facilities.
-- david
------------------------------
From: [EMAIL PROTECTED] (Rahjer Dean)
Subject: IP Masq slow with Cablemodem
Date: Mon, 23 Aug 1999 23:56:12 GMT
Hi,
I've got an IP Masquerading setup here that works just fine,
but things are pretty slow and I'm getting a wierd message on the
Linux machine.
Running RH5.2 on a P100/32m, two ISA network cards, and all
the IP Masq stuff setup just fine. Have a cable modem plugged into
the Linux box, and two Win95 machines plugged into a hub then into the
Linux box.
I get long pauses (30-60 seconds) on the Win95 machines while
on the Net (downloading mail, newsgroup headers, etc), and the speed
is noticably slower when things are going. Also, on the Linux box,
when I'm at the command prompt (no X running), I get a repeated
message:
eth1: XMT status = 0xffff8841
Is that normal (doesn't sound right to me)? Is something wrong
with the way my network cards are setup? I've got an Intel
EtherExpress Pro 10+ and a 3Com 509b, both drivers compiled into the
kernel. eth1 is the Intel card going to the hub.
Any help would be greatly appreciated.
Roger
------------------------------
From: "Trudel,Guy" <[EMAIL PROTECTED]>
Subject: Servers not accessible from network
Date: Mon, 23 Aug 1999 18:19:10 -0400
Hi,
I have a RedHat 6.0 system with 2.2.11 kernel. The machine is set up as
a MASQ gateway with DSL link to the internet. Apache, FTP and Telnet
daemons work great. I can access them from the internet and my LAN with
no problem.
The problem is that when I try to run other services such as
phonepatch, fsgs daemon listing at certain port (1084 and 6112) I can
access them from the gateway machine only but not from my LAN and
Internet. This problem was no there when I was using RedHat 5.2. Does
anybody know what would cause this problem and how I should fix it? I
suspect that its related to some new security enhancment of Redhat 6.0.
Please help!
Thanks,
Matthew Ho
[EMAIL PROTECTED]
------------------------------
From: "Trudel,Guy" <[EMAIL PROTECTED]>
Subject: Servers not accessible from network
Date: Mon, 23 Aug 1999 18:18:45 -0400
Hi,
I have a RedHat 6.0 system with 2.2.11 kernel. The machine is set up as
a MASQ gateway with DSL link to the internet. Apache, FTP and Telnet
daemons work great. I can access them from the internet and my LAN with
no problem.
The problem is that when I try to run other services such as
phonepatch, fsgs daemon listing at certain port (1084 and 6112) I can
access them from the gateway machine only but not from my LAN and
Internet. This problem was no there when I was using RedHat 5.2. Does
anybody know what would cause this problem and how I should fix it? I
suspect that its related to some new security enhancment of Redhat 6.0.
Please help!
Thanks,
Matthew Ho
[EMAIL PROTECTED]
------------------------------
From: "Trudel,Guy" <[EMAIL PROTECTED]>
Subject: Servers not accessible from network
Date: Mon, 23 Aug 1999 18:18:12 -0400
Hi,
I have a RedHat 6.0 system with 2.2.11 kernel. The machine is set up as
a MASQ gateway with DSL link to the internet. Apache, FTP and Telnet
daemons work great. I can access them from the internet and my LAN with
no problem.
The problem is that when I try to run other services such as
phonepatch, fsgs daemon listing at certain port (1084 and 6112) I can
access them from the gateway machine only but not from my LAN and
Internet. This problem was no there when I was using RedHat 5.2. Does
anybody know what would cause this problem and how I should fix it? I
suspect that its related to some new security enhancment of Redhat 6.0.
Please help!
Thanks,
Matthew Ho
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Subject: lengthy ARP discovery
Date: Tue, 24 Aug 1999 00:05:06 GMT
Why should ARP tables take up to 2 minutes to reflect a new entry, in
response to a ping or other attempt to concact a target by its IP
address?
My understanding is that the activities are supposed to be high
priority if not immediate. First, that the underlying routing system
on the pinging machine issues an ARP request without delay as soon as
it tries but fails to resolve the IP address into a corresponding
ethernet address. Second, that machines receiving the request (ie, all
that are connected to the LAN) are to read it, and answer it if they
know the answer-- also without delay.
I'm issuing "pings" and the target IP address shows up in the arp
table immediately but with "(incomplete)" in the ethernet address
field. A whole minute later, perhaps, the ethernet address finally
shows up in the table. This is happening on 2 linux machines in 2
different LANs and I don't remember having seen such a problem before
on either.
Any wisdom?
------------------------------
From: "Tom Holmes" <[EMAIL PROTECTED]>
Subject: 3c905c - Tornado, Please Help
Date: Mon, 23 Aug 1999 20:11:17 -0400
Hi, I am one of the unfortunates who didn't get their network identified
right away.
It is a 10/100 3Com 3c905c-TX card. The BIOS recognizes the card with PnP
and gives it IRQ 11. I confirmed that the card was working fine, and I
changed the card to only be 10MBps and half-duplex since my hub is only
10MBps also. The card works fine with Windows 95/98/NT.
I am using a Dual Pentium Tyan Tomcat III with 64MB of RAM, and two P166
Classic. This board has 4 PCI slots (1-4) and 4 ISA slots. I have the card
in slot 2 (and I can't put any card in PCI Slot 1 because of the case I
bought).
I got the new 3c59x.c drivers and re-compiled, no luck.
I re-compiled the 2.2.11 kernel and loaded the drivers for the Vortex 3com
cards, no luck.
I can't do an insmod.
Is the 3c59x.o the right module? The /proc/pci tells me that the card is
recognized at port 11, but that it is an unknown device.
Should I turn off plug and play and re-install 2.0.36? Is the Plug and Play
causing the problem.
Any help would be much appreciated. The only reason I got this card is
because I thought the card would be identified right away. I got tired of
having to re-compile the rtl8139.c file and I am havin MORE problems with
the 3Com card.
Please help. Thanks.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: Cracks for Linux?
Date: 23 Aug 1999 17:29:35 -0700
[EMAIL PROTECTED] (Jason Earl) writes:
[snip]
> I got it to work by typing in (as root):
>
> cd /
> rm -rf *
>
> After that I didn't have any problems with OSS/Linux asking me to register it.
> I think that it removes the nagware config file or something.
it didn't seem to work for me...
rm: /bin/rm: Text file busy
are they using some eleet copy protection or something?
andru
--
==========================================================================
| Andru Luvisi | http://libweb.sonoma.edu/ |
| Programmer/Analyst | Library Resources Online |
| Ruben Salazar Library |-----------------------------------------|
| Sonoma State University | http://www.belleprovence.com/ |
| [EMAIL PROTECTED] | Textile imports from Provence, France |
==========================================================================
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.misc
Subject: Re: Is 3com's Office Connect ISDN modem overkill?
Date: Tue, 24 Aug 1999 00:08:26 GMT
Steve,
I am doing the same thing BUT I have not received my ISDN yet (~3 wks to
go). One thing I noticed in reading some vendor www docs is that you
*might* have problems with NAT and running your own Quake server and/or
allowing multiple PC's thru NAT to play in your hosted game -or- on the
same server on the net. The reason is NAT (remember that your ISDN
router will get the REAL IP address from your ISP and not your connected
PC(s) which typically get a private address that the ISDN router
handles). The reason for the problem is a Quake server wouldn't know
which PC on your LAN (asuming an ISDN router and 2+ PC's) is the IP
address your router received when you dialed in. The doc below states: "
Certain Quake servers do not allow multiple users to login using the
same unique IP, so only one Quake user will be allowed in this case.
Moreover, when a Quake server is configured behind SUA (Single User
Account) Prestige will not be able to provide information of that server
on the internet". So, this product (and maybe SUA) is worthless, right?
My reference article is:
http://www.zyxel.com/doc/p202/ap_note/isdn/sua_ap.htm for their Prestige
202 ISDN Router (one of the ones I've looked at).
IF ANYONE KNOWS A WAY AROUND THIS *PLEASE* CORRECT ME (and the vendor)!
In article <7pmvqm$hpe$[EMAIL PROTECTED]>,
"Steve Ferrante" <[EMAIL PROTECTED]> wrote:
> I want to switch from a dialup ISP to ISDN. I heard about bottlenecks
--<snip>--
> If this more expensive than in other areas? Are there other choices
for
> ISDN service besides BA in North Jersey?
In western Virginia, Sprint is charging about $ 45/mth for unlimited
local ISDN 2B1D circuits plus your ISP fee (~$40/mth for 150 hrs dual
channel avg.) The local telco can charge what they are tarriffed for
which means if they are tarriffed for business ISDN but not residential
ISDN you'll pay more! Call them and ask what they are tarriffed for.
> Thanks,
> Steve Ferrante
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: Cracks for Linux? that's unfortunate
Date: 23 Aug 1999 17:34:03 -0700
"Abhishek Srivastava" <[EMAIL PROTECTED]> writes:
> Asking for Crack software for linux is Stupid !!
> when u get *almost* all things for free. To pay
> USD 20 for such a wonderfull system should hurt at all.
>
> The linux community should support the organizations that make
> Linux software.
$20 is fine. what hurts is not getting GPL compatible source code. I
would pay *more* for the driver that comes with the kernel than the
one they sell, because it comes with GPL compatible source.
andru
--
==========================================================================
| Andru Luvisi | http://libweb.sonoma.edu/ |
| Programmer/Analyst | Library Resources Online |
| Ruben Salazar Library |-----------------------------------------|
| Sonoma State University | http://www.belleprovence.com/ |
| [EMAIL PROTECTED] | Textile imports from Provence, France |
==========================================================================
------------------------------
Subject: Firewall setup and network routing
From: [EMAIL PROTECTED] (Jim Smith)
Date: Tue, 24 Aug 1999 00:31:29 GMT
Hi all,
I'm hoping somebody can give a hand with this one. I have two Linux boxes and a
Win98 box setup on my Local Area Network. One Linux box (LINUX-PPP) is set up
to connect to my ISP via PPP and the other one is just for fooling around
(LINUX-DUMMY). My Win98 box (WIN98) is just for applications like Word and
stuff. I would to access to Internet SECURELY from every box on my LAN.
First off all, PPP is set up and working fine. I've downloaded the
firewall/proxy SOCKS and installed it on LINUX-PPP. I've modified the
/etc/inetd.conf file and added the line "socks stream tcp nowait nobody
/usr/local/etc/sockd sockd". Secondly, I've created the file
/etc/sockd.conf and adding the following two lines: "permit 192.120.1.0
255.255.255.0" and "deny 0.0.0.0 0.0.0.0".
First question, is this a secure setup and how does one test the setup itself?
Second question, how do I access the Internet from WIN98 and from LINUX-DUMMY
through LINUX-PPP?
Thanks in advance for your help.
Jim.
------------------------------
** 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
******************************
