Linux-Networking Digest #879, Volume #9 Thu, 14 Jan 99 18:13:50 EST
Contents:
Re: Copying packets over interfaces (Pete Jones)
config pcmcia/pc card and 3c589_cs (TJ)
copying some directory tree to local HD with ftp (shin, dong shik)
Re: Hackers used my linuxserver be hacked gateway How to fixing? (David Barnes)
Help making an IPX router ("John Jarvis")
tcp/ip from Win98 to Linux: Strange behaviour ("Alex Bruschke")
Re: DNS/DHCP behavior (Yan Seiner)
Re: Linux as a simple file sharing device? ("Eugene")
Mail and dial up ([EMAIL PROTECTED])
Re: PCMCIA/D-link 660/RH5.1 ethernet problems (Ulf Leichsenring)
Re: Allowing remote access. ("Eugene")
Beginner with Lots of Toys!!! ("Lord Rayden")
Re: I know nothing (Richard Steiner)
Re: SLOW PPP Connection (Marc)
Re: DNS/DHCP behavior (Norman Patten)
imap is there a safe version? ("Jos Okhuijsen")
Re: How to assign a IP address range to an interface ? (Matt Kressel)
Re: New PPP Program (Matt Kressel)
Re: Linux-Linux networking problem (Matt Kressel)
Re: Port Forwarding (Wisquatuk)
Re: PPP serial connection to nowhere (jfk)
----------------------------------------------------------------------------
From: Pete Jones <[EMAIL PROTECTED]>
Subject: Re: Copying packets over interfaces
Date: 14 Jan 1999 18:30:28 +0200
Ronald S. Kundla Jr. <[EMAIL PROTECTED]> wrote:
: So enabling the bridge functionality and recompiling the kernel didn't
: work?
Hmm, sort of slipped me by, that one. Now it looks like
it's working.
(http://metalab.unc.edu/LDP/HOWTO/mini/Bridge+Firewall.html)
Thanks,
Pete...
------------------------------
From: [EMAIL PROTECTED] (TJ)
Subject: config pcmcia/pc card and 3c589_cs
Date: Thu, 14 Jan 1999 03:53:18 GMT
I want to add the module for my lan/modem card. But first, I need to
set up the pcmcia. I can see that the Texas Instruments card slots
have been initialized, but not the pcmcia module (?). Can someone
tell me explicitly where the instructions are to do this or better yet
give me a generic recipie?
------------------------------
From: [EMAIL PROTECTED] (shin, dong shik)
Subject: copying some directory tree to local HD with ftp
Date: Thu, 14 Jan 1999 05:33:46 GMT
Hi,
I want to copy all about 500Mbyte contents from some ftp site to my
local hard disk. That will be something like private mirroring.
I am going to build some script, which connect to ftp server and get
each file. The script works with cron demon and automatically start
and download files written in hoping-files list.
Anybody knows how?
Thanks in advance...
shin, dong shik
---
email : [EMAIL PROTECTED]
------------------------------
From: David Barnes <[EMAIL PROTECTED]>
Subject: Re: Hackers used my linuxserver be hacked gateway How to fixing?
Date: Thu, 14 Jan 1999 13:23:39 -0500
Thanks for your reply. I appreciate it. I had a firewall on the system itself
set up with the following parameters:
ipfwadm -F -p deny
ipfwadm -F -a m -S 192.168.1.0/24 -D 0.0.0.0/0
But I did not have any of the Redhat 5.1 patches loaded.
First, is that firewall configuration right, and second, even with the firewall
configuration right, could the other security issues, such as the NFS one you
mentioned still be exploited in spite of the firewall?
Thank you.
David Barnes
[EMAIL PROTECTED]
Harald Michael Fuchs wrote:
> .....the safest way still is to set up a firewall.......
>
> cu,
>
> Harry
>
> /\_/\ Dipl.-Ing. Harald M. Fuchs <[EMAIL PROTECTED]>
> ( o.o ) http://oeh.tu-graz.ac.at/~harry
> > ^ < private phone: +43 (316) 84 72 73
> Sally business phone: +43 (316) 873 5843, fax: 873 5805
------------------------------
Subject: Help making an IPX router
From: "John Jarvis" <[EMAIL PROTECTED]>
Date: Thu, 14 Jan 1999 13:00:42 -0500
Hello -
I have the task of connecting a Novell server (using IPX) to our state's NDS
tree, from behind our masq box. We are currently using a Linux machine to
masquerade our Ethernet IP LAN over a token ring 'backbone' (which has run
flawlessly - with only one accidental fat-finger reboot - for over a year).
Unfortunately, I can't get IPX to work. Maybe some IPX guru out there can
tell me what I'm doing wrong. I have set up IPX on the machine using the
following commands:
ipx_configure --auto_interface=on
ipx_interface add -p eth0 802.2 A2
ipx_route add a1 a2 006008a39cb5 (the number obtained from node_address
below)
Here is a printout of my cat /proc/net/ipx_itnerface:
# cat ipx_interface
Network Node_Address Primary Device Frame_Type
000000A1 000629A8F1B8 No tr0 802.2TR
000000A2 006008A39CB5 Yes eth0 802.2
Here is a partial printout of /proc/net/ipx_route:
# cat ipx_route |more
Network Router_Net Router_Node
342D51BC 000000A1 000728321105
00000056 000000A1 0009D5260114
00009000 000000A1 000728321105
(.... pages and pages of more routes, and mine are the last 2)
36486886 000000A1 000083655D5F
000000A2 Directly Connected
000000A1 000000A2 006008A39CB5
And finally, here is my /proc/net/ipx:
# cat ipx
Local_Address Remote_Address Tx_Queue Rx_Queue State Uid
000000A1:0453 Not_Connected 00000000 00000000 07 000
000000A1:0452 Not_Connected 00000000 00000000 07 000
000000A2:0453 Not_Connected 00000000 00000000 07 000
000000A2:0452 Not_Connected 00000000 00000000 07 000
Also, ifconfig does not show IPX under my tr0 adapter. It is listed under my
eth0 adapter as:
IPX/Ethernet 802.2 addr:000000A2:006008A39CB5
... but no such entry shows up under the tr0 adapter (don't know why it
shows up under /proc/net/ipx_interface though).
IPX is installed in the kernel, and our backbone is using both IP and IPX.
Here are my questions, since the above is not working:
1 - what am I doing wrong?
2 - how will I know if I have it?
3 - is there anything I need to know from the 'other side' of the router (on
the backbone)?
Thanks for any help,
John Jarvis
IS Specialist (although I don't feel like it right now:)
WV Retirement Board
------------------------------
From: "Alex Bruschke" <[EMAIL PROTECTED]>
Subject: tcp/ip from Win98 to Linux: Strange behaviour
Date: Thu, 14 Jan 1999 06:59:29 +0100
Hi,
I have a tiny (2 pc) tcp/ip network. One of them is a Win98 pc (adress
192.168.0.1) and the other a Linux pc (adress 192.168.0.2). When I use ping
from my Win98 pc to the Linuc pc there is no problem.
But when I use ftp or telnet (ftp 192.168.0.2) Win98 first tries to dial in
on my internet provider and when I cancel this only then it works fine and I
can connect to the Linux pc.
Two questions:
How can I prevent Win98 from first dialling to the internet when I try to
use the local tcp/ip connection ?
Why doesn't this happen when I use ping ?
------------------------------
From: Yan Seiner <[EMAIL PROTECTED]>
Subject: Re: DNS/DHCP behavior
Date: Thu, 14 Jan 1999 14:17:29 -0500
Reply-To: [EMAIL PROTECTED]
Thanks. I won't mess with it then. That jsut saved me a bunch of work
:-)
Yan
Brian McCauley wrote:
>
> Yan Seiner <[EMAIL PROTECTED]> writes:
>
> > I have both DNS (named) and DHCP (dhcpd) running. On the linux box, DNS
> > fails to resolve any requests for machines given IPs by DHCP.
> >
> > DNS requests over the network are resovled.
> >
> > SO, on the linux box, ping othello produces unknown host, but on a NT WS
> > on the network ping othello produces the correct result.
>
> AFAIK this is because it falls back to WINS not DNS.
>
> Like I said last week in answer to this question, the DHCP/dynamic-DNS
> stuff is still in the IETF-draft stage.
>
> Please read before you post.
>
> --
> \\ ( ) No male bovine | Email: [EMAIL PROTECTED]
> . _\\__[oo faeces from | Phones: +44 121 471 3789 (home)
> .__/ \\ /\@ /~) /~[ /\/[ | +44 121 627 2173 (voice) 2175 (fax)
> . l___\\ /~~) /~~[ / [ | PGP-fp: D7 03 2A 4B D8 3A 05 37...
> # ll l\\ ~~~~ ~ ~ ~ ~ | http://www.wcl.bham.ac.uk/~bam/
> ###LL LL\\ (Brian McCauley) |
------------------------------
From: "Eugene" <[EMAIL PROTECTED]>
Subject: Re: Linux as a simple file sharing device?
Date: Wed, 13 Jan 1999 23:25:25 -0500
samba is the tool for you. Not sure if the howto is available, but check out
www.linux.org/help anyway.
search for samba on freshmeat.net (it should come with your distribution
though)
Brian Smith wrote in message <77jmle$gcc$[EMAIL PROTECTED]>...
>I hope someone can give me some info on using a Linux computer to simply
>share files with Win32 machines.
>
>My company sells software to dentists; the typical user is 3-10 computers.
>We use the Btrieve database engine for all our files. We used to be able to
>put all the shared data files on a Win95 machine and use peer-to-peer
>networking with all the other Win95 machines, and things worked great. Then
>we switched to a new version of Btrieve, and all H*** broke loose; random
>file and record locking problems, lost connections to the server, etc.
>However, our few clients who still run Novell Netware as their server (with
>Win95 computers running our software) are experiencing none of the above,
>and they seem to run our software a lot faster. The only problem is,
Netware
>is quite expensive. So...
>
>We're wondering if a Linux box, set up so that Win95 and NT computers could
>share its files, would work as well for this file sharing task as Netware?
>If so, we want someone to build us a "black box" that would have Linux
>pre-installed, and we'd simply copy our install program onto it, and ship
it
>to clients to use as their "PracticeWorks Server." (PracticeWorks is the
>name of our program.) They'd attach it to their peer-to-peer network (or
>vice versa, I guess), then run our install program from it, and away they'd
>go. We don't even really want the box to have a monitor or keyboard; in
>concept, it would be exactly like the Snap! Server from Meridian
>Technologies. (We tried one; worked great, except it was too slow.)
>
>While we would make no secret of the fact that it's a Linux box, we
wouldn't
>make a big deal out of it, because we'd not want them to use it for
anything
>else. (One of the problems we have is people messing up a perfectly good
>installation of our software -- dental offices typically don't have anyone
>qualified to work on Windows, let alone Linux.)
>
>Can anyone help me? I need some input on the speed issue especially; does
>anyone know of any testing that's been done on Linux vs. Netware vs. NT for
>basic file services? Also, how to find a company that could build and
>service these boxes all over the country. Of course, since they would be so
>single-purpose, "service" might just consist of sending a loaner to replace
>the broken one, then swap them again.
>
>Sorry for being long-winded. I'll check here for replies, but I would
prefer
>an e-mail to [EMAIL PROTECTED], if that's not too inconvenient.
>
>Thanks,
>
>Brian Smith
>Director of Software Development
>PracticeWorks
>
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Mail and dial up
Date: Thu, 14 Jan 1999 21:19:35 GMT
Could anyone help me configuring fetch and send mail to get mailvia a
dial-up ppp connection?
Thanks
Tom
------------------------------
From: [EMAIL PROTECTED] (Ulf Leichsenring)
Subject: Re: PCMCIA/D-link 660/RH5.1 ethernet problems
Date: Thu, 14 Jan 1999 21:21:32 GMT
On 07 Jan 1999 10:36:32 -0800, [EMAIL PROTECTED] (Bernard J.Kozioziemski)
wrote:
>
>Greets,
>I have an Acer Extensa 366D laptop that I'd like to hook up to a cable
>modem. I currently have a D-link PCMCIA DE-660 Ethernet card, but I cannot
>seem to get things to work. I tried with both the RedHat 5.1 kernel and
>pcmcia-3.0.0 package, as well as kernel 2.0.36, without success. Upon
>booting or inserting the card, it is detected and identified as the D-Link
>DE-660 Ethernet Card. Modules pcmcia/pcnet_cs.o and net/8390.o are
>inserted. /var/log/messages shows: eth0: NE2000 Compatible: port 0x300, irq
>5, hw_addr 00:80:c8:8B:DF:96
>
>ifconfig lists eth0 with the correct addresses. The problem is that no
>traffic seems to come in or out... a ping to an ip address doesn't find an
>machine, and trying to ping my computer from the net fails as well. I've
>set the addresses and such in netcfg, looked at the docs for pcmcia
>package, fiddled, etc, all without any luck. If someone could provide some
>pointers, symptoms to look for, or anything, I'd apprecite it. Thanks,
>
>Bernie Kozioziemski
>[EMAIL PROTECTED]
>
Hi Bernie
Sorry, I couldn't help you, but I have the same weired things
happening using a 3Com 3C574TX (10/100 PCMCIA Ethernet) on a Toshiba
Tecra 8000 Notebook with Kernel 2.0.35/2.0.36 and pcmcia 3.0.6 (SuSE
distribution). The card is identified by cardmgr, ifconfig shows
everything allright but no traffic is sent or received.
I even tried another Tecra and another 3c574 to make sure there's no
hardware damage.
It seems to be a problem in the pcmcia subsystem. When I use a 3c589
instead, everything's ok.
I hope, somebody can give us some tricks/patches to make things work.
Bye
Ulf Leichsenring
Active Security
Lufthansa Systems AS GmbH
Schützenwall 1
D-22844 Norderstedt
Tel.: +49-40-5070-7859
Fax: +49-40-5070-7880
email: [EMAIL PROTECTED]
Internet: http://www.lhsystemsas.de
------------------------------
From: "Eugene" <[EMAIL PROTECTED]>
Subject: Re: Allowing remote access.
Date: Wed, 13 Jan 1999 23:32:58 -0500
it is (or should be) configured already. Try telnetting into it.
(BTW, by default root login via telnet is disabled for security reasons).
none wrote in message ...
>I'm currently running RH5.2. I was wondering how to configure it to accept
>telnet access. I connect to the internet throught a WAN with a dynamic IP.
>
>I appologise to the group if this is a rediculous question. If not, please
>point me to the nearest FAQ/How-to. I don't even know what to look for.
>
>Thanks,
>Aaron Mitchell
>
>[EMAIL PROTECTED]
>
>
------------------------------
From: "Lord Rayden" <[EMAIL PROTECTED]>
Subject: Beginner with Lots of Toys!!!
Date: Sun, 10 Jan 1999 19:06:35 -0600
hello everyone, I am a beginner in the Linux world but I already love it. I
have several great books for reference and even though the learning is
slow, I am starting to get it!
I have access to several high powered machines including one laptop (all
have windows and linux partitioned in). I have come across 3-4 laptops that
are 386/486 machines with no CD-ROM drives. Unfortunately, they have 150 MB
hard drives but only about 4 MB memory, one of them (a desktop) has a
network card. In my heart, I feel like I can use Linux to enhance my setup.
CAN YOU GIVE ME SOME IDEAS OF WHAT I CAN DO WITH THESE MACHINES? (Vague
question, but I am sure you get my drift).
Thank you in advance.
-CC-
------------------------------
From: [EMAIL PROTECTED] (Richard Steiner)
Subject: Re: I know nothing
Reply-To: [EMAIL PROTECTED]
Date: Wed, 13 Jan 1999 22:45:54 -0600
Here in comp.os.linux.networking, [EMAIL PROTECTED] (Greg)
spake unto us, saying:
>I am about to configure a server (486 dx4-100) to be my router.
Ah. I've done that. :-) (Well, not as *your* router <grin>).
A couple of sites that might be of interest to you are the following:
http://ipmasq.cjb.net/
http://www.indyramp.com/masq/
http://www.eunuchs.org/linux/ip_masq/index.html
--
-Rich Steiner >>>---> [EMAIL PROTECTED] >>>---> Bloomington, MN
OS/2 + Linux (Slackware+RedHat+SuSE) + FreeBSD + Solaris +
WinNT4 + Win95 + PC/GEOS + Executor = PC Hobbyist Heaven!
Johnny take a walk with your sister the moon...
------------------------------
From: Marc <[EMAIL PROTECTED]>
Subject: Re: SLOW PPP Connection
Date: Thu, 14 Jan 1999 08:42:19 +0100
Reply-To: [EMAIL PROTECTED]
Did you try to use setserial [dev] spd_vhi of spd_hi
Frank Kuehnel wrote:
> Hi Folks,
>
> I use RedHat 5.2 with the 2.0.36-3 kernel. Running PPP 2.3.5 works fine
> except the transfer rates. My external USR Sportster 56k connects at
> some
> 50k to the provider, the serial connection runs fine at 115k there are
> no
> overruns for the incomming frames.
>
> However, if I compare the resulting transfer rates for sending packages
> and
> receiving packages it it looks like 1.8k/sec versus 180bytes/sec with
> many
> incomming frames tossed because of "bad fcs". No matter how explicitly I
>
> tell the modem to be initialized or what explicit parameter I provide to
> pppd,
> like "asyncmap 0", the transfer rates remain the same. The are both way
> off
> from the results using Windows, 5k/sec.
>
> Jan 13 18:54:53 action kernel: PPP: ppp line discipline successfully
> unregistered
> Jan 13 18:56:33 action ifup-ppp: pppd started for ppp0 on /dev/modem at
> 115200
> Jan 13 18:56:33 action kernel: CSLIP: code copyright 1989 Regents of the
> University of California
> Jan 13 18:56:33 action kernel: PPP: version 2.2.0 (dynamic channel
> allocation)
> Jan 13 18:56:33 action kernel: PPP Dynamic channel allocation code
> copyright 1995 Caldera, Inc.
> Jan 13 18:56:33 action kernel: PPP line discipline registered.
> Jan 13 18:56:33 action kernel: registered device ppp0
> Jan 13 18:56:33 action pppd[1145]: pppd 2.3.5 started by root, uid 0
> ...
> Jan 13 18:57:02 action pppd[1145]: Serial connection established.
> Jan 13 18:57:03 action pppd[1145]: Using interface ppp0
> Jan 13 18:57:03 action pppd[1145]: Connect: ppp0 <--> /dev/modem
> Jan 13 18:57:03 action pppd[1145]: sent [LCP ConfReq id=0x1 <asyncmap
> 0x0> <magic 0xffff8001> <pcomp> <accomp>]
>
> .. attempt to read data
>
> Jan 13 19:09:04 action kernel: ppp_tty_read: called buf=08064610 nr=1504
>
> Jan 13 19:09:04 action kernel: ppp_tty_read: no data (EAGAIN)
> Jan 13 19:09:04 action pppd[1221]: sent [LCP EchoReq id=0x13
> magic=0x5feb]
> Jan 13 19:09:04 action kernel: ppp: successfully queued 10 bytes, flags
> = f01004f
> Jan 13 19:09:04 action kernel: ppp_tty_read: called buf=08064610 nr=1504
>
> Jan 13 19:09:04 action kernel: ppp_tty_read: len = 10
> Jan 13 19:09:04 action kernel: ppp_tty_read: passing 12 bytes up
> Jan 13 19:09:04 action pppd[1221]: rcvd [LCP EchoRep id=0x13
> magic=0x9863b5e7]
> Jan 13 19:09:10 action kernel: ppp_dev_xmit [ppp0]: skb 01776ee0
> Jan 13 19:09:10 action kernel: ppp_dev_xmit_lower: fcs is 89d3
> Jan 13 19:09:10 action kernel: ppp_dev_xmit: writing 17 chars
> Jan 13 19:09:10 action kernel: ppp_dev_xmit [ppp0]: skb 01776ee0
> Jan 13 19:09:10 action kernel: ppp_dev_xmit_lower: fcs is 63cf
> Jan 13 19:09:10 action kernel: ppp_dev_xmit: writing 33 chars
> Jan 13 19:09:10 action kernel: ppp_dev_xmit [ppp0]: skb 01776ee0
> Jan 13 19:09:10 action kernel: ppp_dev_xmit_lower: fcs is 9bf9
> Jan 13 19:09:10 action kernel: ppp_dev_xmit: writing 30 chars
> Jan 13 19:09:10 action kernel: ppp_dev_xmit [ppp0]: skb 02e39e9c
> Jan 13 19:09:10 action kernel: ppp_dev_xmit_lower: fcs is a01b
> Jan 13 19:09:10 action kernel: ppp_dev_xmit: writing 48 chars
> Jan 13 19:09:10 action kernel: ppp_dev_xmit [ppp0]: skb 02e39e8c
> Jan 13 19:09:10 action kernel: ppp_dev_xmit_lower: fcs is 96ed
> Jan 13 19:09:10 action kernel: ppp_dev_xmit: writing 11 chars
> Jan 13 19:09:11 action kernel: ppp: frame with bad fcs, excess = b2de
> Jan 13 19:09:14 action kernel: ppp: frame with bad fcs, excess = 3666
> Jan 13 19:09:20 action kernel: ppp: frame with bad fcs, excess = 6ccd
> Jan 13 19:09:32 action kernel: ppp: frame with bad fcs, excess = b319
> Jan 13 19:09:34 action kernel: ppp_dev_xmit_lower: fcs is f4f
> Jan 13 19:09:34 action kernel: ppp_dev_xmit: writing 14 chars
> Jan 13 19:09:34 action kernel: ppp_tty_ioctl: read demand dial info
> Jan 13 19:09:34 action kernel: ppp_tty_read: called buf=08064610 nr=1504
>
> Jan 13 19:09:34 action kernel: ppp_tty_read: no data (EAGAIN)
> Jan 13 19:09:34 action pppd[1221]: sent [LCP EchoReq id=0x14
> magic=0x5feb]
>
> ... finally gave up
>
> Since I have a direct comparison with results from Windows using the
> same
> hardware, I really doubt that the line is jammed every time I use Linux.
>
> At least I`ve read that a couple of people had or have a similar
> problem.
>
> Does anybody know what is going on?
>
> Thanks in advance
>
> Frank Kuehnel
------------------------------
From: Norman Patten <[EMAIL PROTECTED]>
Subject: Re: DNS/DHCP behavior
Date: Thu, 14 Jan 1999 14:08:56 -0800
This is a multi-part message in MIME format.
==============C15A4557D91788679C960BDF
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
I would like to help, but you need to post some information on your
/etc/resolv.conf and named.conf. your resolv.conf should look something
like this.
domain domain.com
nameserver 127.0.0.1 # check yourself first
nameserver x.x.x.x #another nameserver
Yan Seiner wrote:
>
> I have both DNS (named) and DHCP (dhcpd) running. On the linux box, DNS
> fails to resolve any requests for machines given IPs by DHCP.
>
> DNS requests over the network are resovled.
>
> SO, on the linux box, ping othello produces unknown host, but on a NT WS
> on the network ping othello produces the correct result.
>
> named is bound to eth0, and apparently it is ignoring any requests
> coming in locally.
>
> How do I get DNS to resolve the names on the linux box?
>
> Yan
==============C15A4557D91788679C960BDF
Content-Type: text/x-vcard; charset=us-ascii;
name="nepattenNOXSPAM.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Norman Patten
Content-Disposition: attachment;
filename="nepattenNOXSPAM.vcf"
begin:vcard
n:Patten;Norman
x-mozilla-html:FALSE
org:sequent computers
version:2.1
email;internet:[EMAIL PROTECTED]
note:Remove NOXSPAM to reply
x-mozilla-cpt:;0
fn:Norman Patten
end:vcard
==============C15A4557D91788679C960BDF==
------------------------------
From: "Jos Okhuijsen" <[EMAIL PROTECTED]>
Subject: imap is there a safe version?
Date: Thu, 14 Jan 1999 22:24:37 GMT
Hi there,
Is there a safe imap server somewhere, or is the protocol inherently
dangerous?
Regards,
Jos Okhuijsen
------------------------------
From: Matt Kressel <[EMAIL PROTECTED]>
Subject: Re: How to assign a IP address range to an interface ?
Date: Thu, 14 Jan 1999 20:57:38 GMT
[EMAIL PROTECTED] wrote:
>
> Hi,
>
> I need to recompile my 2.1.117 kernel and assign a range of ip address to an
> interface. How can I do that ? What's the kernel hack ?
>
> Example: eth0: 10.0.0.1 to 10.0.0.254
>
> Thank you to answer me also by email
>
"Network Aliasing" and see also in linux source:
Documentation/networking/alias.txt
-Matt
--
Matthew O. Kressel | INTERNET: [EMAIL PROTECTED]
+--------- Northrop Grumman Corporation, Bethpage, NY ---------+
+--------- TEL: (516) 346-9101 FAX: (516) 346-9740 ------------+
------------------------------
Crossposted-To:
alt.2600,comp.os.linux.setup,comp.os.linux.development.apps,comp.os.linux,alt.os.linux
From: Matt Kressel <[EMAIL PROTECTED]>
Subject: Re: New PPP Program
Date: Thu, 14 Jan 1999 21:01:26 GMT
Stuart Harris wrote:
>
> I'm about to start developing a ppp program (shell/perl)
> which will at first only work with REDNET,
> The ppp program (pppconnect) will hopefully grow very big!
> the general idea is that its more a terminal based control
> panel rather than 'enter this then this and ooh what about this'
Before you reinvent the wheel, make sure to check out what's out there!
There are a ton of these types of things there already. Perhaps you can
improve on them?
-Matt
--
Matthew O. Kressel | INTERNET: [EMAIL PROTECTED]
+--------- Northrop Grumman Corporation, Bethpage, NY ---------+
+--------- TEL: (516) 346-9101 FAX: (516) 346-9740 ------------+
------------------------------
From: Matt Kressel <[EMAIL PROTECTED]>
Subject: Re: Linux-Linux networking problem
Date: Thu, 14 Jan 1999 20:45:52 GMT
Jesse Hughes wrote:
>
> Hey ho.
>
> This is my third posting of this problem. The suggestions I've
> received gave me something to look at, but I couldn't find anything
> wrong. The routing tables for both machines look okay - They're
> appended here.
>
> Here's the statement of the problem:
> I have two Linux boxes (one a dual boot) and just purchased two 3c509s
> with a hub. I'm having some trouble with the dual boot machine.
>
> The dedicated Linux box has address 10.0.0.1 and the dual boot
> 10.0.0.2. When 1 pings 2, tcpdump shows messages being sent from 1 to
> 2, but no response from 2. Ping reports similarly (sometimes, I get a
> packet returned, but rarely. So far, it hasn't happened when tcpdump
> has been running).
>
> When 2 pings 1, tcpdump shows a normal request and reply pattern, but
> ping responds oddly. First, only even numbered requests are shown as
> having replies, and their replies have duplicates. Second, the
> reported time of the return trip is huge. It's almost as if the whole
> packet has been shifted by one bit or something.
>
> This problem is only present when the dual boot (2) is running Linux.
> Everything seems to work peachy when Win95 is running. I conclude
> that something is screwy with my Linux setup on 2. However, it
> doesn't seem to be the kernel -- I tried installing 1's kernel in 2,
> as a test. No difference in the symptoms.
>
> Here's the results of netstat -rn:
>
> (10.0.0.2)
> Kernel routing table
> Destination Gateway Genmask Flags Metric Ref Use Iface
> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 1 lo
> 10.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 eth0
> 0.0.0.0 10.0.0.1 0.0.0.0 UG 0 0 0 eth0
>
Looks like some of the problem is here ^^^^, as you seem to be
specifying "1" as your gateway. If 10.0.0.1 is not your gateway then
all packets should be sent to eth0. Did you also compile the right
driver for your card? Also, your localnet destination netmask should be
255.255.255.0, most of the time. I would say that a correct table, with
"1" as a gateway should look like:
Kernel routing table
Destination Gateway Genmask Flags Metric Ref
Use Iface
127.0.0.0 0.0.0.0 255.0.0.0 U 0
0 1 lo
10.0.0.0 0.0.0.0 255.255.255.0 U 0
0 0 eth0
0.0.0.0 10.0.0.1 0.0.0.0 UG 0
0 0 eth0
Additionally, the subnet 192.168.x.x is typically reserved for private
LAN use. Try setting the netmask on both systems to 255.255.255.0. Any
information on your Win95 setting may help too. Probably the problem is
the gateway setting above though. Unless "1" is specifically told to
route packets from "2" it will not do so.
> (10.0.0.1)
> Kernel routing table
> Destination Gateway Genmask Flags Metric Ref Use Iface
> 128.2.13.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 4 lo
> 10.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 eth0
> 0.0.0.0 128.2.13.1 0.0.0.0 UG 0 0 168 ppp0
>
> 128.2.13.1 is my P-t-P address. I don't understand routing well, but
> this looks okay to me. Am I mistaken?
>
HTH,
-Matt
--
Matthew O. Kressel | INTERNET: [EMAIL PROTECTED]
+--------- Northrop Grumman Corporation, Bethpage, NY ---------+
+--------- TEL: (516) 346-9101 FAX: (516) 346-9740 ------------+
------------------------------
From: Wisquatuk <[EMAIL PROTECTED]>
Subject: Re: Port Forwarding
Date: Thu, 14 Jan 1999 22:06:53 GMT
Danz <[EMAIL PROTECTED]> wrote:
> Is there a way to have it resolve to a domain name? [...] I am on a
> cable modem using IP Masq and wish to have my web server on the
> inside using a fake ip (192.168.1.3).
Well, you'll have to do one of two things; either install your
own DNS and follow their instructions on setting up domains, or put
the proper entries in the /etc/hosts of every machine attempting to
resolve the webserver's address.
I'd recommend the latter approach for a small network, and the
former for a very large one (as it's easy to update the records for
the whole network that way) or one with machines using Lose95 (as I
don't believe there's a way to map addresses using those, but I could
be wrong). Also, having your own DNS reduces load on your line to the
internet, and will make looking up the same address over and over a
bit faster -- for example, when your browser looks a site up to get a
webpage, then to get an image, then to get another image, etc. Good
luck!
--
Wisquatuk (name[1..4]@netrover.com to e-mail)
------------------------------
From: jfk <[EMAIL PROTECTED]>
Subject: Re: PPP serial connection to nowhere
Date: Wed, 13 Jan 1999 09:33:56 +0000
Dennis J. Haggerty wrote:
> Hi. I've got what is probably an unusual configuration. I have a serial
> connection into our ISP which is another division of our company. I'm
> running Redhat 5.1 and using KPPP to connect to the ISP. The problem is I
> cannot see any sites beyond our web server. I've tried altering the gateway
> and the DNS addresses to no avail. If I set the default gateway to the IP
> address of our router as suggested by the administrator ppp fails. The PC is
> also on a LAN and the UNIX host and the ISP both have the same domain name
> though different IP addresses. This set up works fine from home with a
> dialup into the ISP. Can anyone tell me how to correct this situation?
>
> TIA
>
> Dennis
HOW CAN I RECIVE MY E-MAIL USING 'CRONTAB ' WITHOUT TELNETING HOST ?
------------------------------
** 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
******************************
