Linux-Networking Digest #36, Volume #12 Wed, 28 Jul 99 14:13:42 EDT
Contents:
Help with finding network driver!! 3Com model 3CXFE575BT ("Stephen Jones")
Re: Works statically, but can't get DHCP working (Bob Tennent)
Re: Masquerading using odd ip address ("Jeff Volckaert")
KDE & firewall ("Michael.Rueckbrodt")
ppp proxyarp (Matt Templeton)
glibc + ppp = half-speed? ("David E. Smith")
Re: FTP deamon and NAT (Attik System)
Re: Remote Printing and RedHat 6.0 ([EMAIL PROTECTED])
Re: rsh, rcp Reliant Unix <-> Linux ("WKuhn")
Registering a domain name ("YouDontKnowWho")
Re: 40Sec Delay When FTP gets the USER on MASQ:ed machines (Michael Shuldman)
MS Proxy Client (Paul Cook)
Re: How to config AnonFTP for default binary transfer mode? (Monte Phillips)
NIS + Samba: Any pitfalls? (Chris Hobbs)
Squid & PPP links ("Matt Willsher")
Re: FTP server supporting continue-broken-download? (John Thompson)
Re: Registering a domain name ("Tim Hall")
Re: 2 ppp's / 2 isp accts (Clifford Kite)
Re: Works statically, but can't get DHCP working (Tim Koruna)
Re: Appletalk stopped working. Can't figure out how to repair it. (Frank Chew)
got NNTP and NEWS server timeout using Masq'g ("David Yuan")
----------------------------------------------------------------------------
From: "Stephen Jones" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,linux.redhat.install
Subject: Help with finding network driver!! 3Com model 3CXFE575BT
Date: Wed, 28 Jul 1999 11:04:55 -0400
I am trying to get linux (Redhat 6.0) working on my laptop (Dell Latitude
Cpi) and I can't find a driver that will work with my NIC. It is a 3Com
Megahertz 10/100 LAN CardBus PC Card (model 3CXFE575BT).
Anyone know where I can find a driver for this card?
Thanks
Stephen
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Bob Tennent)
Subject: Re: Works statically, but can't get DHCP working
Date: 28 Jul 1999 15:11:11 GMT
Reply-To: rdt(a)cs.queensu.ca
On Wed, 28 Jul 1999 14:14:10 GMT, [EMAIL PROTECTED] wrote:
>Trying to get a RH5.2 system on a Rogers@Home cable modem to work
>through DHCP. This cable modem has been servicing a singular Win95
>system without much problems for months now through DHCP. Now, I want
>to share the connection amongst a bunch of computers through the RedHat
>working as an IP masquerading server.
>
>Okay, I've been trying to get the Linux working through DHCP, and I
>went through every inch of the DHCP mini HOWTO manual, and nothing has
>worked so far. The only thing that has worked is for me to turn off
>DHCP and just manually configure everything statically. It works
>beautifully, and various of the local participants say I should just
>use it that way (statically), but I don't want to be the a**h*l* that
>breaks the network's standards.
>
>I think the problem is that the DHCP server at Rogers is using a
>special feature of Windows to identify itself to the lan. The Rogers
>technicians went and specifically setup the Windows machine's computer
>name and workgroup settings to particular values in the Networking
>control panel prior to enabling DHCP on the Win95 box, and I'm trying
>to recreate that on Linux but with no success so far. Previously I was
>under the impression that they must be using the machine's MAC address
>to authenticate to DHCP. The Win95 machine and the Linux machine aren't
>the same machines, they have different MAC addresses. I'm not sure what
>to think: is it the MAC address that is the problem or the machine
>name/workgroup name? Any ideas?
>
Try switching the network cards. I believe the MAC address that you're
speaking of belongs to the card. But I have the same problem as you
and am using the right card, so that may not be the (only) issue.
Bob T.
------------------------------
From: "Jeff Volckaert" <[EMAIL PROTECTED]>
Subject: Re: Masquerading using odd ip address
Date: Wed, 28 Jul 1999 10:49:56 -0400
No problem, I use .20 for my PIX firewall. Just make sure you point your
default gateway to that IP so the PC can find a route to the internet.
Jeff Volckaert
Bill Somerville <[EMAIL PROTECTED]> wrote in message
news:F6an3.331$[EMAIL PROTECTED]...
> As far as I know, you can use any valid private IP address you want. I'm
> using 192.168.0.10 for my server.
>
> -- Bill
>
> Scott W. Petersen - N9SLA <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > We would like to use and ip address ending in .250 rather than .1 on
> > the linux masq server.
> >
> > .1 is in use by a major server and would not like to change it if
> possible.
> >
> > Can we use .250 or do we have to use .1?
> >
> > Scott
>
>
------------------------------
Date: Wed, 28 Jul 1999 17:30:12 +0200
From: "Michael.Rueckbrodt" <[EMAIL PROTECTED]>
Subject: KDE & firewall
Hi folks,
I'm looking for a firewall with a GUI under KDE.
what is the name of the program and where can i find it?
Thnx in advance
------------------------------
From: Matt Templeton <[EMAIL PROTECTED]>
Subject: ppp proxyarp
Date: Wed, 28 Jul 1999 15:20:09 +0000
I am using pptpd to set up a a vpn for a windows98 client dialing into
the internet. So far I am NOT doing any encryption.
In my /var/log/messages file I get the following messages when I try to
connect from a windows 98 client
CHAP peer authentication succeeded for <user name>
Can not determine ethernet address for proxy ARP
local IP address 192.168.2.100
remote IP address 192.168.2.150
vpn allows for the connection of reserved IP addresses such as 192.168.2
over the Internet. The connection works fine for most IP connections but
I need proxyarp to work right so I can get windows to login to a MS
network.
the lan network address that is connected to the Linux box is 192.168.2
The only really strange thing that I can find is the netmask on the
windows side gets set to 255.255.255.0 while the subnet mask on the
Linux side gets set to 255.255.255.255. I have tried to use the netmask
option in the /etc/ppp/options file to set the netmask to 255.255.255.0
but it does not seem to take...
can anyone help me past this problem???
------------------------------
From: "David E. Smith" <[EMAIL PROTECTED]>
Subject: glibc + ppp = half-speed?
Date: 28 Jul 1999 15:13:24 GMT
I doubt the subject line is correct, but I'm at a loss as to what else
might be the problem.
I have a working PPP setup, on a hideously beaten Slackware system. pppd
is version 2.3.5, kernel 2.0.33, and for dialing I use WvDial 1.20.
So I finally decided to PAY for Internet access, and found a company that
supports V.90, to get the full "56" Kbps. Given the phone lines and all,
it averages about 45Kbps. But that's not the problem.
I cleaned off one of my extra hard drive partitions and started installing
Debian (2.1). I use the same kernel for both systems (yay lilo!), and the
"stable" versions of everything else from Debian. pppd 2.3.5, wvdial 1.20.
Same modem init strings. Same /bin/setserial configuration. Everything I
can think of is the same.
When I dial in to the same ISP under the Debian setup, I can't get more
than about 20Kbps out of my modem. Period. I've tested it with pings
(roughly twice as long), and large file transfers from the same site. >10M
files, to (hopefully) minimize fluctuations due to phases of the moon.
I've tested it at different times of day, from/to different sites. Always,
it's the same result: the newer setup is about half the speed of the
current Slack-based one.
I have two theories:
1. I have the WORST luck in the world, such that I keep getting noisy
phone lines every time I happen to boot the Debian system, and cleaner
ones when I boot Slack.
2. Something I can't account for in Debian/glibc/whatever causes my modem
to start smoking crack.
Any ideas? I can't think of other things I might be overlooking --
certainly nothing that would consume half a modem's worth of bandwidth...
...dave
------------------------------
From: Attik System <[EMAIL PROTECTED]>
Subject: Re: FTP deamon and NAT
Date: Wed, 28 Jul 1999 15:45:19 GMT
In article <7nlchk$r8q$[EMAIL PROTECTED]>,
Attik System <[EMAIL PROTECTED]> wrote:
> Does anyone have idea what else I could check to find the problem? I
> have the impression the problem is due to some kind of permissions,
> I don't know where. "chroot()" maybe?
Hello again,
Someone suggested me to test different MTU values through private
email, and this is what I did:
1) reboot the server
2) try sending a 800K file to the server, from behind the NAT router,
with an MTU of 1500: it fails.
3) try again: it fails.
4) try again: it fails.
5) change the MTU of my machine to 500. (ifconfig eth0 mtu 500)
6) try again: OKAY!!!
5) change the MTU of my machine back to 1500.
6) try again: OKAY this time!!!
I could understand it would always fail with an MTU of 1500, but why
does it work with 1500 after it worked with 500? This is crazy...
Is there some utility (MTU path discovery something) that could allow
me to find where exactly there is a problem in the connection? I cannot
say if the problem is due to my router, to the server, or to the ip
tunnel someone installed, maybe the Ethernet switch? Kernel 2.0.36?
2.2.10?
--
Attik System
Philippe Lang
Switzerland
http://www.attiksystem.ch
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Remote Printing and RedHat 6.0
Date: Wed, 28 Jul 1999 15:54:47 GMT
I checked the bug tracker on the redhat site. Go to
http://developer.redhat.com and then go to bug track. You'll have to
register.
It seems that this is a known problem with the version of lpd that is
distributed with Redhat 6.0. One of the comments was that this was
fixed in the rawhide version. If you do a search you can find the
information. Some one supplied the information for patching it.
I don't know why they've not issued a fix for this yet.
James Hubbard
[EMAIL PROTECTED]
In article <7na6qs$qc6$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> I've seen this question posted in various spots, and I'm currently
> running into the problem myself. Currently, my workaround is to spool
> from remote print servers to the primary print server using Samba.
I've
> tinkered with modifying permissions on the spool directories, but to
no
> avail. I'd forward the question to Red Hat support if I knew where to
> send it.
>
> Regards,
>
> Glenn
>
> In article <7n53a3$3tt$[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] wrote:
> > I have several linux boxes running file and print services. I have
> > remote printing setup so that print jobs destined for the one of the
> > other networks goes to the local server, then to other network
server,
> > and finally it is sent to the printer.
> >
> > My problem is that I recently upgraded one of the servers to RedHat
> 6.0
> > and now it no longer accepts remote print jobs. Here are the error
> > messages:
> >
> > lpd[18145]: Can't create temp cfp file
> > lpd[18145]: w_engineer: can't scan /var/spool/lpd/w_engineer
> >
> > Is this a bug with the lpd? I've noticed that this was listed as
bug
> on
> > the Debian bug list when doing a www.deja.com search. Does anybody
> know
> > what I can do to fix this?
> >
> > Thanks
> > James Hubbard
> >
> > Sent via Deja.com http://www.deja.com/
> > Share what you know. Learn what you don't.
> >
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: "WKuhn" <[EMAIL PROTECTED]>
Crossposted-To: de.comp.os.unix.sinix
Subject: Re: rsh, rcp Reliant Unix <-> Linux
Date: Wed, 28 Jul 1999 16:57:25 +0200
Bei Reliant Unix gibt es die Datei /etc/default/login, wo man grunds�tzliche
Einstellungen bzgl. login usw. parametrisieren kann.
Gru�
Wolfgang
Roland Weber <[EMAIL PROTECTED]> schrieb in im Newsbeitrag:
7nn6he$k67$[EMAIL PROTECTED]
> Hallo,
> ich habe heute versucht v. meinen Reliant Unix Kisten (mit 5.44 u. 5.43)
> mit rsh, und rcp auf einen Linux PC (SuSe 6.1) zu kommen.
> Hat nicht funktioniert, .rhosts bzw. hosts.equiv scheinen ignoriert zu
> werden,
> ich werde immer nach einem Passwort gefragt. In der anderen Richtung
> (Linux -> Reliant) genau das selbe. Die Zugriffsrechte der Dateien sind
> jeweils ok.
> Koennen die beiden vieleicht schlicht nicht miteinander??
>
>
> Roland Weber
>
>
>
------------------------------
From: "YouDontKnowWho" <[EMAIL PROTECTED]>
Subject: Registering a domain name
Date: Wed, 28 Jul 1999 15:56:21 GMT
I'm interested in registering my own domain name. At least, that's
what I think I need to do in order to be able to have an official DNS
entry somewhere so people can find my site. Basically, I want to set
up all my own stuff at home and have the ability for other people to
access it.
How do I go about getting an IP address and an official domain name?
What is the best way to get that accomplished?
Thanks!
--
Principle of Minimum Access: "That which is not explicitly permitted
is denied."
ANNOUNCER: And now we return to our regularly scheduled, uncommonly
entertaining thread...
------------------------------
From: [EMAIL PROTECTED] (Michael Shuldman)
Crossposted-To:
alt.os.linux,comp.security.firewalls,alt.os.linux.slakware,alt.linux.slakware
Subject: Re: 40Sec Delay When FTP gets the USER on MASQ:ed machines
Date: 28 Jul 1999 16:03:47 GMT
In comp.security.firewalls Dave Perks ([EMAIL PROTECTED]) wrote:
> Jesper Dybdal wrote:
> >
> > "TRY" <[EMAIL PROTECTED]> wrote:
> >
> > >You may see this delay due to the FTP server trying to establish an ident
> > >conection to your machine to "authenticate" you, if your firewall is set up
> > >to discard incoming ident requests. If you choose to reject ident messages
> > >(I think it is called reject in Linux parlance -- I mean: make the firewall
> > >return a "network unreachable"/"host unreachable" message) instead and the
> > >reason for the delay is ident messages, you should see a substantial
> > >speed-up due to this.
>
> > I set it up this way once, using ipfwadm to reject connections to
> > the auth/ident port.
> >
> > It worked fine with _almost_ every mail server we communicated
> > with.
> >
> > But there was at least one mail server out there who refused to
> > talk to us when the auth request was rejected in that way.
>
> Yes, "network unreachable" code 0 or "host unreachable" code 1 can
> discourage the originating system... better to use the more accurate
> "port unreachable" code 3.
No. The correct response is RST (reset), see rfc793.
--
_ //
\X/ -- Michael Shuldman <[EMAIL PROTECTED]>
------------------------------
Date: Wed, 28 Jul 1999 19:39:46 +1000
From: Paul Cook <[EMAIL PROTECTED]>
Subject: MS Proxy Client
Hi.
I am at a University which implements the use of MS Proxy Server on an
NT box. Each client requires that Proxy Client be installed for logging
and monitoring of data. Is it possible to either configure Linux (Redhat
6) or download software for clients so that they can use this service??
I really want to get Linux going on the net using this setup.
Any help is much appreciated.
Paul Cook
------------------------------
From: [EMAIL PROTECTED] (Monte Phillips)
Subject: Re: How to config AnonFTP for default binary transfer mode?
Date: Wed, 28 Jul 1999 15:23:46 GMT
Look in the AnonFTP manual, it is in there.
"Steve Snyder" <[EMAIL PROTECTED]> wrote:
>How does one configure the anonymous ftp server that comes with RedHat
>v6.0 such that binary is the default transmisstion mode? I'm getting
>tired of people uploading ZIP files to my site in text-mode format.
------------------------------
From: Chris Hobbs <[EMAIL PROTECTED]>
Subject: NIS + Samba: Any pitfalls?
Date: Wed, 28 Jul 1999 09:07:39 -0700
I've not recieved all of my hardware yet, so I haven't been able to try
this out yet - I'm hoping someone here can shed some light on it before
I start down the wrong track...
I'm setting up the following network on a high school campus. I'm
interested in using both NIS (to allow the app servers to authenticate
against one passwd database) and Samba as PDC (to allow WinNT machines
to access /home/*) on the file server.
===================
| 120 X Terminals |
===================
|
|
===================
| 6 Application |
| Servers |
| (NIS Clients) |
===================
|
|
=================== ================
| File Server | | 35 WinNT |
| (NIS + Samba |-------| Workstations |
| Servers) | | |
=================== ================
Are there any problems with running both NIS Server and Client software
on the File Server, and still have that box act as PDC for the WinNT
boxes, or should I look at setting up a separate box (one of the app
servers?) as the NIS server, and just make the file server an NIS client
and Samba server.
Looking forward to ideas and suggestions,
--
Chris Hobbs [EMAIL PROTECTED]|Network Admin | Web/Postmaster
Silver Valley Unified School District |MCP - TCP/IP | J.A.P.H.
http://www.silvervalley.k12.ca.us/chobbs |A witty saying proves nothing.
- Voltaire
------------------------------
From: "Matt Willsher" <[EMAIL PROTECTED]>
Subject: Squid & PPP links
Date: Wed, 28 Jul 1999 16:59:58 +0100
A simple question:
Is Squid worth using on a stand alone machine sitting on a 56kbps dialup
machine? If so, what sort of cache would be useful, assuming the machine has
6Gb of slack space and browsing tends to be to the same sites or ones of a
similar theme. What sort of speed increase would be possible, on average?
Whould it be better than Netscape?
Lots of questions. Can anyone help?
Matt.
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Subject: Re: FTP server supporting continue-broken-download?
Date: Wed, 28 Jul 1999 09:24:56 -0600
Our House wrote:
> Is there a FTP server Linux which supports continuing broken downloads?
wu-ftpd does and comes with most distributions.
--
-John ([EMAIL PROTECTED])
------------------------------
From: "Tim Hall" <[EMAIL PROTECTED]>
Subject: Re: Registering a domain name
Date: Wed, 28 Jul 1999 12:31:14 -0400
Go to www.networksolutions.com. They handle domain registrations (.com,
etc). You have to pay a little ever year but is not much.
YouDontKnowWho <[EMAIL PROTECTED]> wrote in message
news:FKFn3.27946$[EMAIL PROTECTED]...
> I'm interested in registering my own domain name. At least, that's
> what I think I need to do in order to be able to have an official DNS
> entry somewhere so people can find my site. Basically, I want to set
> up all my own stuff at home and have the ability for other people to
> access it.
>
> How do I go about getting an IP address and an official domain name?
> What is the best way to get that accomplished?
>
> Thanks!
>
> --
> Principle of Minimum Access: "That which is not explicitly permitted
> is denied."
>
> ANNOUNCER: And now we return to our regularly scheduled, uncommonly
> entertaining thread...
>
>
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: 2 ppp's / 2 isp accts
Date: 28 Jul 1999 11:26:55 -0500
Ron Bombard ([EMAIL PROTECTED]) wrote:
: I have 2 accounts with the same ISP. And they've been pretty
: accomidating in the past.
: If you could pass along any info, or point me in the correct direction
: as to what is required, I'd appreicate it.
There are two solutions I know about, Multilink PPP (MP) and load
balancing.
There are two favors of MP for joining independent links to the same
ISP, a Terminal Adapter (TA) or a card. The TA is like a modem in that
it only needs pppd, the aggregation of the PPP links is done in the TA.
A card needs a kernel driver and a modified pppd, I *think*. Either can
be a solution in Linux. Here are some links for card MP:
ftp://ftp.east.telecom.kz/pub/src/networking/ppp/multilink/
http://linux-mp.terz.de
The .kz site is being actively developed.
AFAIK, the load balancing available in Linux consists solely of EQL. This
is described in kernel source file linux/Documentation/networking/eql.txt.
There doesn't seem to be any further development from what is described
there. I think there is an eql tarball somewhere on sunsite. Here's a
link to a how-to that may be the same as the eql.txt file.
http://abies.com/eql-howto.html
I tried to post this earlier but I don't think it got sent. But if there
is another similar to this then it did get sent. :^
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* To extract lines: View file with "vi -R". Move cursor to first line.
Press "v". Move cursor to mark lines (Esc unmarks). Write lines to
fubar with ":w fubar <Enter>". Exit with ":q <Enter>". */
------------------------------
From: Tim Koruna <[EMAIL PROTECTED]>
Subject: Re: Works statically, but can't get DHCP working
Date: Wed, 28 Jul 1999 12:27:30 -0400
[EMAIL PROTECTED] wrote:
> I think the problem is that the DHCP server at Rogers is using a
> special feature of Windows to identify itself to the lan. The Rogers
> technicians went and specifically setup the Windows machine's computer
> name and workgroup settings to particular values in the Networking
> control panel prior to enabling DHCP on the Win95 box, and I'm trying
> to recreate that on Linux but with no success so far. Previously I was
> under the impression that they must be using the machine's MAC address
> to authenticate to DHCP. The Win95 machine and the Linux machine aren't
> the same machines, they have different MAC addresses. I'm not sure what
> to think: is it the MAC address that is the problem or the machine
> name/workgroup name? Any ideas?
This is guesswork, but it's informed guesswork <G>.
The dhcpd daemon has an flag called 'boot-unknown-clients' that, if set
to false, will only assign addresses to "known" clients. Known clients
will have an entry in the dhcpd.conf file that specifies their MAC
address and may also specify the machine name. In short, it could be
both the name *and* the MAC address that must be configured properly.
What happens if you swap ethernet cards in the Win95 machine? Does DHCP
still work? If so, it's not a MAC address issue. What happens if you
change the host name of the Win95 machine? Does it work then? If not,
then the name is relevant. You can try changing each of those in
isolation on the Win95 machine to determine if it's the MAC address, the
machine name, or both.
Hope this helps.
TK.
------------------------------
From: Frank Chew <[EMAIL PROTECTED]>
Subject: Re: Appletalk stopped working. Can't figure out how to repair it.
Date: Wed, 28 Jul 1999 12:39:05 -0400
Well, I have fixed the first three error messages by adding the full
paths.
I still get the error:
/etc/rc.d/rc3.d/S79atalk: [: =: unary operator expected
/etc/rc.d/rc3.d/S79atalk: [: =: unary operator expected
During the boot. I do not see a "[:" string in the S79atalk file, what
should I be looking for? Any ideas? Thanks.
-Frank
[EMAIL PROTECTED]
------------------------------
From: "David Yuan" <[EMAIL PROTECTED]>
Subject: got NNTP and NEWS server timeout using Masq'g
Date: Wed, 28 Jul 1999 16:58:34 GMT
hi,
my system is RH6.0 using IP Masq'g with ipchains. telnet, ftp, www all seem
working but pop mail reader and news reader from the client machines. It
keeps getting timeout when downloading data from those server. I am
wondering is there a special module or ipchain command on the gateway?
i read through all IP Mask'g documents i can get and discussions for this
issue. it seems no other people have the problem.
Thanks,
David
------------------------------
** 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
******************************
