Linux-Networking Digest #918, Volume #11 Fri, 16 Jul 99 19:13:36 EDT
Contents:
Re: Shareware DNS Server source???? (Ken MacFarlane)
Radius Dilemma ([EMAIL PROTECTED])
Re: domainname? (peter)
Re: PPP-After get connected and ppp0 configured, I can't see route table !!!
(Clifford Kite)
Re: NFS problems in Redhat Linux 6 (Tommi Komulainen)
win95 / linux IP routing? (Ron Bombard)
internal routing, gateway problem (Christopher Kolar)
Re: Firewalls when I don't control the entire subnet (Matthew Hanselman)
Re: cable modem w/ nonstatic IP (peter)
copy from floppy to hard drive??? (mike schilling)
Re: ppp as non-root (Gerald Krause)
Re: Apache: You do not have permission to access ([EMAIL PROTECTED])
Re: My Dissapointment to find Linux not a viable solution
([EMAIL PROTECTED])
Re: I need help on configuring a RAID 1 (Ryan Hughes)
Re: Need help connecting 2 PC's please (Rogue Eagle)
Re: Another problem with multiple NICS ([EMAIL PROTECTED])
Re: My Dissapointment to find Linux not a viable solution ("Jason Chard")
Newbie Q: Getting message 'RPC program not registered ' when trying to mount (Peter)
Re: NT can't login, can telnet to Linux PC ([EMAIL PROTECTED])
Recognizing and configuring NICs on a crossover ``notwork'' (Greg Tabaka)
Re: Linux PPP server problems (The Sweep)
----------------------------------------------------------------------------
Crossposted-To: comp.unix.solaris
Date: Fri, 16 Jul 1999 16:38:26 -0400
From: Ken MacFarlane <[EMAIL PROTECTED]>
Subject: Re: Shareware DNS Server source????
[mailed & posted]
Try BIND at http://www.isc.org/.
On Tue, 13 Jul 1999 [EMAIL PROTECTED] wrote:
> Does anybody know how I can get my hands on
> the source for a shareware DNS server (preferably
> multi-threaded)???
--
Ken MacFarlane
<[EMAIL PROTECTED]>
http://members.dca.net/ksm/
------------------------------
From: [EMAIL PROTECTED]
Subject: Radius Dilemma
Date: Fri, 16 Jul 1999 20:33:16 GMT
We have a primary Radius server running on a Sparc2 (old software) that
has been working great. The previous people to set it up pretty much
hacked it to death to make it work for them. Apparently it uses the
normal "users" file, but a "uusers" for the Shadow passwords.
To get a backup online, I installed Red Hat 6 on a Sparc5 that we had
laying around. To keep things consistant, I copied over the same
Radius sources and finally got them to compile. Now, if I hand enter
the password in the "users" file it works just great. But if I set it
to "UNIX" to use the Shadow passwords, it fails every time. I compiled
the Radius binaries with and without Shadow support, same outcome for
both sets of executables.
I know the Red Hat box is using /etc/shadow for the Shadow passwords.
Is there a way to compile Radius to use this "uusers" file instead?
Maybe that could help. I am looking for any help! Thanks.
--
[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (peter)
Subject: Re: domainname?
Date: Fri, 16 Jul 1999 21:28:18 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] told
us...
> I am trying to setup domainname on redhat 5.2
> linux by issueing "domainname NAME", cat > /etc/defaultdomain with the
> same NAME and set the domainame NAME in /etc/sysconfig/network, but
> after rebooting it still doesn't know the domainname.
>
are you using NIS ? thats what domainname is for.
man hostname will give you the needed information.
peter
=================
pilsl@
ANTISPAM
goldfisch.atat.at
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: PPP-After get connected and ppp0 configured, I can't see route table !!!
Date: 16 Jul 1999 16:12:56 -0500
Terry Cox ([EMAIL PROTECTED]) wrote:
: I am having a similar problem. I connect to my ISP via PPP. A defaultroute
: through my ppp0 interface is not set up. I have to manually set it up: #
: route add default gw 165.234.17.28 . After I do that, I can get it to work.
: I use 'ifconfig' to find out what my P-t-P ip address is. I am trying to
: figure out how I can get my default route to always be my ppp0 interface. I
: am lurking in the newsgroup to see if anyone else is have the same problem.
Use the pppd defaultroute option and remove any default route to a local
network. A network-specific route works fine for the local network, e.g.,
/sbin/route add -net 123.45.67.0 netmask 255.255.255.0
Just use your local network address and netmask instead of the bogus
one above.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* Speak softly and carry a +6 two-handed sword. */
------------------------------
From: Tommi Komulainen <[EMAIL PROTECTED]>
Subject: Re: NFS problems in Redhat Linux 6
Date: 16 Jul 1999 23:31:02 +0300
"Dr. Ram Samudrala" <[EMAIL PROTECTED]> writes:
> Does anyone know why the "nfssvc not implemented" error occurs even
> though NFS is compiled into the kernel? And how can I compile 2.2.10
> to not have this error? Thanks!
I guess that some parts of the kernel don't get properly (re)compiled
after configuration changes. I once had a similar problem (not with
nfs though) which was resolved by removing all the files from
linux/include/linux/modules directory before issuing make config dep
clean etc.
Works for me at least, I'm running 2.2.10 with knfsd compiled as a module.
Good luck.
--
-=# Tommi Komulainen ## [EMAIL PROTECTED] #=-
------------------------------
From: Ron Bombard <[EMAIL PROTECTED]>
Crossposted-To: redhat.networking.general
Subject: win95 / linux IP routing?
Date: 16 Jul 1999 21:37:51 GMT
Greetings!
I have a RedHat 6.0 machine with a ppp connection to my ISP. My Linux
box has a cache-only named server running, with my isp's DNS server
listed in my /etc/resolv.conf file.
Works fine for browsing and email and whatnot. On the same network I
have my PC.
I'm trying to connect my PC to the internet using my Linux box's
connection.
In the network settings on my PC, I have my linux box listed as the
Gateway. I have my Linux box listed in the DNS server search table. I
have my linux box listed as my WINS server.
I can ping my linux box from the pc. I can ping my pc from my linux
box.
I can ping a hostname on the internet (ie: ping www.commercial.com )
and it will do the address resolution but the ping times out.
# ping www.commercial.com
Pinging www.commercial.com [198.69.28.230] with 32 bytes of data:
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Any suggestions? Or if you can point me to a good doc, I'd appreciate
it.
Thanks
Ron
--
Ron Bombard, Network Administrator
[EMAIL PROTECTED]
PO Box 2567, Glens Falls, Ny 12801
http://members.theglobe.com/virtual_ron
Sometimes loosing a wife can be hard... in my
case it was nearly impossible!!!
===================================================
_O_ _____ _<>_ ___
/ \ | | / \ / _ \
|==/=\==| |[/_\]| |==\==/==| | / \ |
| O O | / O O \ | >< | | |"| |
\ V / /\ - /\ ,-\ () /-. \ X /
/`---'\ /`---'\ V( `-====-' )V /`---'\
O'_:_`O O'M|M`O (_____:|_____) O'_|_`O
-- -- -- -- ---- ---- -- --
STAN KYLE CARTMAN KENNY
------------------------------
From: [EMAIL PROTECTED] (Christopher Kolar)
Subject: internal routing, gateway problem
Date: Fri, 16 Jul 1999 21:20:17 GMT
[Note: IP numbers have been changed to placeholders.]
I am running COL 2.2 on my desktop. My machine is on one of our class
C network with the address 123.456.789.220. Our mail/web/news servers
are at 987.654.321.x. Our gateway is officially 987.654.321.1. OK so
far. For a long time everything (well, the networking anyway) was
working with these settings under NT.
When I converted from NT, the interface configuration utility
complained about the gateway, and I could not see anything on the
network. When I changed the gateway to 123.456.789.1 I could again
reach the outside world.
The problem is that when I stay on our side of the gateway I can
connect to servers but then packets get dropped and the connection
hangs. For fun I got a 987.654.321.x address and changed the gateway
back to the 987. number, but I still had the same problem with hosts
that were on our side of the router -- long traceroute times, dropped
packets, and then timed out sockets; so no working internal network
connections, only external.
I am guessing that there is a simple routing issue that needs to be
addressed, but I am at the end of my understanding. Thanks for any
help that you can suggest. I have it set to use the default path, and
my broadcast address is .255. The only weirdness is that when I
toggle the default path option is replaces the gateway with
123.456.789.254 instead of .1.
Thanks in advance for any suggestions on what to look into. It isn't
so bad for doing things in the outside world, but it would be nice to
read news and check mail -- for work purposes that is.
Cheers,
--chris
--
/////\\\\\/////\\\\\
Christopher G. Kolar
Director of Instructional Technology
Aurora University, Aurora, Illinois
[EMAIL PROTECTED] -- www.aurora.edu
[Public Key http://certserver.pgp.com:11371/pks/lookup2?op=index&search=0x5B61A799]
------------------------------
From: [EMAIL PROTECTED] (Matthew Hanselman)
Subject: Re: Firewalls when I don't control the entire subnet
Date: 16 Jul 1999 21:28:14 GMT
I thought of public addresses I guess because I don't know enough about
how this stuff works....
We want a bit of security; I know this is suboptimal, but currently we
have 1 server masquerading for a private 192.168.1 network. By disabling
telnet, etc to the masquerading machine, the private network is at least safer
than just leaving them on the net (I think?)
Problem is, masquerading breaks UDP and FTP within Netscape and IE. Perhaps
I should instead set it up to be a firewall and proxy everything? If I do
that, will the destination site see the proxy IP address or the firewalled
machine's IP address?
You can tell I'm new to this. :) Thanks for your input.
- Matt
In article <7mo3fa$pki$[EMAIL PROTECTED]>, Greg Leblanc <[EMAIL PROTECTED]> writes:
|> I'll have to poke around and find a subnet mask calculator to answer
|> that part of the question. It IS possible to firewall that way. Why do
|> you need/want public (routable) addresses on the other side of the
|> firewall? That will make it much easier to attack the machines behind
|> the firewall, and unless you have servers behind it, probably add little
|> functionality.
|> Greg
------------------------------
From: [EMAIL PROTECTED] (peter)
Subject: Re: cable modem w/ nonstatic IP
Date: Fri, 16 Jul 1999 21:26:27 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] told us...
> Thanks for your interest. Is there a way to share an internet connection
> (cable modem) using IP masq & Samba; even though my "level of service" does
> not provide a staticIP? That is my cable provider uses DHCP to assign me a
> new IP every session. I would be willing to leave my RedHat 6.0 box turned
> on all the time. I have purchased two 3c905's and an 8 port hub but have
> not yet setup my home LAN. What is the best solution? This is such a basic
> question it is scary!! If you have general software and/or hardware
> solutions a quick description would be appreciated. Thank you very much in
> advance.
> Stew
>
whats the problem ?
you have two cards in your server. one for the outside and one for the inside.
configure your inside LAN as if there was not outside. assign the static
private IP-adresses. (ie: 192.168.1.1 for the server and .x for the others) and
dont forget to add the needed routes for internal traffic. (do not set any
default-route)
then assign the outside card as if there was no inside, put dhcpd in your
startupfiles and assign default-route to the outside-card. (this route-
assignment does not need the IP for this card)
and then set up IP-MASQ. for this you just need to now the IP of your internal
LAN (ie 192.168.1.0) and the rest will be done by your routingtable.
information for ip-masq: http://www.tor.shaw.wave.ca/~ambrose/
peter
=================
pilsl@
ANTISPAM
goldfisch.atat.at
------------------------------
From: mike schilling <[EMAIL PROTECTED]>
Subject: copy from floppy to hard drive???
Date: Fri, 16 Jul 1999 16:35:17 -0500
what is the syntax to copy a file from floppy into a directory on a
Linux machine???
------------------------------
From: Gerald Krause <[EMAIL PROTECTED]>
Subject: Re: ppp as non-root
Date: Fri, 16 Jul 1999 23:33:12 +0100
Hi,
my way (Suse 6.0):
a) find out which *group(s)* /usr/sbin/pppd and the ppp-scripts
in /etc/ppp belonging to and verify that this group(s) have
*execution* rights to the files!
b) if the files belongs to "root" (as "group" not "owner"!) i
suggest you to create a new group (like "dialout") and change
the files to this new group
c) *add* this special group to your normal account
now you should be able to run ppp without loggin in as root
Gerald
[EMAIL PROTECTED] wrote:
>
> I have PPP setup to dial my ISP under RH6.0, and it works like a
> champ... as long as I am logged in as root. When I login as a regular
> user, I cannot access PPP. When I start ppp as root, and then logout
> and login as a user, I cannot access the internet. I can however, log
> back out and relogin as root and it is still connected and working fine.
>
> Any help is appreciated.
>
> Steve
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Apache: You do not have permission to access
Date: Fri, 16 Jul 1999 22:02:55 GMT
THanks It works!
In article <MPG.11f9aac7afe9042989699@server>,
[EMAIL PROTECTED] (tomislav) wrote:
> In article <7mnf7c$gji$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> > I have Apache installed and running, I can view ther html in the
httpd
> > directory, but users cannot view the html of there public_html
directory.
> > I set the permissions of the public-html directory to 755 and made
sure
> > that in /etc/httpd/conf/srm.conf the user directories are set to
> > public_html. The error message is You do not have permission to
access
> > filename on this server.
>
> You need to set the the permissions on the files in the public_html
> directory to 644.
>
> chmod 755 public_html
> cd public_html
> chmod 644 *
>
> --
> [EMAIL PROTECTED]
> http://tojo.home.dhs.org/pgp.asc
>
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.advocacy,comp.security.firewalls
Subject: Re: My Dissapointment to find Linux not a viable solution
Date: Fri, 16 Jul 1999 22:06:53 GMT
This last response sounds like something from the dark ages. Fast,
effective file server based av solutions have been around for years!!
Yes, viruses execute on the client. No, you do not have to execute
viruses to detect them. Since over 90% of virus, trojan and worm
incidents originate as files attached to emails, you can stop 90% of
virus incidents by scanning email traffic for viruses. we strongly
recommend scanning of all SMTP, FTP and HTTP - something our InterScan
VirusWall does - on Linus, Solaris, HP UX, DEX UX and NT.
As far as I know, InterScan is the only virus scanner scanning SMTP,
FTP and HTTP traffic available on Linux.
Desktop solutions are still important - users still have floppy disks
and other ways of getting files that don't go through your email server
and or firewall. However, the problem with desktop virus scanning
includes:
1. users may turn it off, change setting, not update them - or
interfere with the update process. Trend Micro's OfficeScan allows
administrators to centrally manage desktop virus protectio - taking end
users out of the proces.
2. Viruses such as Melissa and the Explorer worm spread too fast to
update desktops/home users. You need to have virus protection built
into the infrastructure of the environment - that can be
centrally/remotely updated and managed.
Dan Schrader
Trend Micro
http://www.antivirus.com
In article <[EMAIL PROTECTED]>,
mlw <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] wrote:
> >
> > NOT, the anti virus should run on incoming files to the network,
either
> > as a CVP or proxy between the LAN and the FW. Yes, having AV on the
> > desktop keeps floppy type viruses from propagating inside the
network
> > but inbound files from the internet should be scanned in 1 place and
> > updates need to be done MUCH more than every month or so!
> >
>
> Sorry, but you are wrong.
>
> Viruses spread by having their code executed by a client machine,
which
> then spreads the virus to other files to which the client has access.
> Having an anti-virus scanner running on the client machine is the only
> way to make sure the code is either not executed or unable to spread.
A
> virus can not spread if it is not executed. The file server does not
> usually execute the files it is serving. This is why Linux can server
> Windows machines. A virus infected executable can sit on a file server
> and not infect other network files until it is run by a client machine
> with write access to previously mentioned network files.
>
> Making program files and document templates read-only on the NOS is
very
> good idea no matter how you look at it.
>
> Lastly, you do not want a file server scanning for viruses on all
files.
> There is no way a file server can server perform its functions and
scan
> for viruses without being so slow that no one would use it.
>
> Think of it this way: A virus scanner uses up 4% of CPU time while
> running on an active client computer. Say the server system is 4 times
> faster than the client system. It would take 100 active connections to
> bottleneck the server in virus scanning alone. No, not a wise thing to
> do. (Yes I know the calculations are very simplistic, however the load
> is probably a much higher figure.)
>
> As for other things you say:
>
> "Inbound files from the internet" As in FTP, HTML, or Email? Sure
> programs are going to get on to a client machine, but they will not be
> executed by the firewall or file server. They will be executed by the
> client machine, which has anti-virus software on it.
>
> "Updates need to be done much more than every month or so!" Ok, what
> ever, not a big deal. Update every day and have users reboot every
> morning. There is client code for Windows which will allow you to do
> this.
>
> --
> Mohawk Software
> Windows 95, Windows NT, UNIX, Linux. Applications, drivers, support.
> Visit http://www.mohawksoft.com
>
--
Dan Schrader
Trend Micro
http://www.antivirus.com
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Ryan Hughes <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: I need help on configuring a RAID 1
Date: Fri, 16 Jul 1999 21:31:03 GMT
Ray,
The kernels that come with SuSE tend to suck. Try downloading a newer
kernel.
-Ryan
Raimund Sacherer wrote:
> Hi, the patch did not work --> when compiling the kernel, i get errors.
> I have the SuSE 6.0 linux.
>
> What is a pristine kernel source package?
>
> best regards
> The Ray
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
Date: Fri, 16 Jul 1999 22:23:32 +0000
From: Rogue Eagle <[EMAIL PROTECTED]>
Subject: Re: Need help connecting 2 PC's please
If you're talking about the 3-com cards, I'd be more than happy to, but they're
not mine. They are the property of my employer.
Sorry,
Steve
Monte Phillips wrote:
> I am willing to have you mail those things to me. Hell I'll even pay
> the postage.
> <G>
>
> Rogue Eagle <[EMAIL PROTECTED]> wrote:
> >Nevermind about this problem. I didn't get it to work with the 3-com
> >cards (3c509), so I installed Linksys 10/100 nic's in both PC's. I brought
> >up the interfaces just like before, only low and behold, this time it
> >works!!!!!!! I'd still be interested to hear any ideas anyone might have
> >about why the 3-com cards didn't work.
> >Thanks,
> >Steve
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Another problem with multiple NICS
Date: Fri, 16 Jul 1999 14:50:14 -0400
I think this deserves FAQ merit.
I am having similar problems. I can't seem to get my eth1 to be recognized.
When I do an ifconfig for eth1 I get an error about not known device.
my lilo.conf reads.
append="ether=10,0x300,eth0 ether=5,0x320,eth1"
is this correct? What else needs to be done.
-Elie
Charles Stack wrote:
> This subject would probably make an excellent "HOW-TO".
>
> Anyway, I just installed RH 6 on a P90. In it, I've got two Bay Network's
> 10/100 ethernet cards. The first is set to IP address 10.0.0.5. The second
> is 10.0.0.6. Both have a subnet mask of 255.0.0.0. And, I've added the
> customary append line to lilo.conf.
>
> When the machine boots, I see both lo, eth0 and eth1 interfaces start. If I
> do a ifconfig on either of the two ethernet adapters, I see they are both
> active and each has its own address and IRQ. And, I can communicate just
> fine through eth0. The problem is with eth1.
>
> The network lights are on for both cards. I can ping the IP address
> 10.0.0.6 from the linux box and get a reply. I just can't seem to be able
> to ping the card from anywhere on the network. I can't even ping another
> workstation through eth1.
>
> I'm sure this has to have a relatively simple solution, but I don't know
> where to start.
>
> Can someone please help.
>
> TIA,
>
> Charles
------------------------------
From: "Jason Chard" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.security.firewalls
Subject: Re: My Dissapointment to find Linux not a viable solution
Date: Fri, 16 Jul 1999 21:44:16 GMT
The only effective method in a large scale environment that I have come
across is to use a combination of all of the methods listed.
All antivirus software must be updated as often as possible. - If the
manufacturer updates it once a month then this is the rate at which we
update it. Emerjency updates must be applied as soon as they are avalible.
All incoming email must be checked for viruses. While I don't know of any
mail gateway scanners that will search inside of a compressed file (yet)
this is not as big a deal as it seems. Most people don't compress files
(or documents) that they are sending in an office environment. There are
packages out there but most of the better ones run on NT. But that doesn't
say you cant use a linux box as your firewall.
The server must be scanned for viruses. Contrary to popular belief any
place you store PC files should be scanned. Sometimes users disable there
virus scanners. Technicians forget to put it on there own systems. Remote
users haven't installed the software. BTW whoever posted earlyer about the
server being brought to a halt by 100 active connections and a virus scanner
but I have had servers with over 200 active connections (p100 w128) serving
files and print jobs and never had it run consitently at 100% (It did
occasionaly peak to 100% but thats to be expected in any environment).
The workstation must be checked for viruses. This is where viruses cause
the most problems hence this is where viruses must be checked for most
often. It is also a good idea to do periodic checks of departements to make
sure that users are using the antivirus software (its not turned of). Your
tech support personell can probably make this part of there regular rounds
(as they repair users systems).
Users must be educated. Send all of your users an email when each update to
the virus scanner is going to happen. Let them know what can happen if they
disable there virus scanner. Let them know what viruses have been found in
the building recently. This also tends to make users fearfull of what would
happen if they don't use antivirus software. Fear is also a good motivater
to prod users into backups (if they do them local to the workstation) but
thats another story.
Gary Flynn wrote in message <[EMAIL PROTECTED]>...
>
>I'm not sure which post to reply to :)
>
>Someone said block it before it gets in with firewalls, someone
>said do it on the server, and someone said do it on the client.
>
>1. Any detection mechansism depends upon knowing the code
> sequence. With people publishing source code for viruses
> and trojans, this can't be guaranteed. Worse, if someone
> knowledgable really wanted to target you, they'd write
> their own (fairly easy these days) and you'd never know
> it as they sucked whatever information was available to
> your computer out the net or scribble all over your
> hard drive.
>
>2. Worse, encryption or compression will further neutralize any
> pattern checking control measure.
>
>3. The ONLY solution is to educate your users on the dangers
> of executing unknown programs and to help them configure
> their desktops to prevent this from happening automatically.
------------------------------
From: [EMAIL PROTECTED] (Peter)
Subject: Newbie Q: Getting message 'RPC program not registered ' when trying to mount
Date: Fri, 16 Jul 1999 22:57:38 GMT
I have recently managed to get networking to connect between my laptop
and desktop machines, I can ping each machine but if I try to mount a
NFS filesystem from either end I get the error ' RPC program not
registered' can someone tell me how I do this, I have found no
reference to this in the manuals.
Any help gratefully accepted
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: NT can't login, can telnet to Linux PC
Date: Fri, 16 Jul 1999 21:00:18 GMT
>From another post, I found that NT needed to be configured to use a
plain text (not encrypted) passwords. Eventually, I found how to do
that. A registration entry is needed and the .REG file is in
/usr/doc/samba.../docs directory. Applying this registry file and
rebooting allowed me to login from NT.
Rick
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Greg Tabaka <[EMAIL PROTECTED]>
Subject: Recognizing and configuring NICs on a crossover ``notwork''
Date: Fri, 16 Jul 1999 21:51:34 GMT
Hi,
After a system crash (Windows 98, relax) I have to move my OS install
files from one computer to another. The recieving one is Redhat Linux
5.2 running kernel 2.0.36 (the crash screwed my new kernel and I have to
boot to this old one which remarkably works) and the sender is Redhat
5.1 with kernel 2.0.34. One is an i486 and the other is a K6 which are
both Intel compatible.
Anyhow, I bought a patch cord, cut it and soldered it to be a cross-
over cord to eliminate a hub. I had two of the same NICs (compatible
with NE2000) lying around so I threw one in each computer. I ran the
DOS diganostics on both and it said my cable was fine. (Unbelievable
when you considered I soldered the wires with a 12-inch copper pipe with
a wet rag wrapped around one end and with the other end sitting on the
stove.) I configured both NICs to use I/O base 0x300 and IRQ 3 since
they were free on both machines.
But I don't know how to set this up in Linux. I read the Ethernet
howto and that didn't help and the Networking FAQ is a step beyond
where I am now.
(The K6/Redhat 5.2 machine can dial the Internet but the 486/RH5.1
machine cannot if this is relevant. The cards are 16-bit ISA boards
from Maxtech [www.maxcorp.com] and their model numbers are NXP-16BT.
I am using 10BaseT. The boards are plug and play but I disabled PnP on
both in favor of my I/O and IRQ settings. I have enabled Full-Duplex
transmission because I doubt that between 2 computers and 4 feet of
wire that I will get many collisions.)
How do I get Linux to recognize these cards?
(I added this to both my /etc/conf.modules:
alias eth0 ne
options ne io=0x300 irq=3
but I don't know if it worked. I rebooted the machine after
modifing this file; is there some program that needs to be run
to recognize these changes?)
How do I set this up in Network Configurator?
What IP numbers and netmasks should I assign to each of the two?
After this is set up, I should be able to use ftp and telnet to
connect to either machine, correct?
Thank you.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (The Sweep)
Subject: Re: Linux PPP server problems
Date: 16 Jul 1999 22:24:46 GMT
Reply-To: nobody
In article <7mo1um$[EMAIL PROTECTED]>,
kite@NoSpam.% inetport.com (Clifford Kite) writes:
> Tom Griep ([EMAIL PROTECTED]) wrote:
>: Thanks for the help Clifford. I got it setup but I have one
>: problem. After I dial in and hang up the phone, the modem
>: in the server doesn't hang up and the ppp daemon is still running.
>: If I kill the pppd daemon, the modem hangs up and I can dialin again.
>
> A guess is that you need the pppd -detach option. Here is a link that
> you might want to check out:
>
> http://www.swcp.com/~jgentry/pers.html
>
> --
> Clifford Kite <kite@inet%port.com> Not a guru. (tm)
> /* Editing with vi is a lot better than using a huge swiss army knife.
> Use +} to wrap paragraphs in vi. */
Thanks, thats what did it. I also had to set the modem to
reset on DTR to make it work.
Sweep
=============================================================
Remove you know what for my Email address.
E-Mail: [EMAIL PROTECTED]
------------------------------
** 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
******************************
