Linux-Networking Digest #278, Volume #11 Tue, 25 May 99 12:13:35 EDT
Contents:
Can't send mail across domains (Ken Horton)
opening socket in a device ("Andreas Moroder")
ppp script won't recognize BUSY string ([EMAIL PROTECTED])
Re: NFS mount problem. (Thierry Sengstag)
Newbie - Sendmail configuration (Ken Horton)
Re: remote root telnet in RedHat 6.0 (Mihaly Gyulai)
Re: Help in setting up IP masquerading ("Loki")
Re: Wingate and Linux ("Rinaldi J. Montessi")
Re: IP Masquerade/Routing ("George Georgakis")
Re: Linux and LLC communications... ([EMAIL PROTECTED])
Re: Redirection using ipchains? (bill davidsen)
Re: SAMBA Success Story - repost (Andrew Williams)
Dual T1's? ("Gary Rule")
Re: PCMCIA and ether card initialization ("Tim Underwood")
Re: 3com network problem- stumped ("Chris Thornburg")
help with novell (Charles Hornat)
----------------------------------------------------------------------------
Date: Tue, 25 May 1999 08:22:15 -0700
From: Ken Horton <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Can't send mail across domains
Hello,
I'm a newbie, so please bear with me.
I have been attempting to send mail to a Linux server from a Novell
server. Both are accessed by Win clients. Everything is working fine
with the exception of sending out from Novell to Linux.
I'm able to retrieve mail to the Novell network, so I assume the POP3
server is working. My guess is that the SMTP (sendmail) is not
configured properly. It may even be the link to the ISP (Linux has a
dedicated connection).
Everything else is working which is what is so puzzling. Please help...
Kurt White
Horton Engineering
-----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
http://www.newsfeeds.com The Largest Usenet Servers in the World!
======== Over 73,000 Newsgroups = Including Dedicated Binaries Servers =======
------------------------------
From: "Andreas Moroder" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.system,comp.os.linux.development,it.comp.linux.development,linux.dev.kernel
Subject: opening socket in a device
Date: 25 May 1999 11:50:13 GMT
Hi,
we have big problems with lpd and out HP JetEX print servers that don't
comunicate a standard way with linux.
The only solution seems to me to write a device driver that talks to the
JetEX.
Now I have two parts ready:
1. A skeleton device driver that prints only messages to the console whe in
open and close teh device.
2. The real program that comunicates with the HP JetEX.
This prog uses the socket and the connect funtion to connect to the
printserver.
If I join this to programs and make insmod the loader tells me it does not
know where to find "socket" and "connect".
What kernel functions can I use instead of "socket" and "connect" ?
"The Linux Programmer's Guide" in the LDP tells me that both are part of
the syscalls available. Are they or aren't they.
I you can help me please send the answer to
amoroder@[nospam]se-nord.provinz.bz.it
Thank you
Andreas Moroder
------------------------------
From: [EMAIL PROTECTED]
Subject: ppp script won't recognize BUSY string
Date: Tue, 25 May 1999 08:16:28 GMT
Hi, all.
Well, the title says it all...
I dial to my school's modem pool which has way less modems
than it should. And I always get a busy signal for the first
or second attempt to dial in.
The thing is my ppp script does not recognize the busy signal.
I thought it might be a problem with something else, like my modem
setting, but when I run minicom it terminates dialing right away
with BUSY string.
Here's how it looks like (well, essentially identical to ppp-howto)
...
exec /usr/sbin/chat -v
TIMEOUT 3 \
ABORT '\nBUSY\r' \
ABORT '\nNO ANSWER\r' \
'' \rAT \
'OK-+++\c-OK' ATH0 \
TIMEOUT 40 \
OK ATDT$TELEPHONE \
...
Other than that, the script runs nice and smooth. :-)
... and also...
Is there any ppp util program that does auto-dialing when the line
is busy until it goes through? I guess I could write a perl script
or something... but I guess I shouldn't bother if someone else
already did the same thing...
Thanks for your help!
Regard,
DAE
p.s. I broke my right index finger yesterday and, man... typing
without it is just a torture!
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
From: Thierry Sengstag <[EMAIL PROTECTED]>
Crossposted-To: comp.sys.hp.hpux,comp.os.linux.setup
Subject: Re: NFS mount problem.
Date: Mon, 24 May 1999 21:57:42 +0200
Reply-To: [EMAIL PROTECTED]
Hi Jim,
Thanks for your answer.
> man route is your friend.
Tough friend, actually... :-)
> You need a route between the two subnets. ie:
>
> on helena - route add -net 195.112.nn.0 dev eth0
> and a return route on homer - route add -net 128.178.nn.0 eth0
I looked at the man page. Actually I guess more than I understand :
eth0 means the ethernet card, but I'm trying to mount the disk
through a phone line. (Yes I know, not very reliable, but for some
reason, I need to do that.) Here the device is ppp0, but to this
argument
my linux box complained "SIOCADDRT: Invalid argument"...
Another point : traceroute mentions 12 nodes between the mount points,
as far as I understand 'route', it is used to build hard paths between
two computers (in other words to tell explicitely through which
computers
the packets have to travel).
Thus the questions :
Do I understand correctly ?
Is it possible (in principle) to do what I want to do ?
How do I specify all the nodes of the path on a single route command ?
(I tried to put some of the addresses listed but traceroute as a
gateway
to the route command, but could not fall on the right one...)
Thanks for any advice,
Thierry
------------------------------
Date: Tue, 25 May 1999 09:40:48 -0700
From: Ken Horton <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Newbie - Sendmail configuration
Hello,
I'm trying to use RH Linux 5.2 with a dedicated internet connection to
relay e-mail from a Novell server to the rest of the known world. Novell
clients can get e-mail just fine (I assume that POP3 is working). In
fact, everything works great with the exception of sending mail to the
Linux box from Novell Netware.
I think SMTP (sendmail) may need to be configured either to accept mail
from the Novell server, or pass it on to the ISP's mail server.
Any ideas?
Kurt White
Horton Engineering
-----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
http://www.newsfeeds.com The Largest Usenet Servers in the World!
======== Over 73,000 Newsgroups = Including Dedicated Binaries Servers =======
------------------------------
From: Mihaly Gyulai <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: remote root telnet in RedHat 6.0
Date: Tue, 25 May 1999 08:53:21 GMT
In article <XJ%13.884$[EMAIL PROTECTED]>,
"Jim Orfanakos" <[EMAIL PROTECTED]> wrote:
> I just installed RedHat 6.0, modified the same files but I still cannot
> telnet in as root (other users can telnet it).
I don't know what file is relevant for allowing this, but...
Allowing a root telnet is a big security risk. If you really need
such function, you have to search for 'ssh' package, which is
recommended for doing this...
--
Mihaly Gyulai
http://www.freeyellow.com/members5/gyulai/
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
From: "Loki" <[EMAIL PROTECTED]>
Subject: Re: Help in setting up IP masquerading
Date: Tue, 25 May 1999 09:50:27 -0400
You need to have the ftp module loaded, here is what I use for Redhat 5.1.
Add this entry to the /etc/rc.d/rc.local.
/sbin/depmod -a
/sbin/modprobe ip_masq_ftp
http://ipmasq.home.ml.org/ is one of the best places to find info on
masquerading.
Loki
------------------------------
From: "Rinaldi J. Montessi" <[EMAIL PROTECTED]>
Subject: Re: Wingate and Linux
Date: Mon, 24 May 1999 08:16:00 -0400
Douglas Dunyan opined
> Hey -
> I am trying to configure a linux client to go through Wingate for
> cable modem services.
>
> I have the network card installed okay.....I think I have tried
> every comibination of routes, hosts, dns, etc....
>
> Here is the scenario:
>
> Wingate Machine:
> nic#1 ip 192.168.0.1
> mask 255.255.255.0
> gateway 24.1.218.77
> dns entries: 24.1.240.33
> 24.1.240.34
>
> nic#2 ip 24.1.218.77
> mask 255.255.252.0
> gateway 24.1.216.1
> dns entries: 24.1.240.33
> 24.1.240.34
>
> Linux box:
> ip 192.168.0.2
> I can ping 192.168.0.1
> 24.1.218.77
> when I traceroute, the first response is Unknown host
> the second shows the ip but goes no further (wingate cant handle a
> traceroute through)
> On my linux box, should I enter info in 'Routes to other Networks' ,
> 'Routes to other hosts' or both? What should the values be?
>
> I can ftp to wingate, and from there, ftp out. That is the total of my
> success.
>
> Any help is definately appreciated.
>
> Doug
Can't be done. Use ipfwadm (2.0) or ipchains (2.+) depending on your
kernel. Configure your win machine as if it were connected directly to the
net except for DUN. If your dual booting the linux box turn off DNS when
in windows. Wingate doesn't care which box has DNS enabled, and it has to
be enabled on the client for ip masq'ing to work.
--
Rinaldi -
Sometimes a cigar is merely a cigar. Sigmund Freud
------------------------------
From: "George Georgakis" <hotmail.com@hurro>
Subject: Re: IP Masquerade/Routing
Date: Mon, 24 May 1999 12:32:52 GMT
If I understand you correctly:
Yes, you can force an IP address on a dial-in connection simply by creating
and editing an options file for the device(s) relating to the modem(s).
I allow dial-in access via modems connected to a Cyclom multiport card. The
device name for the first modem is /dev/ttyC0. I force dial-in connections
to use an address in the private range, in my case, 192.168.1.* The
address 192.168.1.1 is the server's local IP address. Wanting to force the
IP address 192.168.1.100 to the dial-in for /dev/ttyC0, I create an
/etc/ppp/options.ttyC0 with the following line:
192.168.1.1:192.168.1.100
I could also just use:
:192.168.1.100
In your case, you can force the dial-in to take your preferred IP address
with the following line in /etc/ppp/options.tty**
:172.16.254.3
OTOH, if I completely misunderstood your post, please ignore me :)
George
===========================================================================
I never reply by email as a) I don't give out my real email address freely,
and b) it stops other NG users from reading the solutions to problems
I can be contacted thru [EMAIL PROTECTED]
===========================================================================
Curt <[EMAIL PROTECTED]> wrote in article
<RM923.245$[EMAIL PROTECTED]>...
> You might contact the owner of the IP range to write a threatening
letter.
> It appears to be University of Toronto, via whois on www.arin.net
>
> Athol Marshall <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > I have set-up Linux in order to replace a Cisco router. (I need
> > IP masquerading to multiple locations, and Cisco only supports it for 2
> > locations). Other routers only support up to 8 and they're overpriced.
> >
> > I've got IP masquerading working over diald/PPP/ISDN to the Internet.
> >
> > I'm now about to set-up the routes for the various other networks I
need
> > to access.
> >
> > Here's the problem.
> >
> > My clients are very badly behaved and have network addresses in use
> > which may be on the Internet somewhere. ie they're not using private
> > IP addresses.
> >
> > e.g. one of them uses 128.100.0.0, which for all I know may exist
> > somewhere else.
> >
> > If I route 128.100.0.0 via PPP to my client, and I then need to access
> > the "real" 128.100.0.0 for some reason, it clearly won't work.
> >
> > Is there anyway I can alias my client's network address so that I could
> > pretend it was say 172.16.0.0 and have the IP addresses translated
> > before passing packets to PPP?
> >
> > e.g. a host at the other end has IP address 128.100.254.3
> >
> > I would define it in my hosts file or NIS as 172.16.254.3
> >
> > I would define a route to 172.16.0.0 via a diald/PPP interface.
> >
> > The kernel would translate the 172.16 into 128.100 for me, so the
router
> > at the other end would forward the packet correctly.
> >
> > If this is not currently possible, can anyone point me to the
> > appropriate kernel modules so I can try and code it myself?
> >
> > I realise that the proper solution would be to use private IP
addresses,
> > but this is something I have no control over.
> >
> > thanks for any tips/advice,
> >
> > --
> > ======================================================================
> > Athol Marshall, Superior Programming Power Ltd
> > [EMAIL PROTECTED]
> > ======================================================================
> >
> >
> >
>
>
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Linux and LLC communications...
Date: Tue, 25 May 1999 14:08:19 GMT
In article
<[EMAIL PROTECTED]>,
Bernd Huebenett
<[EMAIL PROTECTED]> wrote:
> Hello Andrew,
>
> as far as i know there is no way to
establish a LLC-connection to a host
with linux.
> Maybe you can use tcp/ip instead.
>
> Bye,
> Bernd
>
> Andrew T LeCren wrote:
>
> > Hello-
> > I am currently trying to
implement a program which requires
LLC-layer communications
> > with an external device.
Specifically, I need to be able to
communicate using SAP 4 protocol
> > over a token ring network. The
program is written in C on a Redhat
5.2 machine running
> > kernel 2.0.36.
> >
> > What I need to know is if there
is a way to do LLC-layer, SAP4 level
communications under
> > linux? If so, is this acheivable
via tcp/ip sockets, or is there
another C api/library functions that
can be utilized?
> > If any of you know, or can point
me to resources that will allow me to
find out, I would greatly appreciate
it.
> >
> > Thanks.
> >
> > --
> > Andrew LeCren | "You may
be right, it's all just a waste of
time.
> > XC Development | I guess
that's just a chance I'm prepared to
take
> > Motorola - Ft. Worth | A
danger I'm prepared to face. Cut to
the Chase.
> > [EMAIL PROTECTED] |
- "Cut to the Chase"
>
>
have a look at www.linux-sna.org
maybe it helps you
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
From: [EMAIL PROTECTED] (bill davidsen)
Subject: Re: Redirection using ipchains?
Date: 24 May 1999 22:11:18 GMT
In article <7fY13.15534$[EMAIL PROTECTED]>,
Francois Magnan <[EMAIL PROTECTED]> wrote:
| What you guys are looking for is not done using ipchains. You need the
| ipmasqadm tool to do this. This is called port forwarding.
|
| See:
|
| juanjox.linuxhq.com
My DNS dosn't know about this machine.
--
bill davidsen <[EMAIL PROTECTED]> CTO, TMR Associates, Inc
One common problem is mistyping an email address and creating another
valid, though unintended, recipient. Always check the recipient's
address carefully when sending personal information, such as credit
card numbers, death threats or offers of sexual services.
------------------------------
From: Andrew Williams <[EMAIL PROTECTED]>
Subject: Re: SAMBA Success Story - repost
Date: Tue, 25 May 1999 12:27:54 +0200
Reply-To: [EMAIL PROTECTED]
There are two newsgroups more or less dedicated to Samba -
news:comp.protocols.smb and news:linux.samba - they are usually very
helpful.
W.A. Scheer wrote:
> I've noticed a lot of "Can't Get SAMBA to work" posts in this group. I'm
> reposting the following in hopes that it may help SAMBA newbies like
> myself:
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++
> Well, I'm not really sure why things weren't working in the first place
> but
> here's what I did to get things working:
>
> 1) UNINSTALLED Samba from my RedHat Linux 6.0 'workstation' using RPM.
>
> 2) REINSTALLED Samba ... using RPM.
>
> 3) CHECKED my /etc/inetd.conf file to be certain that this line existed:
>
> swat stream tcp nowait.400 root
> /usr/sbin/swat
> swat
>
> This enables using the 'SWAT' (Samba Web Admin Tool) to configure
> SAMBA.
>
> 4) OPENED SWAT in Netscape Navigator by going to:
>
> http://192.168.0.5:901
>
> Note that '192.168.0.5' is the IP address of my machine. The
> ':901'
> part tells Navigator to talk to port #901
>
> 5) CONFIGURED my 'Global' settings as follows (make sure to 'commit
> changes'):
>
> workgroup = MYGROUP
> netbios name = LINUX
> security = user
> encrypt passwords = yes
>
> 6) Created a directory called /export/share1. You can do this either
> the
> GNOME File Manager tool or
> with the mkdir command.
>
> 7) USED the 'Shares' section in SWAT to create a new share called
> 'share1'
> and configured it as follows
> (remember to 'commit changes'):
>
> path = /export/share1
> read only = no
> guest ok = no
> browseable = yes
> available = yes
>
> 8) I then went to the 'Status' page of SWAT and started the smbd and
> nmbd
> services.
>
> Magically, my new 'share1' now works, and I can access it from my Win98
> workstation!!! Sorry for a long post, but I wanted to be as helpful as
> possible to others in this 'group having problems with SAMBA as well.
>
> W.A. Scheer
--
Mielipiteet omiani - Opinions personal, facts suspect, especially on my
http://www.germanynet.de/teilnehmer/101/69082/samba.html
Simple Samba Solutions web page. ICQ 1722461
__________________________________________________________
| Fight Spam! Join EuroCAUCE: http://www.euro.cauce.org/ |
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
------------------------------
From: "Gary Rule" <[EMAIL PROTECTED]>
Subject: Dual T1's?
Date: Tue, 25 May 1999 14:49:01 GMT
We have a 24/7 Internet business and our ISP has a tendancy to go down once
every month or two. We would like to purchase a second T1 as a backup. Can
anybody give me any information or point me to some, about doing this? I
know that we would have completely different IP numbers, I would like to be
able to direct web traffic to that second set of numbers in case the first
line went down...or is there an easier way?
Thanks in advance,
Gary R
------------------------------
From: "Tim Underwood" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.portable
Subject: Re: PCMCIA and ether card initialization
Date: Mon, 24 May 1999 21:51:13 -0500
Yes, it was in 5.2 as well. You get a message that says "NE2000 card
initialization delayed", because it knows that it will be initialized later,
after the card services have been installed.
Has something changed in this behavior in 6.0? I still get an
"initialization delayed" message in 6.0.
[EMAIL PROTECTED] wrote in message
<[EMAIL PROTECTED]>...
>>
>> The problem is that the card is not seen when linux boots (even after the
>> "delay" startup). But all I have to do is run linuxconf, and have it
>> restart the network, and PRESTO! The card is now there and initialized.
>>
>> Why isn't it initializing properly?
>>
>>
>
>Check the start order of the init scripts. By default the network is
>set up to start before card services.
>
>
>rick
>
------------------------------
From: "Chris Thornburg" <[EMAIL PROTECTED]>
Subject: Re: 3com network problem- stumped
Date: Tue, 25 May 1999 11:52:27 -0400
Yeah, I had the problem in windows.. It was quite easy to fix. My problem
was someone set it to use Rj-45 instead of Coax or autosence. So it was
trying to ping the Rj-45 which had nothing hooked up to it. Thats why I
could ping myself but no one else. Once I changed it to AutoSence it picked
up the Coax just fine.
3C5X9CFG.EXE is proggie i do believe
------------------------------
From: Charles Hornat <[EMAIL PROTECTED]>
Subject: help with novell
Date: Tue, 25 May 1999 12:05:00 -0400
I need to connect my linux redhat 6 to our novel 4 network. How do I do
this? M network card is working and I am connected to the internet. So
IP is working and the card is fine. But where is the network
management?
Thanks
------------------------------
** 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
******************************
