Linux-Networking Digest #639, Volume #11 Wed, 23 Jun 99 18:13:53 EDT
Contents:
Xwin-32 connections behind a masq machine (Steve Sobka)
Re: Realtek RTL8019 network card ??? (Edmondo)
Re: POP Installation (Marc Mutz)
Re: configuration of eth0 & ifconfig (Andrei Bergners)
Samba Configuration ("Terence Parker")
Re: Leafnode problems (Marc Mutz)
Re: Setting up several ISP for PPP connections (Marc Mutz)
Re: NFS write from Linux client stalls on RH,Debian, not on Slackware (Peter
Mutsaers)
Re: Swap over NFS (Marc Mutz)
proxy arp question (Guillaum Dallaire)
Re: no route to host behind linux router. ("David Means")
Linux+Windoze+Mac How To? (Alex Lam)
Re: NFS with Redhat 6 server and (A E Lawrence)
Re: How to??? - ADSL w/SuSE 6.1 (Alex Lam)
Re: Multiple Linux Mail Servers ("Curt")
Re: mgetty for dial-in blocks outgoing traffic ([EMAIL PROTECTED])
Re: NFS with Redhat 6 server and (A E Lawrence)
Re: About secondary name server (Mike Selders)
Re: starting ppp from user accounts (Clifford Kite)
----------------------------------------------------------------------------
From: Steve Sobka <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,redhat.x.general
Subject: Xwin-32 connections behind a masq machine
Date: 23 Jun 1999 20:30:55 GMT
I have been able to get Xwin-32 (http://www.startnet.com) to work without
any problems on my internal IPMasqed LAN. But when I try and connect to
the server from outside my lan, I get no response?
Do I need to add somthing special to any files to allow me to use it from
outside my 192.168.0.x network?
My linux box has a static ip (206.xxx.xxx.xxx) which I can telnet into
just fine...I also have installed SSH, but this problem started before the
install of SSH...
I try to connect to the EXTERNAL static IP address, but I get nothing?
Any help would be great...
.Steve
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: Edmondo <[EMAIL PROTECTED]>
Subject: Re: Realtek RTL8019 network card ???
Date: Fri, 18 Jun 1999 10:25:31 +0200
Reply-To: [EMAIL PROTECTED]
Dmitry Koshkin wrote:
> I am wondering if I can install the network card with RTL8019 chip on
> Redhat 6.0.
> What driver should I use for this.
I think it is a NE2000 compatible card. Compile the kernel with the
NE2000 network card (select it under ISA or PCI network adapter. If
I right remember it is an ISA card).
ciao
edmondo
------------------------------
Date: Wed, 23 Jun 1999 22:07:37 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: POP Installation
Brian E. Parker wrote:
>
> Hello Linux people,
>
[snip]
> What I want to do is to allow access via POP mail. I know exactly
> nothing on the subject, so I don't know if POP-mail comes with Redhat,
> Sendmail, etc,. nor do I know how to activate it if it does exist already on
> the box.
>
> What do I need to do to get started on running POP mail? I don't need
> anything fancy at all, so I don't want to install some 3rd party app that
> does a bunch of things I don't need. We just want very basic POP mail
> services.
>
It should be just a matter of installing the pop daemon, it works
independently of sendmail and should not require any additional changes
to the existing system. It should listen to name popd, but I guess there
are several other popd's out there, just like there is sendmail, smail,
qmail,...
If your installation tool supports it (e.g. debian's does), you can
search the list of available packages for 'pop'. From there on it's just
a matter of taste...
Marc
------------------------------
From: Andrei Bergners <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,alt.linux,comp,linux.redhat,linux.redhat.misc
Subject: Re: configuration of eth0 & ifconfig
Date: Wed, 23 Jun 1999 20:17:11 GMT
Try:
ifconfig <interface> <ip address>
ie., ifconfig etho 192.168.0.2
take care,
Andrei
>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<
On 1/26/99, 11:57:13 PM, "John K" <[EMAIL PROTECTED]> wrote regarding Re: =
configuration of eth0 & ifconfig:
> Thanks for your info.
> Yes, I got this far. The problem is the Linux machine. The Ethernet=20
card is
> not getting a IP number. When I tested the link durring the jumper
> installation with both computers connected under DOS, the link worked.=
> Therefore, the problem is not cables or conflics. The problem is that =
Linux
> is not assigning ant IP number to eth0.
> I made this conclusion because when I try to ping the Linux computer=20
to his
> own IP (192.168.0.2) Host is unreachable.
> So If It can't ping its own IP, Linux is the problem. Any idea how to=
> assign eth0 a IP?
> PS. Wingate provides DCHP.
> >6. If U've gotten this far and ur still saying "this aint helping=20
any" then
> >ur linux machine is the problem... ur internal ethernet card on linux=
=20
isnt
> >getting a ip-number... isnt installed correctly (but this should of=20
shown
> in
> >step 3) or something i dont know about.
------------------------------
From: "Terence Parker" <[EMAIL PROTECTED]>
Subject: Samba Configuration
Date: Wed, 23 Jun 1999 09:33:11 +0800
Firstly - where is the Samba Configuration on Caldera OpenLinux 2.2? Anyone
know? This is not important, but I just started playing with Caldera on my
other machine - and everything is different to my existing Red Hat System.
Secondly, does anyone know how I can (or if I can) restrict IP addresses
using Samba? I know there is that setting near the top of the SMB.CONF file
which allows me to restrict the IP addresses that can access the Samba
Server - but is there a way in which I can set IP restrictions to specific
shared resources? Looking through my SMB.CONF file I only see examples of
restrictions by username and group - but not by IP address.
Thanks in advance for any help,
Terence Parker.
------------------------------
Date: Tue, 22 Jun 1999 22:18:49 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Subject: Re: Leafnode problems
Hi out there!
I finally found out what went wrong: Whenever 'fetch' is running,
leafnode refuses to deliver messages. Once fetch ended, it all worked
beautifully!
Thank goddness!
Marc
------------------------------
Date: Wed, 23 Jun 1999 14:23:11 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Subject: Re: Setting up several ISP for PPP connections
"R. Denoire" wrote:
>
> I am using SuSE 6.0 with updated Kernel 2.2.5 here.
>
> My questions relates the problem that when setting up a ppp connection
> to an Internet Service Provider, the IP address of its DNS server
> should be entered in the file /etc/rc.config. So when connecting to
> the ISP, usually the new connection over ppp is set as defaultroute
> and all name resolutions use the IP addresses given in /etc/rc.config.
>
[snip]
This is no limitation, because you can reach any public DNS server
through any ISP. If you have several ISP, chances are that not all of
them block requests from other clients than the currently logged in
ones. So choosing one of those will only cause DNS lookups to be a
fraction slower when using the 'wrong' ISP.
btw: Setting the DNS server IP's in rc.config does not suffice. You have
to run the config program in order to make sure every app knows them.
Also, the init scripts will or will not be restarted, and even if they
are, they are not by simply changing rc.config.
Marc
------------------------------
From: Peter Mutsaers <[EMAIL PROTECTED]>
Subject: Re: NFS write from Linux client stalls on RH,Debian, not on Slackware
Date: 23 Jun 1999 22:43:59 +0200
>> "j" == joechiu <[EMAIL PROTECTED]> writes:
>> When I mount an IRIX or Solaris NFS server on my Linux box and try to
>> write something larger than 64kb, the writing process stalls and the
>> write never finishes. (reading from a NFS server is OK)
>>
>> This happens with kernels 2.2.5, 2.2.7 and 2.2.10 from RH6 of
>> Debian-current on a Compaq desktop with a ThunderLAN 10/100 network
>> card. FTP-ing large files causes no problems however.
j> Might it be that NFS on the Linux system uses UDP only, and that
j> your client is on a 100 Mbps connection, while the server is on
j> a 10 Mbps connection? (Or perhaps on a saturated 100 Mbps connection)
Indeed it is. Is this a known problem with Linux? Would it be solved
by using an older (more stable) kernel?
I already used 1024 bytes, even tried 128 bytes read/write size.
The servers alas don't support NFS over TCP.
Btw: A FreeBSD NFS client works fine in the same situation (using NFS
version 3 instead of 2).
I guess that for the time being I'll stick with FreeBSD.
Thanks,
--
Peter Mutsaers | Abcoude (Utrecht), | Trust me, I know
[EMAIL PROTECTED] | the Netherlands | what I'm doing.
===============+=====================+==================
Powered by FreeBSD (-current). See http://www.freebsd.org
------------------------------
Date: Wed, 23 Jun 1999 23:17:02 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Subject: Re: Swap over NFS
Andreas Jaehnigen wrote:
>
> Hi,
>
> is everyone out there running a diskless client which swaps over NFS?
> The filesystems are exported from a Sun running Solaris 2.6, on the client
> side runs kernel 2.2.9 (2.2.10 compiling NOW. *grin*).
>
> Every document about diskless clients I read so far doesn't mention this
> topic.... (?!)
>
Simply because it would be tooooo slow.
Marc
------------------------------
From: Guillaum Dallaire <[EMAIL PROTECTED]>
Subject: proxy arp question
Date: Wed, 23 Jun 1999 17:18:08 -0400
Reply-To: [EMAIL PROTECTED]
Hi,
when I add an arp entry with :
arp -Ds x.y.z.w eth0 pub
The others hosts in the lan doesn't "see" the x.y.z.w IP,
I have to add the arp entry on each host manually to make my setup
working (NAT).
It's normal ?
------------------------------
From: "David Means" <[EMAIL PROTECTED]>
Subject: Re: no route to host behind linux router.
Date: 23 Jun 1999 21:19:32 GMT
If you can connect to an FTP site using a non-standard port, but cannot
get "List" or "Get" to work, the problem is very likely that the secondary
(data) connection is being denied. You have two options:
1. Permit (outside) hosts to initiate a TCP connection on those (also
non-standard) ports. This is a security hole that one could navigate a
whale through with just a little thought, but you might want to do it
anyway, just for test.
2. Use the "proxy" command (prefix), so that all connections are
initiated from within the firewall perimeter. This still might fail either
because
the firewall rules do not permit the connect packet, or because the remote
FTPD does not implement the PASV command.
Sleinfeld <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hey.
>
> This is from windows machines behind a linux router :
>
> I'm trying to connect to ftp's with other ports than port 21, and when
> it comes to list, it says : no route to host
>
> I have a ftp site at port 25, one at port 32 and one at port 4500 that
> I want to connect to...
>
> How do I set these ports open ?
>
>
> I get port 21 ftp, web, irc, ICQ, news, mail to work though...
>
> thnx..
>
> Leif Ringstad
> ------
> [EMAIL PROTECTED]
------------------------------
From: Alex Lam <[EMAIL PROTECTED]>
Subject: Linux+Windoze+Mac How To?
Date: Wed, 23 Jun 1999 09:48:44 -0700
Hi,
I need to set up a Linux box to share files between other
Linux/FreeBSD, Windoze and Mac together. I got the Linux/FreeBSD/Windoze
part done through Samba and ipchains. Now how do I add Mac and PowerPC
to the chain?
Thanks.
Alex Lam.
--
*** *** *** *** *** *** ***
Remove all the upper case Xs from my email address if reply by e mail.
**************************************************
*If you receive any spam from my domain name. It's forged.
I DO NOT send spam e mail. But I've found out that my
domain has been forged many times.
**************************************************
------------------------------
From: A E Lawrence <[EMAIL PROTECTED]>
Subject: Re: NFS with Redhat 6 server and
Date: Wed, 23 Jun 1999 22:22:38 +0100
In article
<[EMAIL PROTECTED]>,
"Hebert, Greg (EXCHANGE:KWAY:6C26)"
<[EMAIL PROTECTED]> wrote:
> Jon Paterson wrote:
> >
> > I am having a frustrating problem with a Redhat 6 server that was
> > previously a redhat 5.2 server rebuilt.
> >
> > I have a Linux client (redhat 6.0 also) that is trying to connect to the
> > server and is always getting the "permission denied" message.
> > > > There is nothing wrong with the exports file, I have even deleted it and
> > done the configuration through Linuxconf, and the same error exists.
> >
> > I think that it may have something to do with Knfsd, but am not sure
> > where to look.
> >
> > can anyone help or point me in the right direction?
I haven't seen an answer to this question, but many posts with the same
problem.
There are two basic problems (well one really: Redhat's failure to draw
attention to these changes, unless I have missed it...)
1) knfs requires an explicit mount for each filesystem.
2) The default permissions for /etc/exports have changed. Thus anyone
upgrading from a working nfs Redhat 5.2 to Redhat 6.0 *will* experience
problems. Easy to fix when you know what is wrong. Frustrating and time
wasting otherwise!
Ok. Suppose that you have
/
/disk2
/disk3
as file systems on a Redhat 6.0 box. That is 3 physical discs with, say,
disk1 as mounted as root.
Whereas under the userland nfsd of Redhat6.0, /etc/exports needed a
single entry to permit the whole filesystem including "submounts" to
be exported, under knfs, you need a separate entry for each individual
filesystem. That is 3 entries in the example above: /,/disk2,/disk3.
Before, the default permissions included rw, the default now seems to be
ro.
And if you wish to mount from another box, you now need several
invocations of mount:
mount redhat6.0_name:/ /local/mount/
mount redhat6.0_name:/disk2 /local/mount/disk2
....
Hope this helps.
ael
--
Dr A E Lawrence (from home)
------------------------------
From: Alex Lam <[EMAIL PROTECTED]>
Subject: Re: How to??? - ADSL w/SuSE 6.1
Date: Wed, 23 Jun 1999 09:57:43 -0700
Spicoli wrote:
>
> On Tue, 22 Jun 1999 07:59:54 -0700, Alex Lam <[EMAIL PROTECTED]>
> wrote:
>
> - snip -
>
> >Just login as root, then use YaST to config the network. Save the
> >changes, log out from root. Piece of cake. Should takes no more than 2-3
> >minutes.
> >
> >Forget the SuSE manuel. It tells you absolute nothing, other than the
> >initial installation.
> >
> >Alex Lam.
> >
>
> - snip -
>
> Thanks for the tip Alex! It took all of 3 minutes and now my ADSL is
> up and running!
>
Have fun! :)
Alex Lam.
> Spicoli
>
> spicoli at aracnet dot com
--
*** *** *** *** *** *** ***
Remove all the upper case Xs from my email address if reply by e mail.
**************************************************
*If you receive any spam from my domain name. It's forged.
I DO NOT send spam e mail. But I've found out that my
domain has been forged many times.
**************************************************
------------------------------
Reply-To: "Curt" <[EMAIL PROTECTED]>
From: "Curt" <[EMAIL PROTECTED]>
Subject: Re: Multiple Linux Mail Servers
Date: Wed, 23 Jun 1999 11:43:36 -0500
You might want to use a combination of qmail and fetchmail or procmail to do
the job.
I think you'll find it simpler to deal with than sendmail. www.qmail.org
General HOWTO:
http://metalab.unc.edu/LDP/HOWTO/mini/Offline-Mailing.html
Neil A. Miller <[EMAIL PROTECTED]> wrote in message
news:7kqm7i$ffj$[EMAIL PROTECTED]...
> Hi
>
> I want to set up internal/external e-mail in a company at low cost.
>
> This is my current train of thought.
>
> We are in the process of obtaining a suitable domain name and will want to
> create mail system when we have it. At the moment we are spread over four
> sites and are all interconnected with TCP/IP (A Windows NT domain) using
> ISDN2 and routers. What I have in my mind is to use Linux based mail
servers
> at each site of which one (at head office) would act as the gateway to the
> Internet. This gateway would dial an ISP every 15 minutes or so during the
> day and upload/download external Internet mail. This gateway would also
> distribute mail to the internal site based mail servers as needed. Each
> desktop user's PC would use Outlook Express (or a similar reader) to
check
> for mail on their local Linux mail server
>
> I am guessing that Linux using 'sendmail' to distribute mail & PPP to talk
> to the ISP will cope with this scenario but I could do with a few pointers
> to 'HOWTOs' etc. to get me underway.
>
> Thanks
>
> Neil
>
>
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: mgetty for dial-in blocks outgoing traffic
Date: Wed, 23 Jun 1999 15:55:13 GMT
pppd creates a lock file in var/lock lock...ttyS3
Which is OK, I guess. But ifup ppp0 has no effect.
***AN IDEA** my modem is class1. I remember this because I couldn't get
it to work with mgetty+sendfax a few weeks ago.
Could this be the reason?
However, I have entered the -D option (data only) just in case.
Michael
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: A E Lawrence <[EMAIL PROTECTED]>
Subject: Re: NFS with Redhat 6 server and
Date: Wed, 23 Jun 1999 22:27:43 +0100
A E Lawrence wrote:
>
> In article
> <[EMAIL PROTECTED]>,
> "Hebert, Greg (EXCHANGE:KWAY:6C26)"
> <[EMAIL PROTECTED]> wrote:
> > Jon Paterson wrote:
> > >
> > > I am having a frustrating problem with a Redhat 6 server that was
> > > previously a redhat 5.2 server rebuilt.
> > >
> > > I have a Linux client (redhat 6.0 also) that is trying to connect to the
> > > server and is always getting the "permission denied" message.
> > > > > There is nothing wrong with the exports file, I have even deleted it and
> > > done the configuration through Linuxconf, and the same error exists.
> > >
> > > I think that it may have something to do with Knfsd, but am not sure
> > > where to look.
> > >
> > > can anyone help or point me in the right direction?
>
> I haven't seen an answer to this question, but many posts with the same
> problem.
>
> There are two basic problems (well one really: Redhat's failure to draw
> attention to these changes, unless I have missed it...)
>
> 1) knfs requires an explicit mount for each filesystem.
> 2) The default permissions for /etc/exports have changed. Thus anyone
> upgrading from a working nfs Redhat 5.2 to Redhat 6.0 *will* experience
> problems. Easy to fix when you know what is wrong. Frustrating and time
> wasting otherwise!
And I forgot to mention the new command exportfs (there is a man page)
which is useful in sorting out what is happening. It seems to be needed
to re-read /etc/exports after editing. The old way of sending SIGHUP to
rpc.mountd doesn't seem to work: indeed it kills rpc.mountd. :-(
Of course the man page for exports is out of date. OK, so there were 4
problems.. :-(
I suppose that I should go and consult the source to see whetehr I got
this stuff right. :-)
ael
--
Dr A E Lawrence (from home)
------------------------------
Date: Wed, 23 Jun 1999 10:42:50 -0600
From: Mike Selders <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: About secondary name server
Lim wrote:
> Hello,
> I have two servers running with bind 8.2 (primary & secondary), I use
> nslookup to check each server is worked. But I still have a problem, when I
> add a new record in a zone file at primary server and update the serial
> number, but secondary server do not update the new zone file.
> Any help would be more than appreciated. Thanks
You must manually add the slave zone entries on your secondary for any new
zones you create. The secondary has no idea that you wish it to be a slave
for a specific zone until you've added that zone to its named.conf file. Once
added, updates to that zone on the primary will cause the secondary to update
itself.
-- Mike Selders, Network Administrator
______________________________________
Clark & Thompson Online Services, Inc.
http://www.ctos.com
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: starting ppp from user accounts
Date: 23 Jun 1999 16:07:12 -0500
paul ([EMAIL PROTECTED]) wrote:
: Hello, I'm running a R.H 5.2 setup with a small network, 1 linux box
: with ppp, ipmasq, samba and 3 win95 boxes. Linux runs the whole system
: perfectly.
: The only prob I'm having is starting ppp from a user's a/c. I've
: created a ppp group, added all the users, setup the permissions
: according to the PPP Howto, (pppd suid etc) but no go. There's no error
: messages or any indication of what's wrong. The only way I can start ppp
: is from root. Needless to say I have no desire to give the other users
: root access to Linux.
Do the scripts have full pathnames for pppd and chat? The sbin
directories may not be on a regular users' path. Indications that this
is a problem might be found in /etc/ppp/connect-errors.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* I gave up on politics when no matter who I voted for, I regretted it.
* -- Pepper...and Salt, WSJ */
------------------------------
** 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
******************************
