Linux-Networking Digest #566, Volume #12 Sun, 12 Sep 99 20:13:33 EDT
Contents:
Re: virtual webhosting ("Andrew Smith")
Re: dialing to office-through lan to internet (Floris van Nierop)
PPPd on a single box (Chris Anderson)
Re: 2nd NIC won't load (David Crooke)
Re: dialing to office-through lan to internet (Floris van Nierop)
PPP Server problem using Redhat 6.0 (Nick)
diald uses obsolete (PF_INET,SOCK_PACKET) ("mikes")
��������� ������ ("Alexandr")
Re: nslookup resolves, ping doesn't (Ludger Solbach)
IP Masquerading problem? (Peter Jun)
Re: IPChains, MASQ and FTP help (Warren Young)
Re: URGENT: Dual NIC Problem (Warren Young)
How to getSMC EZ Card 10 out of PnP mode? ([EMAIL PROTECTED])
Re: Fetchmail: fetching for multiple users in one pass (Frank v Waveren)
is ipfwadm's -W option restrictive? ([EMAIL PROTECTED])
Re: SAMBA slowness ("David B. Putnam")
----------------------------------------------------------------------------
From: "Andrew Smith" <[EMAIL PROTECTED]>
Subject: Re: virtual webhosting
Date: Sun, 12 Sep 1999 23:14:35 +0100
hi
if the two domain addresses point to the same site, then the IP addresses
wil be exactly the same. If you want www.foo.com to point to one website and
www.foo-2.com to point to another site, you'll need to either use different
IP addresses (the easiest way) or different ports. you can configure the
server to display a different site depending on which IP/port is used. i'm
not an expert on this, so i can't really tell you what you need to do to set
this up, although i'm sure there was a HOWTO abbout this...
hope this helps,
Andrew
chris stegmaier <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> If I've got a webhost up behind a firewall, and the webhost
> is masqueraded/proxied, then how can I setup virtual hosting.
>
> I still sort of don't understand the virtual hosting thing...
>
> Does is look like this?
>
> www.foo.com 222.222.222.222
> www.foo-2.com 222.222.222.222:1
>
> I thought I saw something like this is the mini HOWTO
> on virtual hosting, but I'm not sure how it works.
>
> Is it all set up on the one NIC/IP address in the webhost?
> How does that work?
>
> Thanks in advance,
>
> chris
------------------------------
From: [EMAIL PROTECTED] (Floris van Nierop)
Subject: Re: dialing to office-through lan to internet
Date: 12 Sep 1999 22:54:50 GMT
On Sun, 12 Sep 1999 17:36:11 -0400, David Heath <[EMAIL PROTECTED]>
coughed, then solemnly spoke:
>Does anybody know of a way to setup a linux box to accept dial-ups from
>home then connect through the linux box on a lan connected to internet
>via fiber connection. I guess it would be setting up your office to act
>as your own isp.
For the first part of the problem (setting up the Linux box to accept
PPP-dialins), you might want to take a look at mgetty:
http://www.leo.org/~doering/mgetty/
Cheers,
Floris
------------------------------
From: [EMAIL PROTECTED] (Chris Anderson)
Subject: PPPd on a single box
Date: 12 Sep 1999 23:17:26 GMT
Reply-To: [EMAIL PROTECTED]
HAs anyone had a pppd link running across a single machine, on
different interfaces?
Chris
------------------------------
From: David Crooke <[EMAIL PROTECTED]>
Subject: Re: 2nd NIC won't load
Date: Sun, 12 Sep 1999 19:40:26 GMT
Peter Cioe wrote:
>
> I got it to work, not sure if this is the correct way.
>
> I copied the /etc/sysconfig/network-scripts/ifcfg-eth0
> and created ifcfg-eth1 edited it to the 2nd NIC's settings. Now when I
> reboot it loads both eth0 and eth1. I also did the same and eth2
> loaded.
>
> Is this the best way to fix my problem?
> Thanks
> Peter
Sounds like that's the "official" way to do things for your particular
Linux distribution (I presume Red Hat or Mandrake).
Jsut dumping a modprobe command in rc.local would work fine in most
cases, but it isn't exactly as elegant.
--
David Crooke, Austin TX, USA. +1 (512) 656 6102
"Open source software - with no walls and fences, who needs Windows
and Gates?"
------------------------------
From: [EMAIL PROTECTED] (Floris van Nierop)
Subject: Re: dialing to office-through lan to internet
Date: 12 Sep 1999 23:05:51 GMT
On 12 Sep 1999 22:54:50 GMT, Floris van Nierop <[EMAIL PROTECTED]> coughed,
then solemnly spoke:
>On Sun, 12 Sep 1999 17:36:11 -0400, David Heath <[EMAIL PROTECTED]>
>coughed, then solemnly spoke:
>>Does anybody know of a way to setup a linux box to accept dial-ups from
>>home then connect through the linux box on a lan connected to internet
>>via fiber connection. I guess it would be setting up your office to act
>>as your own isp.
>
>For the first part of the problem (setting up the Linux box to accept
>PPP-dialins), you might want to take a look at mgetty:
>http://www.leo.org/~doering/mgetty/
I know posting a follow-up to oneself is an early sign of dementia, but:
More specific information about setting up PPP-links through mgetty can
be found at:
http://members.bellatlantic.net/~mrdennis/mgetty.html
Cheers,
Floris
------------------------------
From: Nick <[EMAIL PROTECTED]>
Subject: PPP Server problem using Redhat 6.0
Date: Sun, 12 Sep 1999 17:32:39 -0700
Reply-To: [EMAIL PROTECTED]
==============BE191B96B38DB28872A34016
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Help. I am trying to set up a PPP server
under redhat 6.0 and having all sorts of problems. I had it
working under 5.2 but after installing 6.0, no go. I have ppp
support in the kernel. I can dial in and get shell access (I'm
using uugetty on the server) and that works fine. I can start pppd
and get a serial connection established. Then I can even ping my
server (from home where the computer I am trying to use to dial in
is) and any other computer on our LAN but anything outside our netmask
is unreachable. I know this sounds like an ip forwarding problem
but if it is I don't know how to enable ip forwarding. There
is no question for it in the kernel config I have (2.2.5 i think) and it
only says to echo "1" to /proc/sys/net/ipv4/ip_forward, which I have
done. I also set ip_forward=yes and ipv4_forward=yes in the /etc/sysconfig/network
file. None of this has helped. I can see everything on my LAN but
nothing outside the LAN's netmask. I don't think it's a resolv.conf
or anything like that problem cause I can't even ping ip addresses
outside the netmask. If anyone has any ideas about solving this please
lemme know or post a reply.
Thanks in advance.
Nick ([EMAIL PROTECTED])
==============BE191B96B38DB28872A34016
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML>
<PRE>Help. I am trying to set up a PPP server
under redhat 6.0 and having all sorts of problems. I had it
working under 5.2 but after installing 6.0, no go. I have ppp
support in the kernel. I can dial in and get shell access (I'm
using uugetty on the server) and that works fine. I can start pppd
and get a serial connection established. Then I can even ping my
server (from home where the computer I am trying to use to dial in
is) and any other computer on our LAN but anything outside our netmask
is unreachable. I know this sounds like an ip forwarding problem
but if it is I don't know how to enable ip forwarding. There
is no question for it in the kernel config I have (2.2.5 i think) and it
only says to echo "1" to /proc/sys/net/ipv4/ip_forward, which I have
done. I also set ip_forward=yes and ipv4_forward=yes in the
/etc/sysconfig/network
file. None of this has helped. I can see everything on my LAN but
nothing outside the LAN's netmask. I don't think it's a resolv.conf
or anything like that problem cause I can't even ping ip addresses
outside the netmask. If anyone has any ideas about solving this please
lemme know or post a reply.</PRE>
Thanks in advance.
<P>Nick ([EMAIL PROTECTED])
<BR> </HTML>
==============BE191B96B38DB28872A34016==
------------------------------
From: "mikes" <[EMAIL PROTECTED]>
Crossposted-To: linux.dev.diald
Subject: diald uses obsolete (PF_INET,SOCK_PACKET)
Date: Sun, 12 Sep 1999 18:35:01 -0500
Reply-To: "mikes" <[EMAIL PROTECTED]>
Though my linux box can dial out and connect to my ISP, I see the following
messages in /var/log/messages:
Sep 12 18:03:07 phoenix kernel: diald uses obsolete (PF_INET,SOCK_PACKET)
What does this mean?
It seems to me that diald (or some other process) is sending an obsolete
packet. Is there a fix for this?
Mike
------------------------------
From: "Alexandr" <[EMAIL PROTECTED]>
Subject: ��������� ������
Date: Mon, 13 Sep 1999 02:25:16 +0400
Subject: AllCommunity.com
I want to let you know about an exciting Internet community in the making.
Please take a few minutes to read below. Email me with any questions you
have or go to http://AllCommunity.com/all.members/Alex15/ to learn more and
become part of what may become a powerful consumer force on the Web.
Facts:
A "few" people/companies are making millions of dollars on the Internet.
AllCommunity.com will share the wealth with you. The Internet changes
everything.
Questions:
How many advertisements do you think you get HIT WITH every time you are on
the Web? (hundreds to thousands)
How much money do the advertisers PAY YOU while you are surfing the Web?
(Nothing, ZERO, unless you sign up with AllCommunity.com.)
Does your on-line community PAY YOU to cruise the globe on the WWW?
Would you like to GET PAID for your time online?
Would you like to GET PAID for friends that you referred to this site while
they are online?
This is a radically simple concept.
AllCommunity.com is an Internet company positioned to share the wealth of
the Web.
Go to http://AllCommunity.com/all.members/Alex15/ to learn more and sign up
today.
http://www.internations.net/ru/alex246
------------------------------
From: Ludger Solbach <[EMAIL PROTECTED]>
Subject: Re: nslookup resolves, ping doesn't
Date: Sun, 12 Sep 1999 23:43:26 +0200
> > Probably /etc/reolv.conf isn't
> > setup to use DNS.
It is. It contains valid settings for "search" and "nameserver". I can
ping almost every site on the internet by its name, but there seem to be
a few exceptions. One is
www.zm.bertelsmann.de
Ping does not work:
> ping www.zm.bertelsmann.de
ping: unknown host: www.zm.bertelsmann.de
Nslookup tells me
Non-authoritative answer:
Name: www.zm.highway.bertelsmann.de
Address: 194.174.252.80
Aliases: www.zm.bertelsmann.de
So, apparently "www.zm.bertelsmann.de" is only an alias for
"www.zm.highway.bertelsmann.de". When I ping the original name directly,
I do get a reply. So, my guess was that there's a problem with names
that are aliases. But that can't be the case either since I can ping
"www.mit.edu" directly, even though that's an alias as well.
Clueless,
Ludger.
------------------------------
From: Peter Jun <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.redhat,comp.os.linux.questions
Subject: IP Masquerading problem?
Date: Sun, 12 Sep 1999 23:42:40 GMT
I have setup three Win95 computer network that is hooked up to a Linux
2.2.12/Redhat 6 box running IP Masquerading.
Win95 COMPUTERS ----- (eth1) Linux (eth0) ----- (cable modem) INTERNET
Everything seems to work (I can surf the internet using the win95
computers), except I've noticed occasionally that eth1 will 'reset'
itself (the light on the hub goes off and then a few seconds later goes
on). This intermittent 'disconnection' causes most of the apps (esp.
telnet) to disconnect.
The linux box is a P100 running Redhat 6.0/2.2.12. eth0 is a NE2000
(realtek) compatible 10Mbit card while eth1 is a DEC tulip 10Mbit card.
Does anyone have a solution for this eth1 resetting? Here is the
ifconfig output. Thanks for any help that you can provide!
eth0 Link encap:Ethernet HWaddr 00:00:B4:52:B0:DA
inet addr:24.114.x.x Bcast:24.114.x.x Mask:255.255.254.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:83450 errors:0 dropped:0 overruns:0 frame:688
TX packets:9326 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
Interrupt:12 Base address:0x6800
eth1 Link encap:Ethernet HWaddr 00:E0:29:36:53:AE
inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:8811 errors:0 dropped:0 overruns:0 frame:0
TX packets:9473 errors:16 dropped:0 overruns:0 carrier:32
collisions:161 txqueuelen:100
Interrupt:10 Base address:0x6c00
------------------------------
From: Warren Young <[EMAIL PROTECTED]>
Subject: Re: IPChains, MASQ and FTP help
Date: Tue, 31 Aug 1999 23:21:44 -0600
Joseph Santaniello wrote:
>
> I've got a machine using a simple ipchains rule to masqurade my internal
> network, and it works fine, except for some ftp clients. The IPchains
> HOWTO says in a general sense how to fix this, but I couldn't figure it
> out.
>
> Here's my ipchains command:
>
> /sbin/ipchains -A forward -j MASQ -s 192.168.15.0/24 -d 0.0.0.0/0
>
> What should I add to make ftp and ftp-data work in passive mode?
It's a client configuration issue. Most browsers always do passive
mode, so you never see the setting. Other FTP clients should offer you
the option to change this. If not, they may only support active
connections.
Alternatively, try the following stuff: http://juanjox.linuxhq.com/
This re-writes passing FTP packets so that active connections work
properly.
For a better answer, tell us which FTP clients you're having trouble
with.
--
= Warren -- See the *ix pages at http://www.cyberport.com/~tangent/ix/
=
= ICBM Address: 36.8274040 N, 108.0204086 W, alt. 1714m
------------------------------
From: Warren Young <[EMAIL PROTECTED]>
Subject: Re: URGENT: Dual NIC Problem
Date: Tue, 31 Aug 1999 23:26:50 -0600
[EMAIL PROTECTED] wrote:
>
> I have a Compaq Server with RedHat Linux 6.0.
>
> this m/c has 2 NICs. Both are Compaq Netelligent (One embedded and
> other an Add-on Card). Both are using the TLAN driver.
>
> The problem is that if I bring down eth1 (Embedded card), I can still
> connect to eth0. BUT if I bring down eth0, it also brings down eth1
> severing all connectivity to this machine!
>
> How do I get around this problem?
This may be a problem with the ifdown scripts for the eth0 interface.
Look in /etc/sysconfig/network-scripts. (You'll need to be able to read
Bourne shell scripts. If you can't it's an excellent time to learn.
B-) )
Good luck,
--
= Warren -- See the *ix pages at http://www.cyberport.com/~tangent/ix/
=
= ICBM Address: 36.8274040 N, 108.0204086 W, alt. 1714m
------------------------------
From: [EMAIL PROTECTED]
Subject: How to getSMC EZ Card 10 out of PnP mode?
Date: Sun, 12 Sep 1999 23:12:40 GMT
Hi all,
I'm trying to find the right utility to get my new SMC EZ Card 10 (PCI
version) out of PnP mode so it will work under Linux AND NT (dual-boot).
According to http://lhd.datapower.com/db/dispreport.cgi?DISP?216, I need
some setup utility to do this -- the one that came with the card doesn't
seem to do it, and I can't find the right one on the SMC site.
Can anyone point me to the right utility?
I've also got a 3Com Etherlink III (ISA version) running on the box
right now, and as far as I know both cards are in PnP mode under NT, so
I guess I will have to disable the PnP on the 3com card too, right?
Does anyone have any pointers on getting either/both of these cards
running under Linux (my ultimate goal is to create a home LAN with my
cable modem)?
Thanks,
-ML
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Frank v Waveren)
Subject: Re: Fetchmail: fetching for multiple users in one pass
Crossposted-To: comp.os.linux
Date: Mon, 13 Sep 1999 00:22:17 GMT
In article <Z8WC3.48955$[EMAIL PROTECTED]>,
"YouDontKnowWho" <[EMAIL PROTECTED]> writes:
> We are currently using fetchmail to retrieve user mail from our ISP
> accounts. Each user has a fetchmail cron that checks their mail and
> puts it locally where it belongs.
>
> Well, right now we don't have that many users on the system, so
> setting this up is not that much work or overhead on the system. But
> that will change soon and I'm dreading the day we have to manage
> setting up fetchmail cron jobs and .fetchmailrc files for every person
> we add. Not to mention the processing overhead of all those fetchmail
> cron jobs running at various times!
>
> How can we get fetchmail to, in one pass, check the e-mail account on
> our ISP for each of our users and locally distribute the e-mail
> appropriately? Can fetchmail be set up to do that? We would like to
> have a single configuration file that tells fetchmail what to check
> and who to give the resulting e-mail to. That way, setting e-mail
> delivery for new users on the system involves just adding a new line
> to the configuration file.
>
> If fetchmail cannot do this, is there anything out there that can do
> this?
I think fetchmail can. Have you tried using fetchmailconf? This makes
fetchmail configuration a lot easier. IIRC, you can enter a remote pop/imap/etc
account, and a local user it's got to be addressed to.
--
Frank v Waveren
[EMAIL PROTECTED]
ICQ# 10074100
------------------------------
From: [EMAIL PROTECTED]
Subject: is ipfwadm's -W option restrictive?
Date: Sun, 12 Sep 1999 23:43:15 GMT
Please help me understand ipfwadm's -W option in the following:
ipfwadm -F -a masquerade -W ppp0 -S $LOCALNET
Does this *confine* masqerading operations to traffic passing through
-W's interface? Would http traffic routed to pass through a different
interface get masqueraded or not? (Usual setup, this linux box gives
gateway service to other machines on the LAN it's attached to.)
Till recently the only external connection was ppp0, with the default
route going through it. A DSL connection has been added, and the
default route changed to go there (eth1) instead.
Therefore, packet traffic from a workstation seeking to browse a
website at some random IP addresss now ends up headed out through
eth1. But the "-W ppp0" in the masq setup rule somehow associates
masqerading with the old ppp0-- now a dry riverbed.
>From my customer's reports (I am off site with remote access to the
linux box but not its LAN clients) it seems that masquerading is
operational (ie, they see outside websites) but super slow.
Here's what ipfwadm man page says about -W:
-W name
Optional name of an interface via which a packet is
received, or via which is packet is going to be
sent.
This doesn't tell me if it's restrictive or not as to the interface on
which the rule (masqueraded forwarding in this case) will be
performed.
Anybody know?
------------------------------
From: "David B. Putnam" <[EMAIL PROTECTED]>
Subject: Re: SAMBA slowness
Date: 13 Sep 1999 00:12:03 GMT
Neil A Fraser wrote:
>
> I am running Samba 2.0.3 under RedHat Linux 6.0 on a Celeron 333MHz with 32MB
> memory. Its sole purpose is to provide file and print services to Win98
> clients in a single workgroup on a small 4-client network.
>
> It works fine logically, but it's slow. If I compare the time taken to
> copy 20 MBytes of files to the Samba PC with the time taken to copy the same
> files to one of the Win 98 PCs, the ratio of time taken is about 4:1. And
> that's with noone logged onto the Samba PC, which is otherwise idling.
>
> If two Win 98 clients access the same set of files on the Samba PC (e.g. for
> the accounting system), I sit back and reorganize my desk while I wait for the
> system to load.
>
> I did a straight swap from the PC booting and running Win95 to the
> ame PC running Linux/SAMBA. Both work, but the Win95 peer-peer networking
> seemed quite a lot faster. Otherwise nothing changed, so I cannot suspect
> networking hardware problems.
>
> I'm determined to fix whatever is causing the Samba slowness, and would
> appreciate any comments about what I should be looking for.
>
> Neil
try this
socket options = TCP_NODELAY
in smb.conf
DBP
------------------------------
** 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
******************************
