|
Home > Archive > Red Hat Networking > January 2004 > network terminology
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
network terminology
|
|
|
| Network terminology gets real confusing, when you have more than one
NIC in your Linux box.
eg:
The /etc/hosts file is said to contain IP address to hostname
translation.
But IP-addresses are linked 1-1 to interfaces (eth0, eth1 etc) if you
have 2 NICs.
Only when you have one NIC, there is a 1-1 relation to the host.
So hosts gives each interface card an FQDN...
any comment ?
frgr
Erik
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
> Network terminology gets real confusing, when you have more than one
> NIC in your Linux box.
>
> eg:
> The /etc/hosts file is said to contain IP address to hostname
> translation.
> But IP-addresses are linked 1-1 to interfaces (eth0, eth1 etc) if you
> have 2 NICs.
> Only when you have one NIC, there is a 1-1 relation to the host.
>
> So hosts gives each interface card an FQDN...
>
> any comment ?
>
> frgr
> Erik
>
>
The way you are stating that there is a 1-1 relationship to an interface
is a little ambiguous. Interface and NIC are not quite the same thing.
You can have multiple interfaces on a single NIC for example the first
interface on your first NIC is eth0 and the second is eth0:1 the third
eth0:2 and so forth. So yes there is a 1-1 relationship between IPs and
interfaces but not NICs.
The /etc/hosts file gives you name to IP and IP to name resolution if
the hosts entry in the /etc/nsswitch.conf file has the files parameter.
So if you have the following in your host file:
192.168.1.10 pcname.yourdomain.com pcname annothername
You can refer to the interface that has the IP address 192.168.1.10 as
pcname.yourdomain.com or pcname or annothername. So I do not know what
you are trying to say by stating "hosts gives each interface card an
FQDN" all it does is allows you to associate any legal name to an IP
address for your local computer.
I usually only refer to a FQDN when I am talking about name resolution
via DNS and not with a host file.
Charles LaCour
| |
|
| On Tue, 25 Nov 2003 13:31:05 -0800, the right honourable Charles
LaCour <calacour@cox.net> wrote:
quote:
>€®ik wrote:
>The way you are stating that there is a 1-1 relationship to an interface
>is a little ambiguous. Interface and NIC are not quite the same thing.
> You can have multiple interfaces on a single NIC for example the first
>interface on your first NIC is eth0 and the second is eth0:1 the third
>eth0:2 and so forth. So yes there is a 1-1 relationship between IPs and
>interfaces but not NICs.
>
>The /etc/hosts file gives you name to IP and IP to name resolution if
>the hosts entry in the /etc/nsswitch.conf file has the files parameter.
>
>So if you have the following in your host file:
>192.168.1.10 pcname.yourdomain.com pcname annothername
>
>You can refer to the interface that has the IP address 192.168.1.10 as
>pcname.yourdomain.com or pcname or annothername. So I do not know what
>you are trying to say by stating "hosts gives each interface card an
>FQDN" all it does is allows you to associate any legal name to an IP
>address for your local computer.
>
>I usually only refer to a FQDN when I am talking about name resolution
>via DNS and not with a host file.
>
>Charles LaCour
I didn't know about multiple IP addresses for a NIC...
ok...
thing is, a host is not an interface and not a NIC. So why call the
file "hosts" ? it's about interfaces... not hosts...
things like this make reading up on comp-stuff much harder..
ok, computerists are unequal to linguists too, but language is our
primary medium of communication, IRL and on-line...
frgr
Erik
| |
|
| On Tue, 25 Nov 2003 09:23:28 +0100, the right honourable €®ik <>
wrote:
quote:
>Network terminology gets real confusing, when you have more than one
>NIC in your Linux box.
>
>eg:
>The /etc/hosts file is said to contain IP address to hostname
>translation.
>But IP-addresses are linked 1-1 to interfaces (eth0, eth1 etc) if you
>have 2 NICs.
>Only when you have one NIC, there is a 1-1 relation to the host.
>
>So hosts gives each interface card an FQDN...
>
>any comment ?
>
>frgr
>Erik
>
also:
in ifcfg-eth0 one write features of the interface, as the name
suggests.
Why write the network's GATEWAY in it too ? the gateway is a feature
of a subnet, is it not ?
al these naming things make life uneasy... for me at least..
Erik
| |
| Alexander Dalloz 2004-01-23, 7:48 pm |
| On Tue, 25 Nov 2003 23:07:01 +0100 wrote:
quote:
> On Tue, 25 Nov 2003 13:31:05 -0800, the right honourable Charles
> LaCour <calacour@cox.net> wrote:
>
>
> I didn't know about multiple IP addresses for a NIC...
> ok...
> thing is, a host is not an interface and not a NIC. So why call the
> file "hosts" ? it's about interfaces... not hosts...
No, it gives a FQDN hostname (if wanted a shortname too) an IP address.
quote:
> things like this make reading up on comp-stuff much harder..
>
> ok, computerists are unequal to linguists too, but language is our
> primary medium of communication, IRL and on-line...
Well, you will not changer terminology as it is old as computers and nets.
quote:
> frgr
> Erik
Alexander
--
Alexander Dalloz | Enger, Germany
PGP key valid: made 13.07.1999
PGP fingerprint: 2307 88FD 2D41 038E 7416 14CD E197 6E88 ED69 5653
| |
| Alexander Dalloz 2004-01-23, 7:48 pm |
| On Tue, 25 Nov 2003 23:25:22 +0100 wrote:
quote:
> On Tue, 25 Nov 2003 09:23:28 +0100, the right honourable €®ik <>
> wrote:
>
>
> also:
>
> in ifcfg-eth0 one write features of the interface, as the name
> suggests.
> Why write the network's GATEWAY in it too ? the gateway is a feature
> of a subnet, is it not ?
The GATEWAY entry is for setting up a route (even a default route) and so
for network traffic finding it's way out of the own net into any other.
quote:
> al these naming things make life uneasy... for me at least..
>
>
> Erik
Alexander
--
Alexander Dalloz | Enger, Germany
PGP key valid: made 13.07.1999
PGP fingerprint: 2307 88FD 2D41 038E 7416 14CD E197 6E88 ED69 5653
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
> On Tue, 25 Nov 2003 13:31:05 -0800, the right honourable Charles
> LaCour <calacour@cox.net> wrote:
>
>
>
>
> I didn't know about multiple IP addresses for a NIC...
> ok...
> thing is, a host is not an interface and not a NIC. So why call the
> file "hosts" ? it's about interfaces... not hosts...
>
> things like this make reading up on comp-stuff much harder..
>
> ok, computerists are unequal to linguists too, but language is our
> primary medium of communication, IRL and on-line...
>
>
>
> frgr
> Erik
In general the /etc/hosts file allows the resolution of names to IP
addresses, this includes "hosts" other than the local computer. So the
name hosts still applies.
Other flavors of unix use the hosts file for other purposes as well.
Solaris for example ises the /etc/hosts file to map the host/computer
name to its IP address. In Solaris you have a file /etc/hostname.hme0
that has the host name for the computer on the hme0 interface and
durring boot time Solaris looks in the host file for an entry that
matches what is in the /etc/hostname.hme0 and assigns the IP it finds to
that interface. In Linux the interface initalization is done a little
different, but for concistancy for people like me who use different *nix
OSs it is still used.
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
> On Tue, 25 Nov 2003 09:23:28 +0100, the right honourable €®ik <>
> wrote:
>
>
>
>
> also:
>
> in ifcfg-eth0 one write features of the interface, as the name
> suggests.
> Why write the network's GATEWAY in it too ? the gateway is a feature
> of a subnet, is it not ?
> al these naming things make life uneasy... for me at least..
>
>
> Erik
The GATEWAY entry in the /etc/sysconfig/network-scripts/ifcfg-* files
defines what gateway the interface should use not the gateway for the
network. The GATEWAY entry in the etc/sysconfig/network-scripts/ifcfg-*
files should match with what the actual gateway is on the network that
the interface is on.
The /etc/sysconfig/network-scripts/ifcfg-* files define the information
that the interface needs to have to be able to communicate. For example
the combination of the ipaddr and the netmask tells what IPs are on the
same logical network segment as the interface (if it needs to be sent to
a router or not). The gateway is defined to tell an interface how to
get to IP addresses that are not on its subnet. So if you have an IP
address of 192.168.1.10 and a netmask of 255.255.255.0 and a gateway of
192.168.1.1 the following are true:
- For the interface to communicate to 192.168.1.* there is nothing
special it needs to do.
- For the interface to communicate to any address that is not in the
192.168.1.* range of addresses it will communicate through the gateway
192.168.1.1.
| |
|
| quote:
>
>No, it gives a FQDN hostname (if wanted a shortname too) an IP address.
What is an FQDN hostname ?
Fully Qualified Domain Name hostename
does not sound right.
jees, i just started to try to unravle the nameing issues here and
thought aw give it up its too complicated/confusing...
FQDNs for interfaces
nets given an domain name
interfaces with gateway definitions (ifcfg-eth0)
.....
talking about language....
quote:
>Well, you will not changer terminology as it is old as computers and nets.
no, I do not have that ambition. But it sure would be helpfull if
these confusions were pointed out in the docs.
frgr
Erik
| |
|
| >The GATEWAY entry in the /etc/sysconfig/network-scripts/ifcfg-* filesquote:
>defines what gateway the interface should use not the gateway for the
>network. The GATEWAY entry in the etc/sysconfig/network-scripts/ifcfg-*
So they need not be the same ?
quote:
>files should match with what the actual gateway is on the network that
>the interface is on.
>
So they must be the same ?
or can there be many gateways on a net ? yes there can be, because a
net can be connected to many other nets.
quote:
>The /etc/sysconfig/network-scripts/ifcfg-* files define the information
>that the interface needs to have to be able to communicate. For example
>the combination of the ipaddr and the netmask tells what IPs are on the
>same logical network segment as the interface (if it needs to be sent to
>a router or not). The gateway is defined to tell an interface how to
>get to IP addresses that are not on its subnet. So if you have an IP
>address of 192.168.1.10 and a netmask of 255.255.255.0 and a gateway of
>192.168.1.1 the following are true:
> - For the interface to communicate to 192.168.1.* there is nothing
>special it needs to do.
> - For the interface to communicate to any address that is not in the
>192.168.1.* range of addresses it will communicate through the gateway
>192.168.1.1.
so: itself ?
but in the 2-NIC situation:
eth0 is 10.0.0.150/8
eth1 is 192.168.0.1/28
eth0 is connected to my alcatel-modem (out to the internet) which has
10.0.0.138.
eth1 is connected to a switch with some other machines (192.168.0.x)
plugged in.
what would be the gateway for eth0 and eth1 ?
you suggest that for eth1 it would be 10.0.0.150 if i'm correct....
but my book says, for this 2-NIC situation, neither eth0 or eth1 needs
a gateway definition. One only needs a gateway definition for a
default route (which would be along the 10.0.0.150 line, to the
outside)
This would be gateway 10.0.0.138.
and for the machines behind the switch ? what would be their gateway ?
192.168.0.1 ?
All books describe a 1-NIC situation. 2-NIC setups are difficult to
find...
thanks anyway for your enlightenment.
I'm very language-sensitive. I always find, that using the right words
help tremendously to clarify things.
frgr
Erik
| |
| Alexander Dalloz 2004-01-23, 7:48 pm |
| On Wed, 26 Nov 2003 08:55:32 +0100 wrote:
quote:
>
>
> What is an FQDN hostname ?
> Fully Qualified Domain Name hostename
> does not sound right.
Well, what I ment with "FQDN hostname" was a hostname with a station name
part (close meaning of hostname), a domain part and a TLD part. So for
instance workstation1.provider-domain.com would be such a naming which
follows DNS naming structure which should be followed if computers are
linked to the net and are not just private stations in a closed LAN.
quote:
> Erik
Alexander
--
Alexander Dalloz | Enger, Germany
PGP key valid: made 13.07.1999
PGP fingerprint: 2307 88FD 2D41 038E 7416 14CD E197 6E88 ED69 5653
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
>
>
> So they need not be the same ?
>
>
>
>
>
> So they must be the same ?
> or can there be many gateways on a net ? yes there can be, because a
> net can be connected to many other nets.
>
>
>
>
>
> so: itself ?
>
>
> but in the 2-NIC situation:
> eth0 is 10.0.0.150/8
> eth1 is 192.168.0.1/28
> eth0 is connected to my alcatel-modem (out to the internet) which has
> 10.0.0.138.
> eth1 is connected to a switch with some other machines (192.168.0.x)
> plugged in.
> what would be the gateway for eth0 and eth1 ?
> you suggest that for eth1 it would be 10.0.0.150 if i'm correct....
>
> but my book says, for this 2-NIC situation, neither eth0 or eth1 needs
> a gateway definition. One only needs a gateway definition for a
> default route (which would be along the 10.0.0.150 line, to the
> outside)
> This would be gateway 10.0.0.138.
>
> and for the machines behind the switch ? what would be their gateway ?
> 192.168.0.1 ?
>
> All books describe a 1-NIC situation. 2-NIC setups are difficult to
> find...
>
> thanks anyway for your enlightenment.
>
> I'm very language-sensitive. I always find, that using the right words
> help tremendously to clarify things.
>
> frgr
> Erik
>
The GATEWAY entry in the ifcfg-* file needs to be set to the IP address
for the gateway for the network the inerface is on.
Yes, there can be many gatways on a network but it depends on the
network topology but there is usually a single default gateway on a network.
The Linux box you are setting up knows bout the 10.0.0.150/8 and
192.168.0.1/28 networks and has no need to have a gateway defined co
comunicate on them. If you need to comunicate with a network that is on
the other side of a router you need to define the gateway.
If there is a router on your 192.168.0.1/28 network that conects to
another network then you would need to set the GATEWAY on the eth1
interface to that router otherwise it is not neded.
The eth0 interface would need to have the GATEWAY entry set to your ISPs
default router for that network.
The computers on the 192.168.0.1/28 netwoek other than the Linux one
should have the gateway set to 192.168.0.1.
| |
|
| quote:
>The GATEWAY entry in the ifcfg-* file needs to be set to the IP address
>for the gateway for the network the inerface is on.
>
>Yes, there can be many gatways on a network but it depends on the
>network topology but there is usually a single default gateway on a network.
>
>The Linux box you are setting up knows bout the 10.0.0.150/8 and
>192.168.0.1/28 networks and has no need to have a gateway defined co
>comunicate on them. If you need to comunicate with a network that is on
>the other side of a router you need to define the gateway.
>
>If there is a router on your 192.168.0.1/28 network that conects to
>another network then you would need to set the GATEWAY on the eth1
>interface to that router otherwise it is not neded.
>
>The eth0 interface would need to have the GATEWAY entry set to your ISPs
>default router for that network.
>
>The computers on the 192.168.0.1/28 netwoek other than the Linux one
>should have the gateway set to 192.168.0.1.
thank you for your input.
If i understand correctly:
there is no other router on the 192.168.0.0 network, so my linux box
needs no gateway definition on the eth1 (192.168.0.1) interface
my eth0 (10.0.0.150) does need a gateway definition of 10.0.0.138
because it is directly connected to my modem/router (10.0.0.138) (and
to nothing else) which goes out on the ADSL line to my ISP.
a box that is "bridging" two nets, does itself need no gateway
definition on any interface.
Only the other computers on both nets need a gateway.
That is why eth0 needs 10.0.0.138 as a gateway definition: 10.0.0.150
needs to send packets, not destined for any interface on its net,
out some interface on the 10.0.0.0 net, out into the world.
Right now I have a SUSE linux machine in the same position and it has
for routing table the following, which works well:
Destination Gateway Genmask Flags Metric Ref Use Iface
10.0.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
192.168.0.0 192.168.0.1 255.255.255.0 UG 0 0 0 eth1
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
0.0.0.0 10.0.0.138 0.0.0.0 UG 0 0 0 eth0
I'm trying to get a RH9 machine up and running and it works different
from the SUSE network software.
I still can't figure out how to set up RH9...
So now I'll try out your suggestions.
frgr
Erik
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| In-Reply-To: <s7ocsv0936tgu82j9c1jl87eco6s807je3@4ax.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 8bit
Lines: 143
Message-ID: <TONxb.7908$ZE1.1912@fed1read04>
Date: Fri, 28 Nov 2003 11:59:52 -0800
NNTP-Posting-Host: 68.111.220.15
X-Complaints-To: abuse@cox.net
X-Trace: fed1read04 1070049587 68.111.220.15 (Fri, 28 Nov 2003 14:59:47 EST)
NNTP-Posting-Date: Fri, 28 Nov 2003 14:59:47 EST
Organization: Cox Communications
Xref: intern1.nntp.aus1.giganews.com redhat.networking.general:10827
€®ik wrote:quote:
>
>
> thank you for your input.
>
> If i understand correctly:
>
> there is no other router on the 192.168.0.0 network, so my linux box
> needs no gateway definition on the eth1 (192.168.0.1) interface
>
> my eth0 (10.0.0.150) does need a gateway definition of 10.0.0.138
> because it is directly connected to my modem/router (10.0.0.138) (and
> to nothing else) which goes out on the ADSL line to my ISP.
>
> a box that is "bridging" two nets, does itself need no gateway
> definition on any interface.
>
> Only the other computers on both nets need a gateway.
>
> That is why eth0 needs 10.0.0.138 as a gateway definition: 10.0.0.150
> needs to send packets, not destined for any interface on its net,
> out some interface on the 10.0.0.0 net, out into the world.
>
> Right now I have a SUSE linux machine in the same position and it has
> for routing table the following, which works well:
>
> Destination Gateway Genmask Flags Metric Ref Use Iface
> 10.0.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
> 192.168.0.0 192.168.0.1 255.255.255.0 UG 0 0 0 eth1
> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
> 0.0.0.0 10.0.0.138 0.0.0.0 UG 0 0 0 eth0
>
> I'm trying to get a RH9 machine up and running and it works different
> from the SUSE network software.
> I still can't figure out how to set up RH9...
>
> So now I'll try out your suggestions.
>
> frgr
> Erik
>
>
>
Here is a quick diagram that should fit your network from what you have
said.
----- -----
| | N1 | |
| I |---------| A |
| | | |
----- -----
|
|N2
|
----- N3 -----
| | ----- | |
| C |--| s |--| B |
| | ----- | |
----- | | -----
| |
| |
----- | | -----
| | | | | |
| D |---+ +---| E |
| | | |
----- -----
I - The internet
A - Your modem/router
B - Your Linux server
C,D,E - Other computers on your internal network.
s - Switch/Hub for your internal network
N1 - Your ISP's network
N2 - The network segment between your Linux PC and your modem/router
(10.0.0.150/8 addresses 10.0.0.0 to 10.255.255.255)
N3 - Your internal network (192.168.0.1/28 addresses 192.168.0.0 to
192.168.0.15)
Your Linux server (B) can communicate on N2 and N3 without a gateway, if
it needs to comunicate with N1 it has to do so through the gateway (A)
between N1 and N2.
The computers C, D and E can comunicate with anything on the N3 network
without a gateway but if they need to comunicate with something on N2 or
N1 then they need to have their gateway defined as the IP address of the
interface of B that is on the N3 network.
Looking at your routing table if your internal network is acctually
192.168.0.1/28 then your route for 192.168.0.0 has the wrong net mask.
A /28 netmask is equivilant to 255.255.255.240 limiting you to 14 usable
IP adresses on that network. If you truely want a 255.255.255.0 net
mask that would be equivilant of /24. The route for 10.0.0.0 is has a
an issue with its net mask as well if you want a /8 subnet mask it
should be 255.0.0.0. Other than that the routing table looks good!
Take a look at the file /usr/share/doc/initscripts-*/sysconfig.txt for
all of the parameters in the configuration files under the
/etc/sysconfig directory.
The ifcfg-* files should look something like this:
/etc/sysconfig/network-scripts/ifcfg-eth0
GATEWAY=10.0.0.138
ONBOOT=yes
TYPE=Ethernet
IPADDR=10.0.0.150
DEVICE=eth0
BOOTPROTO=none
NETMASK=255.0.0.0
USERCTL=no
PEERDNS=yes
/etc/sysconfig/network-scripts/ifcfg-eth1
ONBOOT=yes
TYPE=Ethernet
IPADDR=192.168.0.1
DEVICE=eth1
BOOTPROTO=none
NETMASK=255.255.255.240
USERCTL=no
PEERDNS=yes
| |
|
| quote:
>Here is a quick diagram that should fit your network from what you have
>said.
> ----- -----
> | | N1 | |
> | I |---------| A |
> | | | |
> ----- -----
> |
> |N2
> |
> ----- N3 -----
> | | ----- | |
> | C |--| s |--| B |
> | | ----- | |
> ----- | | -----
> | |
> | |
> ----- | | -----
> | | | | | |
> | D |---+ +---| E |
> | | | |
> ----- -----
>
>
>I - The internet
>A - Your modem/router
>B - Your Linux server
>C,D,E - Other computers on your internal network.
>s - Switch/Hub for your internal network
>N1 - Your ISP's network
>N2 - The network segment between your Linux PC and your modem/router
>(10.0.0.150/8 addresses 10.0.0.0 to 10.255.255.255)
>N3 - Your internal network (192.168.0.1/28 addresses 192.168.0.0 to
>192.168.0.15)
>
>Your Linux server (B) can communicate on N2 and N3 without a gateway, if
>it needs to comunicate with N1 it has to do so through the gateway (A)
>between N1 and N2.
>
>The computers C, D and E can comunicate with anything on the N3 network
>without a gateway but if they need to comunicate with something on N2 or
>N1 then they need to have their gateway defined as the IP address of the
>interface of B that is on the N3 network.
>
>Looking at your routing table if your internal network is acctually
>192.168.0.1/28 then your route for 192.168.0.0 has the wrong net mask.
>A /28 netmask is equivilant to 255.255.255.240 limiting you to 14 usable
>IP adresses on that network. If you truely want a 255.255.255.0 net
>mask that would be equivilant of /24. The route for 10.0.0.0 is has a
>an issue with its net mask as well if you want a /8 subnet mask it
>should be 255.0.0.0. Other than that the routing table looks good!
>
>
>Take a look at the file /usr/share/doc/initscripts-*/sysconfig.txt for
>all of the parameters in the configuration files under the
>/etc/sysconfig directory.
>
>The ifcfg-* files should look something like this:
>
>/etc/sysconfig/network-scripts/ifcfg-eth0
>GATEWAY=10.0.0.138
>ONBOOT=yes
>TYPE=Ethernet
>IPADDR=10.0.0.150
>DEVICE=eth0
>BOOTPROTO=none
>NETMASK=255.0.0.0
>USERCTL=no
>PEERDNS=yes
>
>/etc/sysconfig/network-scripts/ifcfg-eth1
>ONBOOT=yes
>TYPE=Ethernet
>IPADDR=192.168.0.1
>DEVICE=eth1
>BOOTPROTO=none
>NETMASK=255.255.255.240
>USERCTL=no
>PEERDNS=yes
i'm thinking of hanging myself.....
I still can't get the machine working as I want it to.
Maybe you can help me out. here are some of the configuration files I
have now:
/etc/sysconfig/network-scripts/ifcfg-eth0:
NAME=alcatel
DEVICE=eth0
BOOTPROTO=none
IPADDR=10.0.0.150
NETMASK=255.0.0.0
ONBOOT=yes
GATEWAY=10.0.0.138
PEERDNS=yes
USERCTL=no
TYPE=Ethernet
/etc/sysconfig/network-scripts/ifcfg-eth1:
NAME=lima
DEVICE=eth1
BOOTPROTO=none
GATEWAY=192.168.0.1
NETMASK=255.255.255.240
IPADDR=192.168.0.1
ONBOOT=yes
TYPE=Ethernet
PEERDNS=yes
USERCTL=no
/etc/sysctl.conf:
net.ipv4.ip_forward = 1
net.ipv4.conf.default.rp_filter = 1
kernel.sysrq = 0
kernel.core_uses_pid = 1
Also I did:
echo "1" > /proc/sys/net/ipv4/ip_forward
The route table (route -n)
Destination Gateway Genmask Flags Metric Ref
Use Iface
192.168.0.0 0.0.0.0 255.255.255.240 U 0 0
0 eth1
10.0.0.0 0.0.0.0 255.0.0.0 U 0 0
0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0
0 lo
0.0.0.0 10.0.0.138 0.0.0.0 UG 0 0
0 eth0
I can ping 10.0.0.150 and 192.168.0.1.
But nothing further: not 10.0.0.138 and no 192.168.0.3, which should
both be reachable.
Looks to me, there is a problem with the gateway definitions.
But how must I change which file to get a GW definition in the route
table ?
in /etc/sysconfig/network-scripts I have no extra route definitions.
You say, the netmask for the 192.168 net would be a problem. I don't
understand: if I need only 14 IP addresses on that net (192.168.0.1 to
0.14) I could do that ... I think...
192.168.0.15 would be the broadcast address and 192.168.0.0 would be
the net number.
right ?
frgr
Erik
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
>
>
>
> i'm thinking of hanging myself.....
>
> I still can't get the machine working as I want it to.
> Maybe you can help me out. here are some of the configuration files I
> have now:
> /etc/sysconfig/network-scripts/ifcfg-eth0:
> NAME=alcatel
> DEVICE=eth0
> BOOTPROTO=none
> IPADDR=10.0.0.150
> NETMASK=255.0.0.0
> ONBOOT=yes
> GATEWAY=10.0.0.138
> PEERDNS=yes
> USERCTL=no
> TYPE=Ethernet
>
> /etc/sysconfig/network-scripts/ifcfg-eth1:
> NAME=lima
> DEVICE=eth1
> BOOTPROTO=none
> GATEWAY=192.168.0.1
> NETMASK=255.255.255.240
> IPADDR=192.168.0.1
> ONBOOT=yes
> TYPE=Ethernet
> PEERDNS=yes
> USERCTL=no
>
> /etc/sysctl.conf:
> net.ipv4.ip_forward = 1
> net.ipv4.conf.default.rp_filter = 1
> kernel.sysrq = 0
> kernel.core_uses_pid = 1
>
>
> Also I did:
> echo "1" > /proc/sys/net/ipv4/ip_forward
>
> The route table (route -n)
>
> Destination Gateway Genmask Flags Metric Ref
> Use Iface
> 192.168.0.0 0.0.0.0 255.255.255.240 U 0 0
> 0 eth1
> 10.0.0.0 0.0.0.0 255.0.0.0 U 0 0
> 0 eth0
> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0
> 0 lo
> 0.0.0.0 10.0.0.138 0.0.0.0 UG 0 0
> 0 eth0
>
> I can ping 10.0.0.150 and 192.168.0.1.
> But nothing further: not 10.0.0.138 and no 192.168.0.3, which should
> both be reachable.
> Looks to me, there is a problem with the gateway definitions.
> But how must I change which file to get a GW definition in the route
> table ?
>
> in /etc/sysconfig/network-scripts I have no extra route definitions.
>
> You say, the netmask for the 192.168 net would be a problem. I don't
> understand: if I need only 14 IP addresses on that net (192.168.0.1 to
> 0.14) I could do that ... I think...
> 192.168.0.15 would be the broadcast address and 192.168.0.0 would be
> the net number.
> right ?
>
> frgr
> Erik
The netmask on all the interfaces on the same network segment have to
match. The netmask determines the broadcast address for that network.
The broadcast address is important for address discovery. When you ping
an IP address on a local network segment if the hardware, MAC, address
for that IP address is not known an ARP packet is sent out addressed to
the broadcast address requesting for the owner of the IP address to
respond. If the broadcast addresses on different computers on the same
network do not match they will not be able to communicate.
You are correct that you would have the useable addresses 192.168.0.1 to
..14 and .15 is the broadcast address. The problem I was mentioning with
the netmask was not that you could not use a 255.255.255.240 netmask but
that the netmask you gave for the interface and in the routing table
were different.
There is no need for the GATEWAY line in your ifcfg-eth1 file.
Can any of the other computers on the 192.168.0.* network ping each
other? Are there any other Linux PCs on the 192.168.0.*? If so run the
command "tcpdump host 192.168.0.1 and arp" as root and then try to ping
it from the PC you are having the network problems. Do you see anything
from the tcpdump command?
Charles LaCour
| |
|
| here is no need for the GATEWAY line in your ifcfg-eth1 file.quote:
>
>Can any of the other computers on the 192.168.0.* network ping each
>other? Are there any other Linux PCs on the 192.168.0.*? If so run the
>command "tcpdump host 192.168.0.1 and arp" as root and then try to ping
>it from the PC you are having the network problems. Do you see anything
>from the tcpdump command?
>
>Charles LaCour
from a Windows XP machine (192.168.0.3) , I can ping around to an
intel netport printserver (192.168.0.10) and one other winXP machine.
But i cannot ping to 192.168.0.1 (Request timeout).
Maybe the ping can go out to 0.1, but it cannot return...
Looks like the problem really is in the setup of the 0.1 RH9-machine.
So I simplified the situation:
I reinstalled the original firewall machine in between the modem and
the internal net. It's a SUSE 8.1 machine, called tango, and it does
its job well.
In the internal net I now have a few WIN machines and two RH9 linux
machines, called lima and charlie.
from the Win machines I can ping everything except the RH9 boxex. from
tango, I can ping the windows machines.
the RH9 machines don't seem to be there when pinging: I can't ping to
them and not from them: network unreachable.
tcpdump on tango does not blink an eye...
tcpdump on tango does report well when pinging from Windows machines.
on charlie and lima (RH9) , ifconfig reports correctly.
I have the distinct impression, something is wrong with the building
up of the route table.
I don't suppose the RH firewall would hold things up, would it ?
frgr
Erik
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
> here is no need for the GATEWAY line in your ifcfg-eth1 file.
>
>
>
>
>
> from a Windows XP machine (192.168.0.3) , I can ping around to an
> intel netport printserver (192.168.0.10) and one other winXP machine.
> But i cannot ping to 192.168.0.1 (Request timeout).
> Maybe the ping can go out to 0.1, but it cannot return...
> Looks like the problem really is in the setup of the 0.1 RH9-machine.
>
> So I simplified the situation:
> I reinstalled the original firewall machine in between the modem and
> the internal net. It's a SUSE 8.1 machine, called tango, and it does
> its job well.
>
> In the internal net I now have a few WIN machines and two RH9 linux
> machines, called lima and charlie.
> from the Win machines I can ping everything except the RH9 boxex. from
> tango, I can ping the windows machines.
> the RH9 machines don't seem to be there when pinging: I can't ping to
> them and not from them: network unreachable.
> tcpdump on tango does not blink an eye...
> tcpdump on tango does report well when pinging from Windows machines.
>
> on charlie and lima (RH9) , ifconfig reports correctly.
>
> I have the distinct impression, something is wrong with the building
> up of the route table.
> I don't suppose the RH firewall would hold things up, would it ?
>
>
> frgr
> Erik
It looks like there is something odd going on here. Please run the
following commands on the Linux machines and include the results in a reply.
ipchains -L
ifconfig -a
netstat -rn
arp -a
Please run the following from a command prompt on the Win XP PC.
ipconfig /all
route print
arp -a
Charles LaCour
| |
|
| quote:
>It looks like there is something odd going on here. Please run the
>following commands on the Linux machines and include the results in a reply.
>
>ipchains -L
>ifconfig -a
>netstat -rn
>arp -a
>
>Please run the following from a command prompt on the Win XP PC.
>
>ipconfig /all
>route print
>arp -a
>
ipchains -L is "unknown command"
ifconfig -a gives the normal output. addresses are ok:
(remember:
- eth1 is unplugged here, eth0 is used
- I use mask 255.255.255.0 for the 192-net
)
****************************************
****************
eth0 Link encap:Ethernet HWaddr 00:04:76:20:DF:DA
inet addr:192.168.0.8 Bcast:192.168.0.255
Mask:255.255.255.0
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:232 errors:0 dropped:0 overruns:0 carrier:232
collisions:0 txqueuelen:100
RX bytes:0 (0.0 b) TX bytes:13920 (13.5 Kb)
Interrupt:9 Base address:0xf800
eth1 Link encap:Ethernet HWaddr 00:60:97:B8:90:3D
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
Interrupt:11 Base address:0xf400
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:227 errors:0 dropped:0 overruns:0 frame:0
TX packets:227 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:25396 (24.8 Kb) TX bytes:25396 (24.8 Kb)
****************************************
****************
netstat -m is unknown option,
but netstat -M gives "no support for ip_masquerade on this system"
arp -a gives nothing, it just returns to the prompt.
on the Windows system, I get the following info for the three commands
you suggested:
****************************************
**********************************
Windows IP Configuration
Host Name . . . . . . . . . . . . : echo
Primary Dns Suffix . . . . . . . :
Node Type . . . . . . . . . . . . : Unknown
IP Routing Enabled. . . . . . . . : No
WINS Proxy Enabled. . . . . . . . : Yes
Ethernet adapter Local Area Connection 4:
Connection-specific DNS Suffix . :
Description . . . . . . . . . . . : Sitecom PCI Fast 10/100
Ethernet Adapter LN-020
Physical Address. . . . . . . . . : 00-50-BF-99-8F-70
Dhcp Enabled. . . . . . . . . . . : No
IP Address. . . . . . . . . . . . : 192.168.0.3
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . : 192.168.0.1
DNS Servers . . . . . . . . . . . : 195.121.1.34
195.121.1.66
========================================
===================================
Interface List
0x1 ........................... MS TCP Loopback interface
0x50002 ...00 50 bf 99 8f 70 ...... Sitecom PCI Fast 10/100 Ethernet
Adapter LN-020 - Packet Scheduler Miniport
========================================
===================================
========================================
===================================
Active Routes:
Network Destination Netmask Gateway Interface
Metric
0.0.0.0 0.0.0.0 192.168.0.1 192.168.0.3
20
66.125.134.118 255.255.255.255 192.168.0.1 192.168.0.3
20
127.0.0.0 255.0.0.0 127.0.0.1 127.0.0.1
1
192.168.0.0 255.255.255.0 192.168.0.3 192.168.0.3
20
192.168.0.3 255.255.255.255 127.0.0.1 127.0.0.1
20
192.168.0.255 255.255.255.255 192.168.0.3 192.168.0.3
20
224.0.0.0 240.0.0.0 192.168.0.3 192.168.0.3
20
255.255.255.255 255.255.255.255 192.168.0.3 192.168.0.3
1
Default Gateway: 192.168.0.1
========================================
===================================
Persistent Routes:
None
Interface: 192.168.0.3 --- 0x50002
Internet Address Physical Address Type
192.168.0.1 00-10-a7-0c-03-a3 dynamic
****************************************
****************************************
| |
|
| What really bothers me is, that it is a clean install from the CD's...
and It's the same problem on both RH9 machines.
I do not think I did something fishy during install...
| |
| Charles LaCour 2004-01-23, 7:48 pm |
| €®ik wrote:quote:
>
>
> ipchains -L is "unknown command"
>
> ifconfig -a gives the normal output. addresses are ok:
> (remember:
> - eth1 is unplugged here, eth0 is used
> - I use mask 255.255.255.0 for the 192-net
> )
> ****************************************
****************
>
>
> eth0 Link encap:Ethernet HWaddr 00:04:76:20:DF:DA
> inet addr:192.168.0.8 Bcast:192.168.0.255
> Mask:255.255.255.0
> UP BROADCAST MULTICAST MTU:1500 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0 frame:0
> TX packets:232 errors:0 dropped:0 overruns:0 carrier:232
> collisions:0 txqueuelen:100
> RX bytes:0 (0.0 b) TX bytes:13920 (13.5 Kb)
> Interrupt:9 Base address:0xf800
>
> eth1 Link encap:Ethernet HWaddr 00:60:97:B8:90:3D
> BROADCAST MULTICAST MTU:1500 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0 frame:0
> TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:100
> RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
> Interrupt:11 Base address:0xf400
>
> lo Link encap:Local Loopback
> inet addr:127.0.0.1 Mask:255.0.0.0
> UP LOOPBACK RUNNING MTU:16436 Metric:1
> RX packets:227 errors:0 dropped:0 overruns:0 frame:0
> TX packets:227 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:0
> RX bytes:25396 (24.8 Kb) TX bytes:25396 (24.8 Kb)
>
>
>
>
> ****************************************
****************
> netstat -m is unknown option,
> but netstat -M gives "no support for ip_masquerade on this system"
>
> arp -a gives nothing, it just returns to the prompt.
>
> on the Windows system, I get the following info for the three commands
> you suggested:
>
> ****************************************
**********************************
>
>
> Windows IP Configuration
>
> Host Name . . . . . . . . . . . . : echo
> Primary Dns Suffix . . . . . . . :
> Node Type . . . . . . . . . . . . : Unknown
> IP Routing Enabled. . . . . . . . : No
> WINS Proxy Enabled. . . . . . . . : Yes
>
> Ethernet adapter Local Area Connection 4:
>
> Connection-specific DNS Suffix . :
> Description . . . . . . . . . . . : Sitecom PCI Fast 10/100
> Ethernet Adapter LN-020
> Physical Address. . . . . . . . . : 00-50-BF-99-8F-70
> Dhcp Enabled. . . . . . . . . . . : No
> IP Address. . . . . . . . . . . . : 192.168.0.3
> Subnet Mask . . . . . . . . . . . : 255.255.255.0
> Default Gateway . . . . . . . . . : 192.168.0.1
> DNS Servers . . . . . . . . . . . : 195.121.1.34
> 195.121.1.66
> ========================================
===================================
> Interface List
> 0x1 ........................... MS TCP Loopback interface
> 0x50002 ...00 50 bf 99 8f 70 ...... Sitecom PCI Fast 10/100 Ethernet
> Adapter LN-020 - Packet Scheduler Miniport
> ========================================
===================================
>
> ========================================
===================================
> Active Routes:
> Network Destination Netmask Gateway Interface
> Metric
> 0.0.0.0 0.0.0.0 192.168.0.1 192.168.0.3
> 20
> 66.125.134.118 255.255.255.255 192.168.0.1 192.168.0.3
> 20
> 127.0.0.0 255.0.0.0 127.0.0.1 127.0.0.1
> 1
> 192.168.0.0 255.255.255.0 192.168.0.3 192.168.0.3
> 20
> 192.168.0.3 255.255.255.255 127.0.0.1 127.0.0.1
> 20
> 192.168.0.255 255.255.255.255 192.168.0.3 192.168.0.3
> 20
> 224.0.0.0 240.0.0.0 192.168.0.3 192.168.0.3
> 20
> 255.255.255.255 255.255.255.255 192.168.0.3 192.168.0.3
> 1
> Default Gateway: 192.168.0.1
> ========================================
===================================
> Persistent Routes:
> None
>
> Interface: 192.168.0.3 --- 0x50002
> Internet Address Physical Address Type
> 192.168.0.1 00-10-a7-0c-03-a3 dynamic
>
> ****************************************
****************************************
There were 2 parameters on the netstat command the first was r for route
and the second was n for numeric not a single m.
Try the folowing commands on the Linux PC and let me know what you get:
arp -s 192.168.0.3 00:50:BF:99:8F:70
arp -a
ping 192.168.0.3
What I am having you do is manually entering the MAC and IP address of
the Win XP PC in your ARP cache and then pining the IP address. This
should bypass the IP to MAC address resolution that is done using the
broadcast address. I am trying to narow down where the problem is.
When you have done the test type the following to remove the static entry:
arp -d 192.168.0.3
Charles LaCour
| |
|
| quote:
>There were 2 parameters on the netstat command the first was r for route
>and the second was n for numeric not a single m.
>
>Try the folowing commands on the Linux PC and let me know what you get:
>arp -s 192.168.0.3 00:50:BF:99:8F:70
>arp -a
>ping 192.168.0.3
>
>What I am having you do is manually entering the MAC and IP address of
>the Win XP PC in your ARP cache and then pining the IP address. This
>should bypass the IP to MAC address resolution that is done using the
>broadcast address. I am trying to narow down where the problem is.
>When you have done the test type the following to remove the static entry:
>arp -d 192.168.0.3
>
>Charles LaCour
here is the output from "netstat -rn" on LIMA (192.168.0.7):
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window
irtt Iface
192.168.0.0 0.0.0.0 255.255.255.0 U 0 0
0 eth1
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0
0 lo
0.0.0.0 192.168.0.1 0.0.0.0 UG 0 0
0 eth1
then I did:
arp -s 192.168.0.3 00:50:BF:99:8F:70
after which "arp -a" gives:
? (192.168.0.3) at 00:50:BF:99:8F:70 [ether] PERM on eth1
lima.terwiel.loc (192.168.0.1) at <incomplete> on eth1
pinging 192.168.0.3 results in:
PING 192.168.0.3 (192.168.0.3) 56(84) bytes of data
then nothing: PING hangs.
Right now, I'm re-installing RH9 with KDE on another machine.
See what's possible with the utilities in KDE...
But I'll keep hammering on LIMA :-)
I find it very weird, that a fresh install from CD gives such
problems..
frgr
Erik
| |
|
| I may have found the problem.
It may be a cabling error on my part.
I think I switched two MAC-address stickers on the two NICs on LIMA.
and thereby misinterpreting eth0 and eth1.
I'm not sure yet.
I'll report back.
| |
|
| OK, it's much better now.
Seems I switched the MAC addresses om the eth0 and eth1 cards.
But it's not all wel yet.
If I put the RH9 machine (LIMA) in between mij internal net
192.168.0.0/24 and the ADSL modem (10.0.0.138), then , from the LIMA
machine I can ping anywhere: to www.yahoo.com (DNS works too), to
10.0.0.138, 10.0.0.150 and to any machine on the 192.168.0.0/24 net.
But not from the machine ECHO (192.168.0.3).
From that one, I can only ping around in the 192.168.0.0/24 net as wel
as to 10.0.0.150, but not to the modem 10.0.0.138 or beyond.
also, if I tracert 10.0.0.138 from ECHO (a WinXP machine), it only
hops to 192.168.0.1.
On the RH9 machine LIMA (192.168.0.1) with tcpdump, I listen to eth1
for ping traffic (ping 10.0.0.138) from ECHO (192.168.0.3) and I can
see the ICMP pings coming in.
Listening to eth0 (10.0.0.150), which goes to the modem (10.0.0.138),
I can see the same ping traffic coming back too, but ECHO reports a
timeout. So pings get stopped on the way back, I suppose.
Maybe routing doesn't know how to get traffic to 192.168.0.1...
looks like something goes wrong with address translation , does it not
?
| |
| Alexander Dalloz 2004-01-23, 7:48 pm |
| On Thu, 04 Dec 2003 18:18:10 +0100 wrote:
quote:
> OK, it's much better now.
> Seems I switched the MAC addresses om the eth0 and eth1 cards.
> But it's not all wel yet.
>
> If I put the RH9 machine (LIMA) in between mij internal net
> 192.168.0.0/24 and the ADSL modem (10.0.0.138), then , from the LIMA
> machine I can ping anywhere: to www.yahoo.com (DNS works too), to
> 10.0.0.138, 10.0.0.150 and to any machine on the 192.168.0.0/24 net.
>
> But not from the machine ECHO (192.168.0.3).
> From that one, I can only ping around in the 192.168.0.0/24 net as wel
> as to 10.0.0.150, but not to the modem 10.0.0.138 or beyond.
>
>
> also, if I tracert 10.0.0.138 from ECHO (a WinXP machine), it only
> hops to 192.168.0.1.
>
>
> On the RH9 machine LIMA (192.168.0.1) with tcpdump, I listen to eth1
> for ping traffic (ping 10.0.0.138) from ECHO (192.168.0.3) and I can
> see the ICMP pings coming in.
>
> Listening to eth0 (10.0.0.150), which goes to the modem (10.0.0.138),
> I can see the same ping traffic coming back too, but ECHO reports a
> timeout. So pings get stopped on the way back, I suppose.
>
> Maybe routing doesn't know how to get traffic to 192.168.0.1...
>
> looks like something goes wrong with address translation , does it not
> ?
Be sure you have on your RH9 machine set on ip forwarding and masquerading:
1) in /etc/sysctl.conf
# Controls IP packet forwarding
net.ipv4.ip_forward = 1
2) i.e. in /etc/sysconfig/iptables
[0:0] -A POSTROUTING -o eth0 -j MASQUERADE
Alexander
--
Alexander Dalloz | Enger, Germany
PGP key valid: made 13.07.1999
PGP fingerprint: 2307 88FD 2D41 038E 7416 14CD E197 6E88 ED69 5653
| |
|
| quote:
>Be sure you have on your RH9 machine set on ip forwarding and masquerading:
>
>1) in /etc/sysctl.conf
> # Controls IP packet forwarding
> net.ipv4.ip_forward = 1
>2) i.e. in /etc/sysconfig/iptables
> [0:0] -A POSTROUTING -o eth0 -j MASQUERADE
>
>Alexander
Thanx for your attention.
yes, sysctl.conf is ok, and /proc/sys/net/ipv4/ip_forward is 1 too.
the /etc/sysconfig/iptables thing is not set as you said.
First thing I do when I get the machine working is studying IPTABLES.
where in iptables do I put this line ?
| |
| Alexander Dalloz 2004-01-23, 7:48 pm |
| On Thu, 04 Dec 2003 22:20:56 +0100 wrote:
quote:
>
>
>
> Thanx for your attention.
>
> yes, sysctl.conf is ok, and /proc/sys/net/ipv4/ip_forward is 1 too.
Ok, good.
quote:
> the /etc/sysconfig/iptables thing is not set as you said.
> First thing I do when I get the machine working is studying IPTABLES.
>
> where in iptables do I put this line ?
iptables -t nat -A POSTROUTING -o $NATDEV -j MASQUERADE
is the command line, where $NATDEV is the device over which you run
masqueraded/nat'ed connections. In you case it seems to be eth0.
You can run that command anytime, but respect that iptables handles the
rules from first to last until one matches. So if any other rule before
would match this rule would not take place. But normally with masquerading
in the nat table this is not the case.
Btw. you list the iptables rules with:
iptables L -n -v
iptables -t nat -n -v
Alexander
--
Alexander Dalloz | Enger, Germany
PGP key valid: made 13.07.1999
PGP fingerprint: 2307 88FD 2D41 038E 7416 14CD E197 6E88 ED69 5653
| |
|
| At last, I succeeded in setting up the routing RH9 box.
here are the two /etc/sysconfig/network-scripts/ifcfg-ethx files:
NAME=alcatel
DEVICE=eth0
BOOTPROTO=none
BROADCAST=10.255.255.255
IPADDR=10.0.0.150
NETMASK=255.0.0.0
NETWORK=10.0.0.0
ONBOOT=yes
HWADDR=00:60:B0:C3:CC:BF
USERCTL=no
PEERDNS=no
TYPE=Ethernet
and:
NAME=lima
DEVICE=eth1
BOOTPROTO=none
BROADCAST=192.168.0.255
IPADDR=192.168.0.1
NETMASK=255.255.255.0
NETWORK=192.168.0.0
ONBOOT=yes
HWADDR=00:60:B0:B5:6D:4B
USERCTL=no
PEERDNS=no
TYPE=Ethernet
and this is the /etc/sysconfig/network file:
NETWORKING=yes
HOSTNAME=lima
NOZEROCONF=yes
GATEWAY=10.0.0.138
and /proc/sys/net/ipv4/ip_forward must be set to 1, of course.
Setting up another RH9 box with KDE, and configuring it that way,
showed a lot of details in the ifcfg-ethx files.
Next step is setting up the Netfilter stuff.
thanks for all your help.
I learned a lot in the process.
Erik
|
|
|
|
|