routing problem

When I dial into my Linux device over ppp, I need to be able to access
devices connected to the eth0 interface. On the device, I can ping
devices on eth0, but not on the connected device. Here is a summary:

[root@Linux ide]# cat /proc/sys/net/ipv4/ip_forward
1

[root@Linux ide]# ifconfig
eth0 Link encap:Ethernet HWaddr 00:50:15:03:00:62
inet addr:10.0.0.2 Bcast:10.255.255.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:2424 errors:0 dropped:0 overruns:0 frame:0
TX packets:3229 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
Base address:0x300

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:3904 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:0

ppp0 Link encap:Point-to-Point Protocol
inet addr:2.15.3.85 P-t-P:2.15.3.84 Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
RX packets:125 errors:0 dropped:0 overruns:0 frame:0
TX packets:88 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3

[root@Linux ide]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use
Iface
2.15.3.84 * 255.255.255.255 UH 0 0 0
ppp0
10.0.0.0 * 255.255.255.0 U 0 0 0
eth0
127.0.0.0 * 255.0.0.0 U 0 0 0
lo

[root@Linux ide]# arp
Address HWtype HWaddress Flags Mask
Iface
10.0.0.1 ether 08:00:19:00:E1:77 C
eth0

PPP adapter linuxmodem:

Connection-specific DNS Suffix . :
IP Address. . . . . . . . . . . . : 2.15.3.84
Subnet Mask . . . . . . . . . . . : 255.255.255.255
Default Gateway . . . . . . . . . : 2.15.3.84

I want to be able to ping 10.0.0.1 from the dialed in device:


C:\>ping 10.0.0.1

Pinging 10.0.0.1 with 32 bytes of data:

Request timed out.

Ping statistics for 10.0.0.1:
Packets: Sent = 1, Received = 0, Lost = 1 (100% loss),

When I am on the device, I get a response:

[root@Linux ide]# ping 10.0.0.1
PING 10.0.0.1 (10.0.0.1) from 10.0.0.2 : 56 data bytes
64 bytes from 10.0.0.1: icmp_seq=0 ttl=128 time=1.4 ms
64 bytes from 10.0.0.1: icmp_seq=1 ttl=128 time=1.2 ms
64 bytes from 10.0.0.1: icmp_seq=2 ttl=128 time=1.2 ms

--- 10.0.0.1 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 1.2/1.2/1.4 ms

Any suggestions?

bryan <bryanhahn@yahoo.com> wrote:
> When I dial into my Linux device over ppp, I need to be able to access
> devices connected to the eth0 interface. On the device, I can ping
> devices on eth0, but not on the connected device. Here is a summary:

> [root@Linux ide]# cat /proc/sys/net/ipv4/ip_forward
> 1

On the "Linux device," add the pppd option proxyarp and use :10.0.0.x
to specify the peer IP address, where x doesn't belong to any device
on the eth0 LAN. Also make sure the "Linux device" pppd doesn't have
the noipdefault option configured. If there is a special need to use
2.15.3.84 for the dialed-in and 2.15.3.85 for the Linux device on the
PPP link then this won't work.

--
Clifford Kite Email: "echo xvgr_yvahk-ccc@ri1.arg|rot13"
PPP-Q&A links, downloads: http://ckite.no-ip.net/
/* I hear and I forget. I see and I remember. I do and I understand.
--Confucius, 551-479 BC */

On Fri, 4 Feb 2005 18:56:44 -0600, Clifford Kite <kite@see.signature.id> wrote:
> bryan <bryanhahn@yahoo.com> wrote:
>> When I dial into my Linux device over ppp, I need to be able to access
>> devices connected to the eth0 interface. On the device, I can ping
>> devices on eth0, but not on the connected device. Here is a summary:
>
>> [root@Linux ide]# cat /proc/sys/net/ipv4/ip_forward
>> 1
>
> On the "Linux device," add the pppd option proxyarp and use :10.0.0.x
> to specify the peer IP address, where x doesn't belong to any device
> on the eth0 LAN. Also make sure the "Linux device" pppd doesn't have
> the noipdefault option configured. If there is a special need to use
> 2.15.3.84 for the dialed-in and 2.15.3.85 for the Linux device on the
> PPP link then this won't work.

It has always worked for me using eth0 (10.0.0.2 in your case) IP local
pppd IP, another unused LAN IP for remote, with proxyarp pppd option.
You generally would NOT use defaultroute pppd option for the server end of
such a connection (but would typically use it on the remote, if not more
specific routing).

Windows networking has not worked over such a connection for me, even when
I had samba doing wins on the pppd box, unless the name of the remote was
lmhosts. So I scripted some things into /etc/ppp/auth-up & ip-up to
assume the remote machine name was same as login name and plug that into
lmhosts (and remove it in ip-down).

Thanks for the information, I will set the ip address to be on the
local LAN (I.E 10.0.0.6).

You are talking about setting the lmhosts file on the client system to
be the username--> IP address, right? So, if my username was bryan,
then the lmhosts file would be the following:

bryan 10.0.0.6

Have you had better luck doing this on non-windows clients? Any chance
you would be willing to share what you put in your ip-up to set the
remote machine name? When I currently connect, my client machine has
the gateway automatically set to 2.15.3.84. Any reason to think the
automatic setting is wrong?

On 5 Feb 2005 07:38:30 -0800, bryan <bryanhahn@yahoo.com> wrote:
> Thanks for the information, I will set the ip address to be on the
> local LAN (I.E 10.0.0.6).
>
> You are talking about setting the lmhosts file on the client system to
> be the username--> IP address, right? So, if my username was bryan,
> then the lmhosts file would be the following:
>
> bryan 10.0.0.6

Yes assuming netbios (Windows) name of remote machine was bryan and login
name for pppd was bryan.

> Have you had better luck doing this on non-windows clients? Any chance
> you would be willing to share what you put in your ip-up to set the
> remote machine name? When I currently connect, my client machine has
> the gateway automatically set to 2.15.3.84. Any reason to think the
> automatic setting is wrong?

Unfortunately the info maniputlating lmhosts is on an old machine in
storage at work, if not erased when upgraded from RH 5.2 to 6.1. I gave
up on that when I discovered I could simply hang a modem on our Cisco
router, configure it for ppp (proxy arp to LAN) and made the remote IP an
ip-helper-address so it would see LAN netbios (Windows) broadcasts. Until
recently, dialin to our LAN was my boss' only internet access from home.

The basic idea is that /etc/ppp/auth-up is passed the username as 3rd
parameter, so I would plug $3 into a file.

Then from /etc/ppp/ip-up (or ip-up.local), I would grab the username from
the file saved by auth-up and append, that username and $5 to lmhosts.

From ip-down (or ip-down.local) I removed the line from lmhosts that
contained $5 (remote IP). I forget exactly how (I may have used Perl,
since I know that better than shell scripts).

I made the recommended changes and the configuration definitely is
different, but I still cannot ping 10.0.0.1 from the client device.
Here is the updated configuration:


PPP adapter linuxmodem:

Connection-specific DNS Suffix . :
IP Address. . . . . . . . . . . . : 10.0.0.7
Subnet Mask . . . . . . . . . . . : 255.255.255.255
Default Gateway . . . . . . . . . : 10.0.0.7

[root@Linux ide]# ifconfig
eth0 Link encap:Ethernet HWaddr 00:50:15:03:00:62
inet addr:10.0.0.2 Bcast:10.255.255.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:81078 errors:0 dropped:0 overruns:0 frame:0
TX packets:107926 errors:0 dropped:0 overruns:0 carrier:0
collisions:4 txqueuelen:100
Base address:0x300

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:3904 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:0

ppp0 Link encap:Point-to-Point Protocol
inet addr:10.0.0.6 P-t-P:10.0.0.7 Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
RX packets:173 errors:0 dropped:0 overruns:0 frame:0
TX packets:119 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3

[root@Linux ide]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use
Iface
10.0.0.7 * 255.255.255.255 UH 0 0 0
ppp0
10.0.0.0 * 255.255.255.0 U 0 0 0
eth0
127.0.0.0 * 255.0.0.0 U 0 0 0
lo

[root@Linux ide]# arp
Address HWtype HWaddress Flags Mask
Iface
10.0.0.1 ether 08:00:19:00:E1:77 C
eth0
10.0.0.7 * * MP
eth0

Any suggestions?

bryan wrote:
> I made the recommended changes and the configuration definitely is
> different, but I still cannot ping 10.0.0.1 from the client device.
> Here is the updated configuration:
>
>
> PPP adapter linuxmodem:
>
> Connection-specific DNS Suffix . :
> IP Address. . . . . . . . . . . . : 10.0.0.7
> Subnet Mask . . . . . . . . . . . : 255.255.255.255
> Default Gateway . . . . . . . . . : 10.0.0.7
>
> [root@Linux ide]# ifconfig
> eth0 Link encap:Ethernet HWaddr 00:50:15:03:00:62
> inet addr:10.0.0.2 Bcast:10.255.255.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:81078 errors:0 dropped:0 overruns:0 frame:0
> TX packets:107926 errors:0 dropped:0 overruns:0 carrier:0
> collisions:4 txqueuelen:100
> Base address:0x300
>
> lo Link encap:Local Loopback
> inet addr:127.0.0.1 Mask:255.0.0.0
> UP LOOPBACK RUNNING MTU:3904 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:0
>
> ppp0 Link encap:Point-to-Point Protocol
> inet addr:10.0.0.6 P-t-P:10.0.0.7 Mask:255.255.255.255
> UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
> RX packets:173 errors:0 dropped:0 overruns:0 frame:0
> TX packets:119 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:3
>
> [root@Linux ide]# route
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric Ref Use
> Iface
> 10.0.0.7 * 255.255.255.255 UH 0 0 0
> ppp0
> 10.0.0.0 * 255.255.255.0 U 0 0 0
> eth0
> 127.0.0.0 * 255.0.0.0 U 0 0 0
> lo
>
> [root@Linux ide]# arp
> Address HWtype HWaddress Flags Mask
> Iface
> 10.0.0.1 ether 08:00:19:00:E1:77 C
> eth0
> 10.0.0.7 * * MP
> eth0

The tables seem OK, except for (4) below. The ARP line with
the MP flags is for proxy ARP for the Windows client.

1. Did you remember to enable ip forwarding in the Linux box?
(cat /proc/sys/net/ipv4/ip_forward to check).

2. Can you ping the Linux box from the Windows client?

3. Is there a firewall active in the Linux box?
(use iptables -nvL to check)

4. The default route is missing from the routing table above.
If the gateway is 10.0.0.1, try:

route add default gw 10.0.0.1

--

Tauno Voipio
tauno voipio (at) iki fi

bryan <bryanhahn@yahoo.com> wrote:
> I made the recommended changes and the configuration definitely is
> different, but I still cannot ping 10.0.0.1 from the client device.
> Here is the updated configuration:


> PPP adapter linuxmodem:

> Connection-specific DNS Suffix . :
> IP Address. . . . . . . . . . . . : 10.0.0.7
> Subnet Mask . . . . . . . . . . . : 255.255.255.255
> Default Gateway . . . . . . . . . : 10.0.0.7

I don't know what "PPP adapter linuxmodem" means.

> [root@Linux ide]# ifconfig
> eth0 Link encap:Ethernet HWaddr 00:50:15:03:00:62
> inet addr:10.0.0.2 Bcast:10.255.255.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:81078 errors:0 dropped:0 overruns:0 frame:0
> TX packets:107926 errors:0 dropped:0 overruns:0 carrier:0
> collisions:4 txqueuelen:100
> Base address:0x300

> lo Link encap:Local Loopback
> inet addr:127.0.0.1 Mask:255.0.0.0
> UP LOOPBACK RUNNING MTU:3904 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:0

> ppp0 Link encap:Point-to-Point Protocol
> inet addr:10.0.0.6 P-t-P:10.0.0.7 Mask:255.255.255.255
> UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
> RX packets:173 errors:0 dropped:0 overruns:0 frame:0
> TX packets:119 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:3

The local PPP link IP address (for "Linux ide") should be the
same as the IP address for eth0 for proxy arp to work. One way to make
this happen is to use the pppd option

10.0.0.2:10.0.0.7

and don't configure the PPP implementation on the other end to supply
an address for "Linux ide".

-- Clifford Kite Email: "echo xvgr_yvahk-ccc@ri1.arg|rot13"
PPP-Q&A links, downloads: http://ckite.no-ip.net/
/* They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety." Benjamin Franklin */

Your suggestions did the trick. Thanks for all your help!!!!!

Clifford Kite wrote:
> bryan <bryanhahn@yahoo.com> wrote:
> > I made the recommended changes and the configuration definitely is
> > different, but I still cannot ping 10.0.0.1 from the client device.
> > Here is the updated configuration:
>
>
> > PPP adapter linuxmodem:
>
> > Connection-specific DNS Suffix . :
> > IP Address. . . . . . . . . . . . : 10.0.0.7
> > Subnet Mask . . . . . . . . . . . : 255.255.255.255
> > Default Gateway . . . . . . . . . : 10.0.0.7
>
> I don't know what "PPP adapter linuxmodem" means.
>
> > [root@Linux ide]# ifconfig
> > eth0 Link encap:Ethernet HWaddr 00:50:15:03:00:62
> > inet addr:10.0.0.2 Bcast:10.255.255.255
Mask:255.255.255.0
> > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> > RX packets:81078 errors:0 dropped:0 overruns:0 frame:0
> > TX packets:107926 errors:0 dropped:0 overruns:0 carrier:0
> > collisions:4 txqueuelen:100
> > Base address:0x300
>
> > lo Link encap:Local Loopback
> > inet addr:127.0.0.1 Mask:255.0.0.0
> > UP LOOPBACK RUNNING MTU:3904 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:0
>
> > ppp0 Link encap:Point-to-Point Protocol
> > inet addr:10.0.0.6 P-t-P:10.0.0.7 Mask:255.255.255.255
> > UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500
Metric:1
> > RX packets:173 errors:0 dropped:0 overruns:0 frame:0
> > TX packets:119 errors:0 dropped:0 overruns:0 carrier:0
> > collisions:0 txqueuelen:3
>
> The local PPP link IP address (for "Linux ide") should be the
> same as the IP address for eth0 for proxy arp to work. One way to
make
> this happen is to use the pppd option
>
> 10.0.0.2:10.0.0.7
>
> and don't configure the PPP implementation on the other end to supply
> an address for "Linux ide".
>
> -- Clifford Kite Email: "echo
xvgr_yvahk-ccc@ri1.arg|rot13"
> PPP-Q&A links, downloads:
http://ckite.no-ip.net/
> /* They that can give up essential liberty to obtain a little
temporary
> safety deserve neither liberty nor safety." Benjamin Franklin */