lwip-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lwip-users] eCos - PPP - TCP - LwIP


From: Jonathan Larmour
Subject: Re: [lwip-users] eCos - PPP - TCP - LwIP
Date: Mon, 19 Feb 2007 17:30:09 +0000
User-agent: Thunderbird 1.5.0.9 (X11/20070102)

Kieran Mansley wrote:
On Mon, 2007-02-19 at 07:16 -0800, Ashish Anand wrote:
[1] No issue of firewall, since I used "service iptables stop".

Ok, sigh. I go back to my other suggestion of more extensive debug I'm afraid.

[2] When I telnet, I get something similar:

ip_input: iphdr->dest 0xfb0000e0 netif->ip_addr 0x401a8c0 (0xe0,
0x1a8c0, 0xfb00000)ip_input: iphdr->dest 0xfb0000e0 netif->ip_addr 0x100007f (0xe0, 0x7f, 0xfb000000) ip_input: packet not for us. ip_forward: not bouncing packets back on incoming interface.
Except that this time, the (hex) IP address is 192.168.1.4 (board IP)
instead of the MDNS IP.

I don't see any different to the previous ip_input debug output you posted.

So even these packets are considered "not for us"!

It's really strange, why all this complication is arising!

I may have got confused here, but let me summarise what I think is
happening:
 - you receive two packets.
 - the first is for a multicast IP address.
 - the second is a SYN to the board IP address.
 - both are rejected as "not for us"

The multicast packet is rejected because it is multicast and we don't
support generic multicast traffic (there are some special cases where it
is supported, but on the whole not).

The second can only be rejected with that error because your board does
not have the IP address you think it does.  The debug statement above
shows the packet's IP address to be 0xfb0000e0 (i.e 14.0.0.251) and the
netif/board IP address to be  0x100007f (ie. 127.0.0.16).

/me passes Kieran a new hex converter ;-).

0xfb0000e0 == 224.0.0.51.
0x100007f == 127.0.0.1.

ip_input is making debug output for each netif interface. So:

ip_input: iphdr->dest 0xfb0000e0 netif->ip_addr 0x401a8c0 (0xe0, 0x1a8c0,
0xfb00000)

means the destination is the multicast addr, but the netif addr is 0x401a8c0 == 192.168.1.4, which must be the PPP interface.

The second line:
ip_input: iphdr->dest 0xfb0000e0 netif->ip_addr 0x100007f (0xe0, 0x7f, 0xfb000000) is ip_input's attempt to match the multicast addr against the loopback interface.

These lines are just ip_inputs attempts to direct the (single) multicast packet, which it fails to do.

The real problem appears to be that the telnet SYN is never even seen.

Jifl
--
eCosCentric    http://www.eCosCentric.com/    The eCos and RedBoot experts
 Meet us at Embedded Systems Conference - SV, April 3rd-5th.  Stand #2248
Company legal info, address and number:   http://www.ecoscentric.com/legal
------["The best things in life aren't things."]------      Opinions==mine




reply via email to

[Prev in Thread] Current Thread [Next in Thread]