[Click] question about test-tun.click
Eddie Kohler
kohler at cs.ucla.edu
Thu Jul 8 09:42:59 EDT 2004
Xavier Brouckaert wrote:
> Hi,
>
> I'm learning click and I'd like to use it. In test-tun.click, I see this
> : ( http://www.pdos.lcs.mit.edu/click/ex/test-tun.html )
> ch[0] -> IPPrint(tun-ok)
> -> IPFilter(allow icmp type echo)
> -> ICMPPingResponder
> -> IPPrint(tun-ping) (*)
> -> tunq;
>
> gav at osiris:~/Coding/click-1.4.1/conf$ sudo click test-tun.click
> tun-ok: 1089295895.570164: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 1)
> tun-ok: 1089295896.580096: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 2)
> tun-ok: 1089295897.580098: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 3)
> tun-ok: 1089295898.580136: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 4)
> tun-ok: 1089295899.580051: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 5)
> tun-ok: 1089295900.580052: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 6)
> tun-ok: 1089295901.580228: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 7)
> tun-ok: 1089295902.580085: 1.0.0.1 > 1.0.0.2: icmp: echo reply (0, 8)
>
> My question is : why do i never see a tun-ping message which should
> correspond to the (*) line in the code ?
Hi Xavier,
Note the IPFilter two lines above. This only lets "icmp echo" packets through
to the ICMPPingResponder. The tun-ok lines are all "icmp echo reply" packets.
So why aren't any ICMP echoes getting to the IPFilter? I believe this is
expected behavior. The Click configuration is generating the ICMP echoes with
ICMPSendPings/ICMPPingSource; the kernel doesn't send packets _generated_ by a
particular tunnel back to that tunnel.
Eddie
>
> I know that it works OK with tethereal :
> gav at osiris:~$ sudo tethereal -i tun0
> Capturing on tun0
> 0.000000 1.0.0.2 -> 1.0.0.1 ICMP Echo (ping) request
> 0.000025 1.0.0.1 -> 1.0.0.2 ICMP Echo (ping) reply
> 1.000043 1.0.0.2 -> 1.0.0.1 ICMP Echo (ping) request
> 1.000067 1.0.0.1 -> 1.0.0.2 ICMP Echo (ping) reply
> 1.999999 1.0.0.2 -> 1.0.0.1 ICMP Echo (ping) request
> 2.000025 1.0.0.1 -> 1.0.0.2 ICMP Echo (ping) reply
>
> So, I don't understand why the first IPPrint works and not the second.
>
> Thank you for shedding light on this,
> Xavier
>
> _______________________________________________
> click mailing list
> click at amsterdam.lcs.mit.edu
> https://amsterdam.lcs.mit.edu/mailman/listinfo/click
More information about the click
mailing list