Re: libpcap and tc filters

[Adam Katz <adamkatz0@gmail.com>]

well, first of all, thanks A LOT for your effort.

second, I just took at the libpcap source code and it seems it's using
the same ETH_P_ALL option when binding to an interface. So based on
what you're saying, the same solution of patching libpcap and
replacing ETH_P_ALL with  ETH_P_IP should also make these rules work
with traffic sent using pure libpcap or any libpcap - based
application.

On Tue, Jul 5, 2011 at 3:47 PM, jamal <hadi@cyberus.ca> wrote:
> On Tue, 2011-07-05 at 06:56 -0400, jamal wrote:
>
>> I downloaded tcpreplay and reproduced the issue with your rules.
>> Will look into it..
>
> Ok - found out whats going on.
> tcprelay sendpacket_open_pf() does bind to ETH_P_ALL.
> You are sending IP packets (the name tcpreplay is misleading,
> this thing replays anything).
> Your filters are for ip packets as in:
> ---
> sudo tc filter add dev eth0 protocol ip parent 1: prio 1 u32 match ip
> dport 22 0xffff flowid 1:1
> ---
>
> You have two options:
>
> 1) If you change that to capture ETH_P_ALL it works.
> i.e
>
> ---
> sudo tc filter add dev eth0 protocol all parent 1: prio 1 u32 match ip
> dport 22 0xffff flowid 1:1
> ---
>
> Of course this is nasty if you are in a busy network, because _all_
> packets not just ip will look at your filters. If it is just an
> experimental setup, it may be a non-issue
>
> 2) Change tcpreplay to take an additional option so it binds to
> ETH_P_IP (and default stays as is today). The authors of the app
> may not like that option - but it is sensible if you know you are
> replaying ip packets.
>
> cheers,
> jamal
>
>
>
>