[P4-dev] Parsing wireshark trace in p4
antonin at barefootnetworks.com
Tue Nov 8 12:16:10 EST 2016
If you mean run a pcap trace through the behavioral-model software switch,
then you may be able to use the --use-files option.
Assuming you have installed the behavioral-model (
https://github.com/p4lang/behavioral-model), it may look something like
simple_switch --log-console --use-files 10 <path to your P4 JSON file> -i
1 at iface1 -i 2 at iface2
If you run simple_switch like this, it will assume 2 interfaces (port
numbers 1 and 2). For port 1, input packets will be read from
iface1_in.pcap and outgoing packets written to iface1_out.pcap. For port 2,
it will be respectively iface2_in.pcap and iface2_out.pcap. So whatever you
put after the '@' in the command line is simply used as a prefix for the
pcap files name. So obviously, your file names need to match what you pass
on the command line.
Finally, the integer value following --use-files is the time in seconds
that the model will wait before it starts processing the packets in your
_in pcap files. This is convenient if you need to install table entries by
running a separate script.
All this information can be found (in a more succinct form) by running
On Tue, Nov 8, 2016 at 8:49 AM, shubham bhardwaj <
bhardwajshubham2011 at gmail.com> wrote:
> I have a small query, I want to parse a wireshark trace of a protocol. so
> how i can give input of trace in switch.
> Thanks & Regards
> Shubham Bhardwaj
> *Technical University of Darmstadt*
> +49 17641749766
> Sent with Mailtrack
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the P4-dev