On Wed, Apr 17, 2002 at 01:18:11PM -0700, Lawrence, Martha wrote:
> Does anyone know of any plans underway to make Ethereal decodes for MPLS
You mean like the decoding done in the "packet-mpls.c" file in the
Ethereal source? :-)
Or are you thinking of something else?
> or Fibre Channel?
Well, there's
http://www.tcpdump.org/lists/workers/2002/03/msg00015.html
where Don Lee of Cray says
I'm hacking IP over Fibre Channel into ethereal, and hence
need to tweak libpcap. I need a number for the DLT_ type. (for
the return from pcap_datalink() ) Do you guys have a number
already in mind? Prior work/art?
Don, is that going to be raw Fibre Channel, in the sense of what I infer
from Friedhelm Schmidt's "The SCSI Bus & IDE Interface, Second Edition"
is called "FC-2" (in a diagram, he calls it "Signalling protocol/frames,
flow control"), which I assume is the stuff with frames whose headers he
diagrams as having the form
+--------------+--------------------------------------------+
| Routing | Destination ID |
+--------------+--------------------------------------------+
| Reserved | Source ID |
+--------------+--------------------------------------------+
| Type | Frame control |
+--------------+--------------+-----------------------------+
| Sequence ID | Data control | Sequence number |
+--------------+--------------+-----------------------------+
| Originator exchange ID | Responder exchange ID |
+-----------------------------+-----------------------------+
| Parameters |
+-----------------------------+-----------------------------+
or is the initial header of the raw frame data going to be something
else, e.g. some IP-specific header, without showing the raw FC header?
(If the latter, I should rename the DLT_ value I gave you from DLT_FC to
DLT_FC_OVER_IP or something such as that, so that DLT_FC could be used
for that kind of raw FC stuff.)
