Hello Yves,
interesting protocols you plan to implement.
HighSpeed IO is not an easy thing, also not for capturing.
There is one thing you should keep in mind with such high speed communication.
I know from my experience with PNIO protocol, that for example the NDIS driver
on windows is not able to capture all packets if you have small ethernet
frames on a 100Mbit network that is reaching full load.
The CPU is not able to parse the packets headers that quickly and you will get
lost packets.
The same problem also occurs on other OSs that do a software based capturing.
The only real solution for that seams to be to use a hardware based capturing.
Store the data in libcap format, so that you can open it afterwards in
wireshark to analyze the data.
But relax, as long as you don't test full load szenarios you will not have
that troubles.
Also keep in mind that you need two network cards if want to do a full duplex
capture. You can use an Ethernet TAP device for that and record the data with
two network cards.
On Linux you have a nice feature for capturing on two cards. You can capture
on a "Pseudo-device that captures on all interfaces: any". This way the data
of both cards get merged together according to their timestamps.
On Monday 16 July 2007 22:05, Yves Geissbühler wrote:
> Hi everyone,
>
> as suggested by the Wireshark Developer Guide I would like to
> announce my intension to write several dissectors to add support for
> the protocols MPA[1], DDP[2] and RDMA[3]. You may also find further
> information under [4].
> If anyone is already working on this please let me know. Thx.
>
> -- Yves
>
> [1] http://www.ietf.org/internet-drafts/draft-ietf-rddp-mpa-08.txt
> [2] http://www.ietf.org/internet-drafts/draft-ietf-rddp-ddp-07.txt
> [3] http://www.ietf.org/internet-drafts/draft-ietf-rddp-rdmap-07.txt
> [4] http://www.zurich.ibm.com/sys/servers/rdma.html
> _______________________________________________
> Wireshark-dev mailing list
> Wireshark-dev@xxxxxxxxxxxxx
> http://www.wireshark.org/mailman/listinfo/wireshark-dev
--
mit freundlichen Grüßen / best regards
Gerhard Gappmeier
ascolab GmbH - automation system communication laboratory
Tel.: +49 9131 691 123
Fax: +49 9131 691 128
Web: http://www.ascolab.com
GPG-Key: http://www.ascolab.com/gpg/gg.asc
