Re: [Linuxptp-devel] ptp4l wrongly takes padding bytes as TLV?

[Rafaël C. <fu...@vi...>]

Hello,

Since I see the subject is ethernet padding I would like to mention what
we use together with PTP:

A timestamping card which adds 9 bytes of VSS-monitoring trailer to the
ethernet packet.

Depending on the exact value of the timestamp, wireshark can misdetect
it and show a "wrong" FCS plus 5 bytes padding, see
https://github.com/boundary/wireshark/blob/master/epan/dissectors/packet-vssmonitoring.c#L139

I'm attaching a small capture where you can see traffic from 2 NICs, one
which adds a trailer to all incoming packets and one which does not.
I am anonymizing the MAC vendor for the timestamping card.

On 31/01/2019 19:24, Jiri Benc wrote:
> On Thu, 31 Jan 2019 16:28:30 +0000, Vincent Li X wrote:
>> we might also need to check again m->header.messageLength is bigger than
>> cnt.
> 
> This might not be a bad idea; if the packet length is inconsistent with
> the PTP or 802.3 standard, a warning can be emitted and the packet
> dropped and not processed further. This would make linuxptp more robust
> against randomly malformed packets and would give some indication to
> users that the other end or intermediate network equipment is broken.
> 
>  Jiri