Re: comment on ULE draft (bridging mode)

[Gorry Fairhurst <gorry@erg.abdn.ac.uk>]

Tarif.Zein-Alabedeen@space.alcatel.fr wrote:

> This concerns §4.7.5 : bridged frame SNDU encapsulation ( fiugres 8 and 9)
> :
> The type field value in the figures is set to 0x0001.
Yes

> This indicates that
> the payload is an ethernet frame 
Yes

> with preserved FCS
The Ethernet frame should be sent using ULE *without* an Ethernet-level 
FCS as described in:

     "The Encapsulator MUST check the
       LAN-FCS value of all frames received, prior to further processing. 
       Frames received with an invalid LAN FCS MUST be discarded. After
       checking, the LAN FCS is then removed (i.e., it is NOT forwarded in
       the bridged SNDU)."

http://www.ietf.org/internet-drafts/draft-fair-ipdvb-ule-02.txt

Is this the sort of behaviour that you desire?

> This leads to a redundant CRC : that of the Ethernet frame and that of the
> SNDU
>
> In Eth/AAL5, it is useual to remove the FCS field from the Eth frame before
> encapsulating it in AAL5 PDU since AAL5 makes its own checksum
> This may also be a good thing for ULE.
>
> The coding for the type field when the FCS is not preserved is 0x0007
> (instead of 0x0001)
What do you mean?

> regards
>
>
> ALCATEL SPACE
> DRT/RST  --  Ingénieur Systèmes
> Tel : 0534356918  /  Fax : 0534355560
> Porte : W.220
>
>  
Best wishes,

Gorry Fairhurst.