[rbridge] TTL only - was RE: New fields in shim header?

[Joe Touch]

Silvano Gai wrote:
> Joe,
> 
>> -----Original Message-----
>> From: Joe Touch [mailto:touch at ISI.EDU]
>> Sent: Thursday, October 19, 2006 10:08 AM
>> To: Silvano Gai
>> Cc: Caitlin Bestler; Erik Nordmark; Ed Bowen; rbridge at postel.org;
> rbridge-
>> bounces at postel.org; Radia Perlman
>> Subject: Re: [rbridge] TTL only - was RE: New fields in shim header?
>>
>>
>>
>> Silvano Gai wrote:
>>> Caitlin,
>>>
>>> ....
>>>
>>>> For the same reason Bernard Aboba cited on wireless. An set of
>>>> RBridges can distribute the new location information more
>>>> efficiently than existing bridges.
>>> When a VM move the first thing it does is an ARP, i.e. it sends a
>>> broadcast packet.
>> ...
>>> I doubt that in this aspect TRILL is more efficient that .1D, or am
> I
>>> missing something?
>> There's no rule to force an ARP upon move, but that would affect all
> 802
>> nets, including TRILL ones, equally badly.
>>
> 
> I don't understand the last sentence, why does an ARP affect 802 nets
> badly?

Nodes that move and do not do ARP immediately affect 802 nets and TRILL
badly. You noted that 'most commercial nodes do this' (paraphrased), but
it's not a requirement.

Joe

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 250 bytes
Desc: OpenPGP digital signature
Url : http://mailman.postel.org/pipermail/rbridge/attachments/20061019/f2be99c0/signature.bin