Re: [RRG] Small comment on draft-jen-apt-00.txt

[Lixia Zhang <[e-mail]>]

On Oct 15, 2007, at 9:10 AM, Michael R Meisel wrote:

>>>> And the standard question: how will this work with SCTP?
>>>
>>> APT will be invisible to end users. This means that it should not  
>>> have any effect on any transport-layer protocol.
>> But SCTP doesn't *want* that invisibility, because it explicitly  
>> chooses
>> to change to an alternative address in order to use an alternative  
>> path.
>> How do you allow SCTP to do its job?

my apology for this belated comment, but a clarification might be  
worthwhile even though it is late.  Michael's comment below is  
exactly right:

- a SCTP node can have multiple IP addresses only when it has
   multiple interfaces (physical or virtual).

- i.e. it is the host-multihoming that allows a SCTP node
   to "explicitly chooses to change to an alternative address".

- A SCTP node, as far as I understand (as co-author of the first SCTP  
spec), does not have a power to choose alternative paths out of  
*site* multihoming--which is what we are talking about here--if those  
paths are not reflected in the node's choice of its IP addresses.



> The same way BGP allows SCTP to do its job today. If you try a  
> different address, you are just as likely to get a different path  
> through the network with APT in place are you are without it. When  
> I say that APT is invisible, I mean that end hosts shouldn't be  
> able to tell whether it's being used or not -- everything should  
> work the same as it does today from their perspective.
>
> -Michael



--
to unsubscribe send a message to [e-mail] with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg