Hi Wendy,

On Monday, December 2, 2013, Wendy Roome wrote:

> It's been quiet lately, so I thought I'd wake you up for the holidays!
>
> {10.4.2} of draft 21 defines the address types "ipv4" and "ipv6", but
> leaves the door open for extensions to define additional address types.
>
> The Network Map service {11.2.1} returns the prefixes in each PID grouped
> by address type, as in
>
>        "PID2": { "ipv4": ["198.51.100.128/25"] }
>
> What should a client do if the ALTO server returns an address type that
> the client does not recognize? As far as I can tell, the protocol doesn't
> say. If a client completely rejects a Network Map because it has an
> unknown address type, that will make it difficult to add extensions.
>
> {8.3.8} does say, "ALTO implementations MUST ignore unknown fields when
> processing ALTO messages." That could be interpreted as meaning "the
> client should ignore unknown address types." But you could also interpret
> that as only applying to JSON dictionaries whose field names are defined
> by the protocol, such as "meta" and "capabilities".
>
> So what about explicitly saying that clients should ignore unknown address
> types in the Network Map?  Perhaps by adding the following paragraph to
> the end of {11.2.1.6}:
>
> "An ALTO client MUST ignore any EndpointAddressGroup whose address type it
> does not recognize. If as a result a PID does not have any address types
> known to the client, the client still MUST recognize that PID name as
> valid, even thought the client does not know what endpoints it contains."


My understanding of "ignoring unknown fields" includes unknown address
types. The
revised para looks good to me as a clarification. We will add it tomorrow,
if no objection
from others before then.

Richard

>
>         - Wendy Roome
>
>
>
>
>
> _______________________________________________
> alto mailing list
> [email protected] <javascript:;>
> https://www.ietf.org/mailman/listinfo/alto
>


-- 
-- 
 =====================================
| Y. Richard Yang <[email protected]>   |
| Professor of Computer Science       |
| http://www.cs.yale.edu/~yry/        |
 =====================================
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to