Re: [alto] Comments on draft-kiesel-alto-reqs-01

[Laird Popkin <laird at pando.com>]

From: "stefano previdi" <sprevidi at cisco.com>

>> LP: The only requirement (IMO) from the app side regarding ALTO availability
>> is that if ALTO cannot respond to a request that it should do so in an easily
>> identifiable way so that the app doesn't wait "forever" for a response that
>> it's not going to get. (e.g. reply with a 'busy' message, rather than silently
>> not responding) :

>well, the cause of the inability to respond may also prevent the ALTO server
>to send anything back...

Certainly true, as a failure case to handle. But it'd be best (IMO) for the spec to define a failure message and encourage it being sent, to improve the odds of not having to wait for timeouts.

> s.

> 
> - Laird Popkin, CTO, Pando Networks
>   mobile: 646/465-0570
> 
> ----- Original Message -----
> From: "Roni Even" <ron.even.tlv at gmail.com>
> To: alto at ietf.org
> Sent: Thursday, November 13, 2008 1:03:49 PM (GMT-0500) America/New_York
> Subject: [alto] Comments on draft-kiesel-alto-reqs-01
> 
> 
> 
> 
> Hi, 
> 
> I read the draft and have some comments on section 4.2.
> 
>   
> 
> . 
> 
> 
>   
> 
> Req 9 is suggested a solution that should be in other document
> 
>   
> 
> Req 10 I suggest that it should say that a client shall not overload the
> server, the current text is normative text.
> 
>   
> 
> Req 11 is normative text and not a requirement.
> 
>   
> 
> Req 12 the first part " An ALTO server, which is operating close to its
> capacity limit, SHOULD be able to inform clients about its impending overload
> situation" is the requirement the rest is normative text.
> 
>   
> 
> Thanks 
> 
> Roni Even 
> 
> 
> _______________________________________________ alto mailing list
> alto at ietf.org https://www.ietf.org/mailman/listinfo/alto
> _______________________________________________
> alto mailing list
> alto at ietf.org
> https://www.ietf.org/mailman/listinfo/alto



_______________________________________________
alto mailing list
alto at ietf.org
https://www.ietf.org/mailman/listinfo/alto