Roberto, can you be more specific about how the proxy and server
interact to redirect traffic using Alt-Svc? Since in this case the
client is speaking HTTP/1.1, are you suggesting that the proxy treat the
Alt-Svc as a hop-by-hop feature? My understanding is that this is an
end-to-end feature.
On 04/10/2014 05:08 PM, Roberto Peon wrote:
> It helps Daniels case if the server is smart enough to begin doing
> alt-svc redirecting before it goes down (i.e. it enters a period where
> it shifts traffic away while processing the traffic it had been given
> previously).
> Of course, I could be misinterpreting, but this is the retry-failure
> case that I'm familiar with!
> -=R
>
>
> On Thu, Apr 10, 2014 at 5:02 PM, Martin Thomson
> <martin.thomson@gmail.com <mailto:martin.thomson@gmail.com>> wrote:
>
> On 10 April 2014 16:50, Roberto Peon <grmocg@gmail.com
> <mailto:grmocg@gmail.com>> wrote:
> > I'll note that REFUSED_STREAM doesn't actually solve the problem
> entirely,
> > since it does require the user (not just the proxy) to retry.
>
> Indeed. But each hop is given that choice (retry this request, or
> fail back to my client).
>
> > It is hoped that alt-svc (potentially to the same name) will
> give us the
> > ability to have the client (proxy or not) reconnect to a new
> server (and
> > thus there should be no need for a LAME_DUCK given that mechanism).
>
> I don't think that helps Daniel's case, just the more general one.
>
>