I believe we should have overload requirements but we are mandating too much . Different from some other IETF protocols, we have years and years of experience with Web and P2P trackers to help our decision. HTTP 503 is working fine.
Therefore a simple requirement that can be met with "The server is currently unavailable (because it is overloaded or down for maintenance)" is fine. I'm even okay with Retry-After as long as it is optional (MAY) as in http://tools.ietf.org/html/draft-ietf-httpbis-p2-semantics-15#page-35 Thanks, Reinaldo On 7/17/11 1:02 PM, "Sebastian Kiesel" <ietf-a...@skiesel.de> wrote: > Do you tink the wording is bad or that we are mandating too much, even > in the latest version (11), which explicitly states that a retry-after > based mechansim fulfills the req? _______________________________________________ alto mailing list alto@ietf.org https://www.ietf.org/mailman/listinfo/alto
