- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Thu, 19 Jun 2014 12:18:19 -0700
- To: "Richard Wheeldon (rwheeldo)" <rwheeldo@cisco.com>
- Cc: Poul-Henning Kamp <phk@phk.freebsd.dk>, Mark Nottingham <mnot@mnot.net>, HTTP Working Group <ietf-http-wg@w3.org>
On 19 June 2014 11:20, Richard Wheeldon (rwheeldo) <rwheeldo@cisco.com> wrote: > I've been working on the assumption that SETTINGS_MAX_CONCURRENT_STREAMS applies only to a single connection up to now. > I'm not sure how else it makes sense. That's definitely the intent here. A distributed system like your own can really only apply any constraints on each individual node separately. The question is: what recourse does a client have if they find themselves exceeding this limit frequently or for long periods of time. And proxies are likely to find themselves in this situation more easily than others, simply because they are making more requests. It might just come down to detecting and whitelisting proxies so that they get more connections or more concurrent streams.
Received on Thursday, 19 June 2014 19:18:47 UTC