W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2014

Re: Client Behavior for SETTINGS_TIMEOUT

From: Patrick McManus <pmcmanus@mozilla.com>
Date: Thu, 19 Jun 2014 09:20:47 -0400
Message-ID: <CAOdDvNoOKis2=nC3UAvmLVnaweyBOT6O4yB4pTpfYBz8XERxwQ@mail.gmail.com>
To: Jason Greene <jason.greene@redhat.com>
Cc: HTTP Working Group <ietf-http-wg@w3.org>
hey Jason -

SETTINGS_TIMEOUT is a connection error - so all streams and the session
need to be torn dorn. There is really nothing to ack at that point.. any
non-idempotent transactions in flight are retried (where possible) on
another connection only if the last-stream-id in the goaway indicates its
safe to do so. Basically its handled exactly like a PROTOCOL_ERROR

-P



On Wed, Jun 18, 2014 at 3:40 PM, Jason Greene <jason.greene@redhat.com>
wrote:

> I was just wondering what client implementers on this list are currently
> doing in the face of a SETTINGS_TIMEOUT. Is it handled gracefully (retry
> after acking which seems the intention)?
>
>
> --
> Jason T. Greene
> WildFly Lead / JBoss EAP Platform Architect
> JBoss, a division of Red Hat
>
>
>
Received on Thursday, 19 June 2014 13:21:13 UTC

This archive was generated by hypermail 2.3.1 : Monday, 9 September 2019 17:48:12 UTC