- From: Jason Greene <jason.greene@redhat.com>
- Date: Fri, 27 Jun 2014 13:41:55 -0500
- To: Martin Thomson <martin.thomson@gmail.com>
- Cc: Michael Sweet <msweet@apple.com>, Greg Wilkins <gregw@intalio.com>, Mark Nottingham <mnot@mnot.net>, HTTP Working Group <ietf-http-wg@w3.org>
On Jun 27, 2014, at 1:33 PM, Martin Thomson <martin.thomson@gmail.com> wrote: > On 27 June 2014 11:31, Jason Greene <jason.greene@redhat.com> wrote: >> Even if you set table size to 0 you still have to process them before you can return a 413, since the setting is delayed. and you might get them before it takes effect. Only way to not process them is GOAWAY + close. > > A table of size zero means that you don't carry any state forward, so > it's safe to discard the frames. Only when it is in effect. A client can connect to the origin and send a ton of continuation frames, followed by additional requests at the same time as the server is sending the setting to set the table size to 0.Thos additional requests can’t be handled because they contain state relative to the CONTINUATION. I suppose the server could reply with a 503 or something. -- Jason T. Greene WildFly Lead / JBoss EAP Platform Architect JBoss, a division of Red Hat
Received on Friday, 27 June 2014 18:42:28 UTC