Re: [hybi] #4: handshake does not work properly with HTTP reverse proxy.

["hybi issue tracker" <[e-mail]>]

#4: handshake does not work properly with HTTP reverse proxy.
-------------------------------------------+--------------------------------
 Reporter:  salvatore.loreto@…             |       Owner:     
     Type:  defect                         |      Status:  new
 Priority:  critical                       |   Milestone:     
Component:  thewebsocketprotocol           |     Version:     
 Severity:  Active WG Document             |    Keywords:     
-------------------------------------------+--------------------------------
Changes (by sm+ietf@…):

 * cc: ian@… (removed)
 * cc: ifette@… (added)


Comment:

 Comment posted by Salvatore Loreto on 17 Aug, 2010:
 there has been a lot of mails and good technical discussion about
 - the need of "8 random bytes" for security reasons
 - the fact that as the "8 random bytes" are used  in the 00 wg version
 they do not work for reverse proxies and are non-HTTP-compliance,
 - the possibility to use the "8 random bytes" in a way that both work with
 reverse proxies and are HTTP-compliance among the others:
 inserting the 8 random bytes in an header
 using the Willy Tarreau analysis/proposal explained

 http://www.ietf.org/mail-archive/web/hybi/current/msg03315.html

-- 
Ticket URL: <http://trac.tools.ietf.org/wg/hybi/trac/ticket/4#comment:3>
hybi <http://tools.ietf.org/hybi/>
The Hypertext-Bidirectional (HyBi) working group will seek
standardization of one approach to maintain bidirectional
communications between the HTTP client, server and intermediate
entities, which will provide more efficiency compared to the current
use of hanging requests.