- From: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
- Date: Thu, 23 Feb 2012 18:27:42 +0000
- To: Zhong Yu <zhong.j.yu@gmail.com>
- Cc: Alvaro Lopez Ortega <alvaro@alobbs.com>, Willy Tarreau <w@1wt.eu>, HTTP Working Group <ietf-http-wg@w3.org>
On 23 Feb 2012, at 18:19, Zhong Yu wrote: > Thank you both. So, "Connection:keep-alive" in 1.0 is very dependable > in practice, server should keep the connection alive. > > If that's the case, shouldn't we remove the alarming languages in > [part 1, A.1.2.]? They might be legitimate concerns in 1999, but no > longer relevant today. > > However, some experimental implementations of HTTP/1.0 > persistent connections are faulty; for example, if a HTTP/1.0 proxy > server doesn't understand Connection, it will erroneously forward > that header to the next inbound server, which would result in a hung > connection. > > ... > > Clients are also encouraged to consider the use of Connection: keep- > alive in requests carefully; while they can enable persistent > connections with HTTP/1.0 servers, clients using them need will need > to monitor the connection for "hung" requests (which indicate that > the client ought stop sending the header), and this mechanism ought > not be used by clients at all when a proxy is being used. Aren't these two different things that you're confusing? Your original question was about how a HTTP/1.1 server should handle the presence of "Connection:" in a HTTP/1.0 request. The text above is warning HTTP/1.0 clients that HTTP/1.0 proxies may not handle the presence of "Connection:" properly. Ben > > > On Thu, Feb 23, 2012 at 11:56 AM, Alvaro Lopez Ortega <alvaro@alobbs.com> wrote: >> On 02/23/2012 05:56 PM, Willy Tarreau wrote: >>> >>> On Thu, Feb 23, 2012 at 10:32:39AM -0600, Zhong Yu wrote: >>>> >>>>> When an HTTP/1.1 server accepts an HTTP/1.0 request, should the server >>>>> always close the connection after sending the response? That seems to >>>>> be the safest way. While the server could try to keep the connection >>>>> by checking for Connection: keep-alive, the spec suggests that it is >>>>> problematic. What's your recommended best practice for this issue? >>> >>> In haproxy we check for "Connection: keep-alive" in 1.0 and >>> "Connection: close" in 1.1 and I've never had any report of >>> unexpected behaviour by doing so. >> >> >> Same thing for the Cherokee Web Server. I believe it's the behavior that >> most of the servers and proxies implement. >> >> All the best, >> Alvaro >
Received on Thursday, 23 February 2012 18:28:07 UTC