https://bz.apache.org/bugzilla/show_bug.cgi?id=63855
--- Comment #17 from Konstantin J. Chernov <k.j.cher...@gmail.com> --- > Could you please test it so that I can propose it for 2.4.x? Looks like everything is working fine with v3. Thanks a lot! :) > This may be what you want (since mod_security prevents 100-continue > forwarding anyway), but it kind of defeats the purpose of 100-continue IMHO. Well, we are not using mod_proxy_http on his own, but only in pair with mod_proxy_balancer, so passing 100-continue through seems a bit weird for me, as apache always did his own 100-continue check, choosing another backend in case of failure. So from my perspective turning the "Proxy100Continue off" in our case is pretty much OK. > Another option is to "SetEnv proxy-sendchunks" as you noticed, but it > requires that the backends supports chunk encoding in requests (shouldn't be > an issue these days). I'd prefer not to use chunked requests/responses at all. It's harder to read/debug (using wireshark, etc.), and we had some issues a few years earlier when the client was sending a plain request over the socket, regexp-parsing the answer, therefore expecting a well-known formatted response with content-length set, and suddenly the response to the client went from "Content-Length: xxx" to chunks. When we are talking about the browsers as the clients, it might be better/faster/etc., as well as those trendy things like HTTP/2, but when handling M2M requests, the only thing they can give us - is a headache :) -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: bugs-unsubscr...@httpd.apache.org For additional commands, e-mail: bugs-h...@httpd.apache.org
