I got it!
SPDY breaks as soon as my Upstream uses GZip! I turned GZip off in Railo and
voila it works. Turn it on and get blank page again.
Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,247692,247893#msg-247893
___
nginx mailing list
nginx@
In Railo you can access some environment and client vars within a struct
called cgi.
If I hit my site with plain ssl it shows cgi.server_port = 80. Should be
443, shouldn't it?
Hitting Tomcat directly shows cgi.server_port= which is fine.
Posted at Nginx Forum:
http://forum.nginx.org/read.php
Damn, seems Firefox didn't clear the cache even if I told him to do so.
Worked a while, but now blank page again.
But it has to be the Tomcat backend 'cause PHP passed thru to fastcgi works.
Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,247692,247739#msg-247739
___
On Wednesday 19 February 2014 06:24:41 p.heppler wrote:
> Hmm, just tried adding
> proxy_set_header If-none-Match $http_if_none_match;
> to my config and now it works with spdy to.
Probably it works in this case because browser just shows to you cached page,
that was retrieved earlier over http o
Hmm, just tried adding
proxy_set_header If-none-Match $http_if_none_match;
to my config and now it works with spdy to.
Why do I need to set this manually when using spdy, but not on "normal"
http/https?
Weird...
Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,247692,247723#msg-247723
I checked Tomcat logs. No errors, just status 200 and zero length. I
appended %v - Local server name to the log to see if $host is passed thru
correct. It is.
And now it's getting weird. I replcaed my complex website (which runs fine
with pure ssl!) with a simple "Hello World" and it works!
Seems
Strange. As said with pure ssl it works, it's just the spdy part. I'll check
Tomcat logs.
Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,247692,247720#msg-247720
___
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo
On Tuesday 18 February 2014 09:54:40 p.heppler wrote:
> Another run with Tomcat, fresh log:
[..]
In both cases your backend returned a "200 OK" response
with empty body:
2014/02/18 15:46:13 [debug] 12839#0: *2 http proxy status 200 "200 OK"
2014/02/18 15:46:13 [debug] 12839#0: *2 http proxy heade
Another run with Tomcat, fresh log:
2014/02/18 15:58:53 [debug] 12993#0: epoll add event: fd:9 op:1 ev:2001
2014/02/18 15:58:57 [debug] 12993#0: post event 7F19759D4078
2014/02/18 15:58:57 [debug] 12993#0: delete posted event 7F19759D4078
2014/02/18 15:58:57 [debug] 12993#0: accept on
I meanwhile replaced Tomcat with Jetty, because Jetty also supports SPDY.
But same result.
Posted at Nginx Forum:
http://forum.nginx.org/read.php?2,247692,247696#msg-247696
___
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinf
I just created a new log. This is from nginx restart to first request:
2014/02/18 15:45:56 [debug] 12839#0: epoll add event: fd:9 op:1 ev:2001
2014/02/18 15:46:12 [debug] 12839#0: post event 7F29CC9D1078
2014/02/18 15:46:12 [debug] 12839#0: delete posted event 7F29CC9D1078
2014/02/18 1
On Tuesday 18 February 2014 09:15:25 p.heppler wrote:
> Hello,
> I compiled nginx 1.5.10 on Centos 6.5 and try to use it as frontend for
> Tomcat.
> If I use pure SSL everything works fine. But as soon as I enable SPDY, I
> only get a blank page.
> And Content-Length Header is 0, but HTTP status co
Hello,
I compiled nginx 1.5.10 on Centos 6.5 and try to use it as frontend for
Tomcat.
If I use pure SSL everything works fine. But as soon as I enable SPDY, I
only get a blank page.
And Content-Length Header is 0, but HTTP status code is 200.
But, the blank page only affects request which are hand
13 matches
Mail list logo
