Hi there!
read reply post pls
Sas
There is an error in my error.log after restarting nginx mainline. It has not been there before and it is related to the plugin push (???) that seems to be part of nginx. I googled the error message and found an old nginx github conversation about this. Here is the error message:
2016/01/21 13:20:27 [info] 25697#25697: Using 32768KiB of shared memory for push module in /etc/nginx/nginx.conf:114
And as I said it is only occurring when nginx is being restarted. Here is the bug tracker info I found about this: https://bugs.launchpad.net/nginx/+bug/1216817
It seems to be related to fastcgi cache incompatiblities. @Prabuddha do you know anything about this? I posted the error-log entry with nginx config here at the github issue thread that is run by push module coder: https://github.com/slact/nchan/issues/83
I am using fastcgi caching in my customer test sites - before live deployment. Those sites are having weird issues since upgrading to mainline.
And another question about http2: Is it possible to use http2 without SSL? One of my customer sites needs iframe implementation that can’t be delivered via https… so in https websites that iframe will only contain an empty white space… is there a way to use http2 without ssl?
EDIT: Answering my own question here:
Does HTTP/2 require encryption?
No. After extensive discussion, the Working Group did not have consensus to require the use of encryption (e.g., TLS) for the new protocol.
However, some implementations have stated that they will only support HTTP/2 when it is used over an encrypted connection, and currently no browser supports HTTP/2 unencrypted.
Thanks!!!
The developer of th nginx push module states the following:
It looks like you’re using Debian’s nginx-extras package, which bundles the Nginx HTTP Push module, an old version of Nchan. The logged message is not an error.
However it is still spamming my error.log and it is confusing… How can I turn this off?
Thanks!