Yichun Zhang
Yichun Zhang
@surfire91 Yeah, I know :)
Yeah, I agree this is useful though we're really moving to cosockets these years. I don't mind merging a modified and rebased version of this PR if it does not...
@techzhou HTTP 2's in-connection multiplexing might be problematic for such pulling techniques. But we'll have some kind of compromises similar to NGINX's non-buffered request handling in HTTP 2.
@techzhou Do you want to contribute a patch for that? :)
@weichangdong Apparently it was HTTP/2 and the error message clearly indicates the lack of support for HTTP/2. It has nothing to do with traditional https.
First of all, thank you for your contribution! But I can see that there are some minor issues in your patch: 1. could you strictly follow the Lua coding style...
@zhaojp0411 you can take a look at my .vimrc file here: http://agentzh.org/misc/.vimrc
@hongliang5316 This is awesome! I'll have a look as soon as I can manage. BTW, you can often get better performance in a whole OpenResty application by adding the following...
@azuryyu The standard `$request_body` variable of the nginx core won't read large request bodies buffered on the disk (by the nginx core). See https://github.com/openresty/echo-nginx-module#echo_request_body
@azuryyu For meaningful values in `$request_body`, you should configure `client_body_buffer_size` as well. It's more about the nginx core's `$request_body` behavior. I'm tired of educating people on standard nginx behaviors. alas.