It soulds like ab is succeeding in generating a load high enough to impact performance already. You need to figure out where the bottlenecks are. My message is intended as a helpful hint that you should not trust the stats from ab. You need to be measuring perf on the nginx and backend servers. Start by looking at the elapsed time in nginx and the response times of the backends.by bryndole - Nginx Mailing List - English
At that level of concurrency you are only testing the throughput of ab and not nginx. Apache bench is a single process and is too slow to test anything other than a single Apache instance. In summary, ab is a piece of crap.by bryndole - Nginx Mailing List - English
Our own investigation indicates that, no, there is not way to directly access the values in the limit_conn zones from "user space." We created a new status page that reports all the existing zones and the current connection counts for each, as well as what the current limits are. It does this by walking the config data structure and building a list of the zones and then creating a simby bryndole - Nginx Mailing List - English
I'm looking for a way get the value of the current number of connections in a given limit_conn zone.There is no variable access that I can find. I'm considering hacking this into the stub_status module, so that all of the active connection zones are displayed with the current number of active connections. Is this value accessible via the lua or perl interface? Having a way to just add thby bryndole - Nginx Mailing List - English
I did some more research, ran a few experiments, and even looked at the code. What I described above is not possible. The limit_conn method fires exactly once for a connection, even if it goes to a "@retry" location. If you have several internal redirects ( rewrite /foo /bar last; ) only the last limit_conn is fired.by bryndole - Nginx Mailing List - English
I'm trying to limit the number of simultaneous connections based on a CGI args. For example some queries are more expensive to process than others, and I'm trying to avoid a few slow/expensive queries to starve out all the other faster queries. I would like to set an overall (server wide) connection limit of 100 connections overall, with 50 connection limit for &opt1 is set, a 30 connectionby bryndole - Nginx Mailing List - English
Maxim Dounin Wrote: ------------------------------------------------------- > For me, 1st google result points to author's site > (catap.ru) with > the patch posted alone. > > http://www.google.com/search?hl=en&q=upstream_coun > t_limit > http://catap.ru/patches/nginx/ngx_http_upstream_co > unt_limit-0.3.patch > Ah, yes. I should have looked more closby bryndole - Nginx Mailing List - English
The only such patch I could find was this: http://openhack.ru/nginx-patched/ Which only works with the memcached_gzip plugin of which it is a part. I too would find this feature extremely helpful. Preventing the situation where some request takes just a bit too long and times out on server or generates the "wrong" type of error and nginx tries upstream server after upstream serveby bryndole - Nginx Mailing List - English
As long as client request was identified as one came from proxy server ("Via" header present) nginx is able to disable or enable it's own gzip depending on various conditions. These conditions are controlled via gzip_proxied directive. Thank you, that makes complete sense now. -Brynby bryndole - Nginx Mailing List - English
I'm new to nginx. After setting up a reverse proxy and reading the documentation I am still left baffled over what the gzip_proxied option does. http://wiki.nginx.org/NginxHttpGzipModule#gzip_proxied gzip_proxied syntax: gzip_proxied ... default: gzip_proxied off context: http, server, location It allows or disallows the compression of the response for the proxy request in the dby bryndole - Nginx Mailing List - English