# Unanswered requests from local host, high load **URL:** \ **Category:** NGINX\ **Created:** [October 28, 2010, 3:50pm UTC](https://www.ruby-forum.com/t/unanswered-requests-from-local-host-high-load/198247 "2010-10-28T15:50:49Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Dobai-Pataky\_BSSSSl](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/77aa72/32.png) [@Dobai-Pataky\_BSSSSl](https://www.ruby-forum.com/u/Dobai-Pataky_BSSSSl)\ **Post date:** [October 28, 2010, 3:50pm UTC](https://www.ruby-forum.com/t/unanswered-requests-from-local-host-high-load/198247/1 "2010-10-28T15:50:49Z")
Dear list, I’m running a public mirror server which currently handles about 5 GBit/sec with 15.000 concurrent connections and 1400 requests/second. We only serve static files and basically all served files are in RAM (e.g. the current Mozilla Firefox update). In high load scenarios nginx sometimes fails to answer simple requests in time, even when they originate from localhost. According to SmokePing’s “HTTP ping”, which just GETs/HEADs the static index.html file, the time needed for this answer changes dramatically and packet loss is involved with up to 17% when the problems occur. The Apache running in the background did not have this problem for queries from localhost. When considering external queries (another instance of SmokePing), I see gaps (100% packet loss) for both Apache and nginx, where nginx also has packet loss up to 20% and higher response times around that time. If you want I can provide the four snippets of SmokePing graphs. Configuration: nginx 0.8.53, 64 workers, no keepalive, 5000 worker connections, epoll, sendfile, tcp\_nodelay Linux 2.6.26 (Debian), max 131072 open files two quadcore Xeons, 60 GByte DDR3, Intel 10 GBit NIC 10 TByte StorageTek disk backend In case you need further information, I’m happy to provide it. I just increased to 96 workers (as I already wanted, I accidently reverted that config today), let’s see what happens… Best regards, ---
**Author:** ![Carsten\_Otto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/ea666f/32.png) [@Carsten\_Otto](https://www.ruby-forum.com/u/Carsten_Otto)\ **Post date:** [October 28, 2010, 5:30pm UTC](https://www.ruby-forum.com/t/unanswered-requests-from-local-host-high-load/198247/2 "2010-10-28T17:30:50Z")
I’m suprised you need so many workers. Sent from my iPod ---
**Author:** ![Carsten\_Otto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/ea666f/32.png) [@Carsten\_Otto](https://www.ruby-forum.com/u/Carsten_Otto)\ **Post date:** [October 28, 2010, 5:37pm UTC](https://www.ruby-forum.com/t/unanswered-requests-from-local-host-high-load/198247/3 "2010-10-28T17:37:44Z")
On Fri, Oct 29, 2010 at 02:27:45AM +1100, Splitice wrote: > I’m suprised you need so many workers. I’m not sure if we need that many workers, I just have them. My current idea is that we hit some limit (max open files?), because I just noticed that both Apache and nginx did not accept/serve new connections for a serious amount of time (two minutes or so). Best regards,