<html><body><div>Ryan,<br></div><div><br data-mce-bogus="1"></div><div>What is the topology of the system that you are describing?<br data-mce-bogus="1"></div><div><br data-mce-bogus="1"></div><div>You mention kong/nginx, an upstream host, a load balancer, clients ...<br data-mce-bogus="1"></div><div><br data-mce-bogus="1"></div><div>Are the load balancers hardware or software devices? <br data-mce-bogus="1"></div><div>Is kong nginx simply forwarding to a load-balancer VIP that fronts multiple upstream systems?<br data-mce-bogus="1"></div><div>Are there any firewalls or intrusion detection systems also in the mix?<br data-mce-bogus="1"></div><div>Are your clients remote? internet based?<br data-mce-bogus="1"></div><div>Are your nginx instances running on physical hosts, virtual machines, cloud based VMs?<br data-mce-bogus="1"></div><div>Are your nginx instances and your upstream system s in the same datacenter or some distance apart?<br data-mce-bogus="1"></div><div><br data-mce-bogus="1"></div><div>Peter<br data-mce-bogus="1"></div><div><br>On May 16, 2017, at 12:57 PM, rcutter <nginx-forum@forum.nginx.org> wrote:<br><br></div><div><blockquote type="cite"><div class="msg-quote"><div class="_stretch"><span class="body-text-content">Hello, I believe I have a tuning issue with NGINX - hoping someone can point<br>me in the right direction!<br><br>About 1 in 60,000 requests being proxied through Kong/NGINX are timing out. <br>These requests are getting to the upstreamed host and are successfully<br>logged in the load balancer in front of this upstreamed host. So either<br>there's a network issue between that load balancer and NGINX or NGINX is<br>simply not able/willing to process the response.<br><br>Assuming this is an NGINX tuning issue, these are my settings (note these<br>hosts have 32 cores). Traffic is not all that high, less than 10 req/sec<br>per instance and requests are usually satisfied in less than a second:<br><br>worker_processes auto;<br>worker_priority -10;<br>daemon off;<br>pid pids/nginx.pid;<br>error_log logs/error.log debug;<br><br>worker_rlimit_nofile 20000;<br><br>events {<br> worker_connections 10000;<br> multi_accept on;<br>}<br><br>Most all other config settings are "default" values. There's nothing in the<br>Kong logs that indicate these dropped responses are being processed by Kong.<br> There's no indication there aren't enough workers. These timeouts do not<br>happen in clusters, they are more like singletons.<br><br>Any advice on things I should look at or diagnosis possibilities? Thanks<br>very much, Ryan<br><br>Posted at Nginx Forum: <a href="https://forum.nginx.org/read.php?2,274267,274267#msg-274267" data-mce-href="https://forum.nginx.org/read.php?2,274267,274267#msg-274267">https://forum.nginx.org/read.php?2,274267,274267#msg-274267</a><br><br>_______________________________________________<br>nginx mailing list<br><a href="mailto:nginx@nginx.org" data-mce-href="mailto:nginx@nginx.org">nginx@nginx.org</a><br><a href="http://mailman.nginx.org/mailman/listinfo/nginx" data-mce-href="http://mailman.nginx.org/mailman/listinfo/nginx">http://mailman.nginx.org/mailman/listinfo/nginx</a><br></span></div></div></blockquote></div></body></html>