# Nginx on high load (upstream prematurely closed connection while reading response header from upstre **URL:** \ **Category:** NGINX\ **Created:** [May 18, 2016, 10:50am UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642 "2016-05-18T10:50:53Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![Sadaf\_N](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/858c86/32.png) [@Sadaf\_N](https://www.ruby-forum.com/u/Sadaf_N)\ **Post date:** [May 18, 2016, 10:50am UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/1 "2016-05-18T10:50:53Z")
Dear Guys, I am facing strange issue, during load test and peak load of more than 3k concurrent users, get below errors in nginx logs continuously. 2016/05/18 11:23:28 [error] 15510#0: \*6853 upstream prematurely closed connection while reading response header from upstream, client: x.x.x.x, server:abc.com, request: “GET / HTTP/1.1”, upstream: “fastcgi://127.0.0.1:9000”, host: “10.50.x.x” And beside of above error frequently, below error come gradually 2016/05/18 11:23:28 [error] 15510#0: \*7431 recv() failed (104: Connection reset by peer) while reading response header from upstream, client: x.x.x.x, server: [abc.com](http://abc.com), request: “POST / HTTP/1.1”, upstream: “fastcgi://127.0.0.1:9000”, host: “10.50.x.x”, referrer: “[http://10.50.x.x](http://10.50.x.x)” But with out peak load, no error is coming and during load besides of above errors continues coming site is working fine and i see 200 response code in access logs, but still wondering why above errors coming only in peak load or during load testing. Below is the require info. I already tune up the sysctl settings for high throughput nginx version: nginx/1.8.1 php-fpm /php version : 5.4.45 echo 0 \> /proc/sys/net/ipv4/tcp\_ecn echo 1 \> /proc/sys/net/ipv4/tcp\_tw\_reuse echo 20 \> /proc/sys/net/ipv4/tcp\_fin\_timeout echo 102400 \> /proc/sys/net/ipv4/tcp\_max\_syn\_backlog echo 102400 \> /proc/sys/net/core/somaxconn echo 102400 \> /proc/sys/net/core/netdev\_max\_backlog echo 5242888 \> /proc/sys/net/netfilter/nf\_conntrack\_max echo 400000 \> /proc/sys/net/ipv4/tcp\_max\_tw\_buckets echo 655300 \> /proc/sys/vm/max\_map\_count echo 11000 65000 \> /proc/sys/net/ipv4/ip\_local\_port\_range Kindly please help & advise. Posted at Nginx Forum: > **[nginx on high load (upstream prematurely closed connection while reading...](https://forum.nginx.org/read.php?2,266978,266978#msg-266978)** > > Dear Guys, I am facing strange issue, during load test and peak load of more than 3k concurrent users, get below errors in nginx logs continuously. 2016/05/18 11:23:28 \[error\] 15510#0: \*6853 upstream prematurely closed connection while reading... ---
**Author:** ![muzi](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/49beb7/32.png) [@muzi](https://www.ruby-forum.com/u/muzi)\ **Post date:** [May 19, 2016, 10:31am UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/2 "2016-05-19T10:31:34Z")
Dear Francis D. , thankyou for your response, upstream is php-fpm server (127.0.0.1) and its also setup to handle huge load, during above errors, i see the site is opening fine. but don’t know where to catch it. PM is also set on demand and maximum childerns set to 90000 to handle the load, and no logs appears in php-fpm. Can you please suggest how to got the issue ? Posted at Nginx Forum: > **[Re: nginx on high load (upstream prematurely closed connection while reading...](https://forum.nginx.org/read.php?2,266978,266995#msg-266995)** > > Dear Guys, I am facing strange issue, during load test and peak load of more than 3k concurrent users, get below errors in nginx logs continuously. 2016/05/18 11:23:28 \[error\] 15510#0: \*6853 upstream prematurely closed connection while reading... ---
**Author:** ![muzi](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/49beb7/32.png) [@muzi](https://www.ruby-forum.com/u/muzi)\ **Post date:** [May 18, 2016, 6:34pm UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/3 "2016-05-18T18:34:07Z")
On Wed, May 18, 2016 at 04:50:24AM -0400, muzi wrote: Hi there, > I am facing strange issue, during load test and peak load of more than 3k > concurrent users, get below errors in nginx logs continuously. > > 2016/05/18 11:23:28 [error] 15510#0: \*6853 upstream prematurely closed > connection while reading response header from upstream, client: x.x.x.x, > server:abc.com, request: “GET / HTTP/1.1”, upstream: > “fastcgi://127.0.0.1:9000”, host: “10.50.x.x” “upstream” here is your fastcgi server. The nginx message is that the fastcgi server broke the connection. > And beside of above error frequently, below error come gradually > > 2016/05/18 11:23:28 [error] 15510#0: \*7431 recv() failed (104: Connection > reset by peer) while reading response header from upstream, client: x.x.x.x, > server: [abc.com](http://abc.com), request: “POST / HTTP/1.1”, upstream: > “fastcgi://127.0.0.1:9000”, host: “10.50.x.x”, referrer: “[http://10.50.x.x](http://10.50.x.x)” And that message is that the fasctcgi server broke the connection. It is probably worth investigating what the fastcgi server thinks is happening – perhaps it is not configured to handle the load it is being given? ## f Francis D. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) ---
**Author:** ![muzi](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/49beb7/32.png) [@muzi](https://www.ruby-forum.com/u/muzi)\ **Post date:** [May 19, 2016, 9:29pm UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/4 "2016-05-19T21:29:21Z")
On Thu, May 19, 2016 at 04:31:14AM -0400, muzi wrote: Hi there, > upstream is php-fpm server > (127.0.0.1) and its also setup to handle huge load, during above errors, i > see the site is opening fine. but don’t know where to catch it. PM is also > set on demand and maximum childerns set to 90000 to handle the load, and no > logs appears in php-fpm. Can you please suggest how to got the issue ? nginx says that upstream is doing something odd. upstream is php-fpm. I suggest checking the php-fpm docs to see how to get it to log what it is doing – if it says nginx is doing something odd, then there is a conflict to investigate. Otherwise, it should says something about the connections that it is closing. If the site is opening fine, maybe there is no problem to worry about. ## f Francis D. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) ---
**Author:** ![muzi](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/49beb7/32.png) [@muzi](https://www.ruby-forum.com/u/muzi)\ **Post date:** [May 23, 2016, 10:58am UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/5 "2016-05-23T10:58:14Z")
Thankyou once again Francis D., i am now ignoring these errors, as i not see any errors in nginx access logs and site is working on high load and see 200 response code in access logs. Posted at Nginx Forum: > **[Re: nginx on high load (upstream prematurely closed connection while reading...](https://forum.nginx.org/read.php?2,266978,267053#msg-267053)** > > Dear Guys, I am facing strange issue, during load test and peak load of more than 3k concurrent users, get below errors in nginx logs continuously. 2016/05/18 11:23:28 \[error\] 15510#0: \*6853 upstream prematurely closed connection while reading... ---
**Author:** ![muzi](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/49beb7/32.png) [@muzi](https://www.ruby-forum.com/u/muzi)\ **Post date:** [May 23, 2016, 11:14am UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/6 "2016-05-23T11:14:01Z")
Thankyou once again Francis D., i am now ignoring these errors, as i not see any errors in nginx access logs and site is working on high load and see 200 response code in access logs. Posted at Nginx Forum: > **[Re: nginx on high load (upstream prematurely closed connection while reading...](https://forum.nginx.org/read.php?2,266978,267051#msg-267051)** > > Dear Guys, I am facing strange issue, during load test and peak load of more than 3k concurrent users, get below errors in nginx logs continuously. 2016/05/18 11:23:28 \[error\] 15510#0: \*6853 upstream prematurely closed connection while reading... ---
**Author:** ![muzi](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/49beb7/32.png) [@muzi](https://www.ruby-forum.com/u/muzi)\ **Post date:** [May 23, 2016, 11:14am UTC](https://www.ruby-forum.com/t/nginx-on-high-load-upstream-prematurely-closed-connection-while-reading-response-header-from-upstre/244642/7 "2016-05-23T11:14:36Z")
Thankyou once again Francis D., i am now ignoring these errors, as i not see any errors in nginx access logs and site is working on high load and see 200 response code in access logs. Posted at Nginx Forum: > **[Re: nginx on high load (upstream prematurely closed connection while reading...](https://forum.nginx.org/read.php?2,266978,267052#msg-267052)** > > Dear Guys, I am facing strange issue, during load test and peak load of more than 3k concurrent users, get below errors in nginx logs continuously. 2016/05/18 11:23:28 \[error\] 15510#0: \*6853 upstream prematurely closed connection while reading...