# Cpu spike on HUP **URL:** \ **Category:** NGINX\ **Created:** [October 29, 2010, 1:57am UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276 "2010-10-29T01:57:02Z")\ **Posts on this page:** 6\ **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 29, 2010, 1:57am UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276/1 "2010-10-29T01:57:02Z")
Whenever nginx is sent a HUP signal, the cpu jumps to 100% then slowly drops until all the nginx worker processes that are trying to shutting down are done. Those nginx process usually take a while to shutdown since we’re serving very large files (200MB-2GB). Any ideas why this is happening? From the strace below it looks like it all in sendfile but I wouldn’t think that would so cpu intense. -gary /usr/local/nginx/sbin/nginx -V nginx version: nginx/0.8.52 built by gcc 4.4.3 (Ubuntu 4.4.3-4ubuntu5) TLS SNI support enabled configure arguments: --with-http\_ssl\_module --with-http\_flv\_module This is from about 10 seconds of strace of one of the nginx process while shutting down. sudo strace -cp 7087 Process 7087 attached - interrupt to quit ^CProcess 7087 detached % time seconds usecs/call calls errors syscall * * * 99.85 0.071681 139 515 sendfile 0.15 0.000110 0 252 epoll\_wait * * * 100.00 0.071791 767 total ---
**Author:** ![Gary\_Barrueto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/3da27b/32.png) [@Gary\_Barrueto](https://www.ruby-forum.com/u/Gary_Barrueto)\ **Post date:** [October 29, 2010, 10:03am UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276/2 "2010-10-29T10:03:14Z")
On Thu, Oct 28, 2010 at 04:55:40PM -0700, Gary Barrueto wrote: > Whenever nginx is sent a HUP signal, the cpu jumps to 100% then slowly > drops until all the nginx worker processes that are trying to shutting > down are done. Those nginx process usually take a while to shutdown > since we’re serving very large files (200MB-2GB). Any ideas why this is > happening? > > From the strace below it looks like it all in sendfile but I wouldn’t > think that would so cpu intense. What kind of CPU is used in these moments: system or user according to top ? > sudo strace -cp 7087 > Process 7087 attached - interrupt to quit > ^CProcess 7087 detached > % time seconds usecs/call calls errors syscall > > * * * > > 99.85 0.071681 139 515 sendfile > 0.15 0.000110 0 252 epoll\_wait > > * * * > > 100.00 0.071791 767 total – Igor S. > **[Igor Sysoev](http://sysoev.ru/en/)** ---
**Author:** ![Gary\_Barrueto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/3da27b/32.png) [@Gary\_Barrueto](https://www.ruby-forum.com/u/Gary_Barrueto)\ **Post date:** [October 29, 2010, 11:04pm UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276/3 "2010-10-29T23:04:00Z")
On 10/29/2010 01:02 AM, Igor S. wrote: > What kind of CPU is used in these moments: system or user according to top ? I just sent a HUP signal to nginx and the cpu shot up to 55% user and another 10% sys. -gary ---
**Author:** ![Gary\_Barrueto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/3da27b/32.png) [@Gary\_Barrueto](https://www.ruby-forum.com/u/Gary_Barrueto)\ **Post date:** [October 30, 2010, 8:08am UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276/4 "2010-10-30T08:08:56Z")
On Fri, Oct 29, 2010 at 02:02:29PM -0700, Gary Barrueto wrote: > > > think that would so cpu intense. > > > What kind of CPU is used in these moments: system or user according to top ? > > I just sent a HUP signal to nginx and the cpu shot up to 55% user and > another 10% sys. Do you send a HUP signal to master process only or to worker processes too ? – Igor S. > **[Igor Sysoev](http://sysoev.ru/en/)** ---
**Author:** ![Gary\_Barrueto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/3da27b/32.png) [@Gary\_Barrueto](https://www.ruby-forum.com/u/Gary_Barrueto)\ **Post date:** [November 3, 2010, 7:06pm UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276/5 "2010-11-03T19:06:46Z")
On 10/30/2010 2:13 AM, Gary Barrueto wrote: > On 10/29/2010 11:07 PM, Igor S. wrote: > > > > I just sent a HUP signal to nginx and the cpu shot up to 55% user and > > > another 10% sys. > > > Do you send a HUP signal to master process only or to worker > > > processes too ? > > > I sent the HUP signal to the master process. > > > -gary Normally, the cpu is at 1% user and system is at 2% when serving 60-90MB/sec. -gary ---
**Author:** ![Gary\_Barrueto](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/3da27b/32.png) [@Gary\_Barrueto](https://www.ruby-forum.com/u/Gary_Barrueto)\ **Post date:** [October 30, 2010, 11:14am UTC](https://www.ruby-forum.com/t/cpu-spike-on-hup/198276/6 "2010-10-30T11:14:47Z")
On 10/29/2010 11:07 PM, Igor S. wrote: > > I just sent a HUP signal to nginx and the cpu shot up to 55% user and > > another 10% sys. > > Do you send a HUP signal to master process only or to worker processes too ? I sent the HUP signal to the master process. -gary