nginx配置不好，并发性很差

热

哥  NGINX是单进程的你不知道吗   你开两千个woker来回切能块吗   每次上下文切换 和 进程本身都有资源开销  

肯定快不了   单核就开一个 worker 双核开两个worker

楼主就是8核啊。。。

另外 worker_cpu_affinity，这个不建议设置，这是个坑。尤其是如果你用的tengine，它有这个东西，是因为这是针对淘宝的硬件特别优化的，其他人不能这么来

就以一个worker来算，8核心CPU，16G内存，并发2000都有失败的情况，那还谈何nginx高并发和C10K

一定是配置方面有问题

net.ipv4.tcp_syncookies = 0

禁用Syncookies 试试，因为在高并发的情况下，内核会认为系统受到了SYN flood攻击，会发送cookies（possible SYN flooding on port 80. Sending cookies），这样会减慢影响请求的速度。

其中远程使用ab压力测试可能受限于网络设备的吞吐量,Nginx的RPS和吞吐量可能都不会太高.

另外Linux上CPU亲缘性绑定是完全可用的,比如taskset命令就可以绑定指定进程到对应的CPU核心.Nginx提供了更便捷的方式,在配置里提供了指令worker_cpu_affinity用于进行绑定.

“ apr_socket_recv: Connection reset by peer (104)”

这个是 ab 开不了那么多并发数吧

首先，你检测一下

ulimit -a

是不是open files的数量不够，如果没问题，检查你的内核参数，用我这个试试：

net.ipv4.ip_local_port_range = 1024 65000
net.core.rmem_max=16777216
net.core.wmem_max=16777216
net.ipv4.tcp_rmem=4096 87380 16777216
net.ipv4.tcp_wmem=4096 65536 16777216
net.ipv4.tcp_fin_timeout = 10
net.ipv4.tcp_tw_recycle = 1
net.ipv4.tcp_timestamps = 0
net.ipv4.tcp_window_scaling = 0
net.ipv4.tcp_sack = 0
net.core.netdev_max_backlog = 30000
net.ipv4.tcp_no_metrics_save=1
net.core.somaxconn = 262144
net.ipv4.tcp_syncookies = 0
net.ipv4.tcp_max_orphans = 262144
net.ipv4.tcp_max_syn_backlog = 262144
net.ipv4.tcp_synack_retries = 2
net.ipv4.tcp_syn_retries = 2

[root@localhost logs]# /sbin/sysctl -p
net.ipv4.ip_forward = 0
net.ipv4.conf.default.rp_filter = 1
net.ipv4.conf.default.accept_source_route = 0
kernel.sysrq = 0
kernel.core_uses_pid = 1
net.ipv4.tcp_syncookies = 0
kernel.msgmnb = 65536
kernel.msgmax = 65536
kernel.shmmax = 68719476736
kernel.shmall = 4294967296
net.ipv4.ip_local_port_range = 1024 65000
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216
net.ipv4.tcp_rmem = 4096 87380 16777216
net.ipv4.tcp_wmem = 4096 65536 16777216
net.ipv4.tcp_fin_timeout = 10
net.ipv4.tcp_tw_recycle = 1
net.ipv4.tcp_timestamps = 0
net.ipv4.tcp_window_scaling = 0
net.ipv4.tcp_sack = 0
net.core.netdev_max_backlog = 30000
net.ipv4.tcp_no_metrics_save = 1
net.core.somaxconn = 262144
net.ipv4.tcp_syncookies = 0
net.ipv4.tcp_max_orphans = 262144
net.ipv4.tcp_max_syn_backlog = 262144
net.ipv4.tcp_synack_retries = 2
net.ipv4.tcp_syn_retries = 2

[root@localhost logs]# ulimit -n
102400

[root@localhost logs]# ../sbin/nginx -s stop
[root@localhost logs]# ../sbin/nginx
[root@localhost logs]# ps -ef|grep nginx
root     59707     1  0 07:22 ?        00:00:00 nginx: master process ../sbin/nginx
nobody   59708 59707  0 07:22 ?        00:00:00 nginx: worker process

Server Software:        nginx/1.7.2
Server Hostname:        172.16.18.114
Server Port:            80


Document Path:          /index.html
Document Length:        61 bytes


Concurrency Level:      2000
Time taken for tests:   81.722 seconds
Complete requests:      500000
Failed requests:        0
Write errors:           0
Keep-Alive requests:    496000
Total transferred:      147980000 bytes
HTML transferred:       30500000 bytes
Requests per second:    6118.34 [#/sec] (mean)
Time per request:       326.886 [ms] (mean)
Time per request:       0.163 [ms] (mean, across all concurrent requests)
Transfer rate:          1768.34 [Kbytes/sec] received

还是一样差，为啥呢？