Introduction to places where nginx can be optimized

This article brings you an introduction to the places where nginx can be optimized. It has certain reference value. Friends in need can refer to it. I hope it will be helpful to you.

worker_processes 8;

nginxThe number of processes is recommended to be specified according to the number ofcpu, which is usually a multiple of it.

worker_cpu_affinity 00000001 00000010 00000100 00001000 00010000 00100000 01000000 10000000;

Assigncputo each process. In the above example,8The process is assigned to8cpu. Of course, you can write multiple, or assign one process to multiplecpu.

worker_rlimit_nofile 102400;

This command refers to the maximum number of file descriptors opened by anginxprocess. The theoretical value should be the maximum number of open files (ulimit n) is divided by the number ofnginxprocesses, butnginxallocates requests It's not that uniform, so it's better to keep it consistent with the value ofulimit n.

use epoll;

Using theI/Omodel ofepoll, needless to say this .

worker_connections 102400;

The maximum number of connections allowed by each process. Theoretically, the maximum number of connections for eachnginxserver isworker_processes*worker_connections.

keepalive_timeout 60;

keepaliveTimeout.

client_header_buffer_size 4k;

The buffer size of the client request header. This can be set according to the paging size of your system. Generally, the header size of a request will not exceed1k,However, since the general system paging is larger than1k, so the paging size is set here. The paging size can be obtained with the commandgetconf PAGESIZE.

open_file_cache max=102400 inactive=20s;

This will specify the cache for open files. It is not enabled by default.maxSpecifies the number of caches. It is recommended to be the same as the number of open files. Sincerely,inactiverefers to how long the file has not been requested before the cache is deleted.

open_file_cache_valid 30s;

This refers to how often to check cached valid information.

open_file_cache_min_uses 1;

open_file_cache#inactiveThe minimum number of times the file is used within the parameter time in the directive, if this number is exceeded, The file descriptor is always

open in the cache. If a file is not used once ininactivetime, it will was removed.

Optimization of kernel parameters

net.ipv4.tcp_max_tw_buckets = 6000

number of timewait, the default is180000.

net.ipv4.ip_local_port_range = 1024 65000

The port range allowed to be opened by the system.

net.ipv4.tcp_tw_recycle = 1

EnabletimewaitFast recycling.

net.ipv4.tcp_tw_reuse = 1

Enable reuse. AllowTIMEWAIT socketsto be reused for newTCPconnections.

net.ipv4.tcp_syncookies = 1

EnableSYN Cookies, when # appears##SYNEnablecookiesto handle when waiting for the queue to overflow.

net.core.somaxconn = 262144

webIn applicationlistenThe function’sbacklogdefaults to thenet.core.somaxconnthat will limit our kernel parameters to128, andnginxdefinitionNGX_LISTEN_BACKLOGdefaults to511, so it is necessary to adjust this value.net.core.netdev_max_backlog = 262144

When each network interface receives packets faster than the kernel can process those packets, The maximum number of packets allowed to be sent to the queue.

net.ipv4.tcp_max_orphans = 262144

The maximum number of

TCPThe socket is not associated with any user file handle. If this number is exceeded, the orphan connection will be reset immediately and a warning message will be printed. This limit is only to prevent simpleDoSattacks. You cannot rely too much on it or artificially reduce this value. You should increase this value(If you increase the memory).net.ipv4.tcp_max_syn_backlog = 262144

The maximum value of recorded connection requests that have not yet received client confirmation information. For systems with

128Mmemory, the default value is1024, and for systems with small memory it is128.net.ipv4.tcp_timestamps = 0

Time stamps can avoid sequence number wrapping. A

1Gbpslink will definitely encounter a previously used sequence number. The timestamp allows the kernel to accept such"Exception"data packets. It needs to be turned off here.net.ipv4.tcp_synack_retries = 1

In order to open the connection to the peer, the kernel needs to send a

SYNAnd comes with aACKthat responds to the previousSYN. This is the second handshake in the so-called three-way handshake. This setting determines the number ofSYN ACKpackets sent before the kernel abandons the connection.net.ipv4.tcp_syn_retries = 1

在内核放弃建立连接之前发送SYN包的数量。

net.ipv4.tcp_fin_timeout = 1

如果套接字由本端要求关闭，这个参数决定了它保持在FINWAIT2状态的时间。对端可以出错并永远不关 闭连接，甚至意外当机。缺省值是60秒。2.2内核的通常值是180秒，你可以按这个设置，但要记住的是， 即使你的机器是一个轻载的WEB服务器，也有因为大量的死套接字而内存溢出的风险，FIN WAIT2的危 险性比FINWAIT1要小，因为它最多只能吃掉1.5K内存，但是它们的生存期长些。

net.ipv4.tcp_keepalive_time = 30

当keepalive起用的时候，TCP发送keepalive消息的频度。缺省是2小时。

一个完整的内核优化配置

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 = 1 kernel.msgmnb = 65536 kernel.msgmax = 65536 kernel.shmmax = 68719476736 kernel.shmall = 4294967296 net.ipv4.tcp_max_tw_buckets = 6000 net.ipv4.tcp_sack = 1 net.ipv4.tcp_window_scaling = 1 net.ipv4.tcp_rmem = 4096 87380 4194304 net.ipv4.tcp_wmem = 4096 16384 4194304 net.core.wmem_default = 8388608 net.core.rmem_default = 8388608 net.core.rmem_max = 16777216 net.core.wmem_max = 16777216

net.core.netdev_max_backlog = 262144 net.core.somaxconn = 262144 net.ipv4.tcp_max_orphans = 3276800 net.ipv4.tcp_max_syn_backlog = 262144 net.ipv4.tcp_timestamps = 0 net.ipv4.tcp_synack_retries = 1 net.ipv4.tcp_syn_retries = 1

The above is the detailed content of Introduction to places where nginx can be optimized. For more information, please follow other related articles on the PHP Chinese website!