This article mainly introduces the optimization skills of Nginx PHP-FPM, which has certain reference value. Now I share it with you. Friends in need can refer to it

Nginx PHP-FPM optimization skills summary

Here is an article I found on the Internet. I have practiced it carefully. There are many places worthy of reference that can be learned. Since the layout of the previous article is very confusing, I will rewrite and organize this article while studying. All rights reserved. All rights reserved to the original author

Unix Domain Socket Communication

I have briefly introduced Unix Domain Socket this communication method before, see: Nginx PHP-FPM DomainSocketConfiguration method
UnixDomainSocketBecause it does not go through the network, it can indeed improve Nginx and php- fpm Communication performance, but it will be unstable under high concurrency.

Nginx will frequently report errors:

connect() to unix:/dev/shm/php-fcgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream

Copy after login

You can improve stability in the following two ways:

1. Increase the The backlog configuration method in nginx and php-fpm
is: of this domain name in the nginx configuration file Under server, add default backlog=1024 after listen 80.
Also configure listen.backlog in php-fpm.conf to 1024, the default is 128.
2. Increase the number of sock files and php-fpm instances and create a new sock file and pass ## in Nginx The #upstream module load-balances requests to two sets of php-fpm instances behind two sock
files.

php-fpm parameter tuning

2.1 Number of processes

# php-fpm初始/空闲/最大worker进程数
 pm.max_children = 300
 pm.start_servers = 20
 pm.min_spare_servers = 5
 pm.max_spare_servers = 35

Copy after login

2.2 Maximum number of requests processed

The maximum number of requests processed refers to one

php The worker process of -fpm will be terminated after processing a number of requests, and the master process will respawn a new one. The main purpose of this configuration is to avoid memory leaks caused by
php third-party libraries referenced by the interpreter or program

pm.max_requests = 10240

Copy after login

2.3 Maximum execution time

Maximum execution The time can be configured in

php.ini and php-fpm.conf. The configuration items are max_execution_time and request_terminate_timeout respectively. For its role and impact, please refer to: Detailed explanation of 502 and 504 errors in Nginx

High CPU usage troubleshooting method of php-fpm

1. CPU usage monitoring method

top command: After directly executing the
top command, enter 1 to see the CPU usage of each core. And the sampling time can be shortened by top -d 0.1. The following
sar seems to be the shortest is 1 second

sar command:

# sar和iostat命令的安装：
 sysstat.x86_64 : The sar and iostat system monitoring commands
 yum install -y sysstat.x86_64
 
# 执行sar -P ALL 1 100。-P ALL表示监控所有核心，1表示每1秒采集，100表示采集100次。
# 输出结果如下：
CPU     %user     %nice   %system   %iowait    %steal     %idle
all     85.54      0.00      5.69      0.00      0.00      8.76
  0     74.75      0.00     25.25      0.00      0.00      0.00
  1     98.00      0.00      2.00      0.00      0.00      0.00
  2     89.22      0.00      3.92      0.00      0.00      6.86
  3     91.00      0.00      2.00      0.00      0.00      7.00
  4     75.00      0.00      9.00      0.00      0.00     16.00
  5     94.95      0.00      5.05      0.00      0.00      0.00
  6     95.00      0.00      4.00      0.00      0.00      1.00
  7     87.88      0.00      4.04      0.00      0.00      8.08
  8     93.94      0.00      3.03      0.00      0.00      3.03
  9     88.00      0.00      3.00      0.00      0.00      9.00
 10     89.11      0.00      2.97      0.00      0.00      7.92
 11     82.35      0.00      3.92      0.00      0.00     13.73
 12     73.27      0.00      7.92      0.00      0.00     18.81
 13     81.44      0.00      4.12      0.00      0.00     14.43
 14     77.23      0.00      6.93      0.00      0.00     15.84
 15     78.79      0.00      4.04      0.00      0.00     17.17

Copy after login

2. Turn on the slow log

Configuration output

php-fpmSlow log, the threshold is 2 seconds:

request_slowlog_timeout = 2
slowlog = log/$pool.log.slow

Copy after login

Use the sort/uniq command to analyze and summarize the php-fpm slow log:

[root@boole log] grep -v "^$" www.log.slow.tmp | cut -d " " -f 3,2 | sort | uniq -c | sort -k1,1nr | head -n 50
   5181 run() /www/test.net/framework/web/filters/CFilter.php:41
   5156 filter() /www/test.net/framework/web/filters/CFilterChain.php:131
   2670 = /www/test.net/index.php
   2636 run() /www/test.net/application/controllers/survey/index.php:665
   2630 action() /www/test.net/application/controllers/survey/index.php:18
   2625 run() /www/test.net/framework/web/actions/CAction.php:75
   2605 runWithParams() /www/test.net/framework/web/CController.php:309
   2604 runAction() /www/test.net/framework/web/filters/CFilterChain.php:134
   2538 run() /www/test.net/framework/web/CController.php:292
   2484 runActionWithFilters() /www/test.net/framework/web/CController.php:266
   2251 run() /www/test.net/framework/web/CWebApplication.php:276
   1799 translate() /www/test.net/application/libraries/Limesurvey_lang.php:118
   1786 load_tables() /www/test.net/application/third_party/php-gettext/gettext.php:254
   1447 runController() /www/test.net/framework/web/CWebApplication.php:135
 
# 参数解释：
     sort:  对单词进行排序
     uniq -c:  显示唯一的行，并在每行行首加上本行在文件中出现的次数
     sort -k1,1nr:  按照第一个字段，数值排序，且为逆序
     head -10:  取前10行数据

Copy after login

3. Use strace to track the process

1. Use

nohup to convert strace to background execution until attach php-fpm Until the process dies:

nohup strace -T -p 13167 > 13167-strace.log &
# 参数说明:
-c 统计每一系统调用的所执行的时间,次数和出错的次数等.
-d 输出strace关于标准错误的调试信息.
-f 跟踪由fork调用所产生的子进程.
-o filename,则所有进程的跟踪结果输出到相应的filename
-F 尝试跟踪vfork调用.在-f时,vfork不被跟踪.
-h 输出简要的帮助信息.
-i 输出系统调用的入口指针.
-q 禁止输出关于脱离的消息.
-r 打印出相对时间关于,,每一个系统调用.
-t 在输出中的每一行前加上时间信息.
-tt 在输出中的每一行前加上时间信息,微秒级.
-ttt 微秒级输出,以秒了表示时间.
-T 显示每一调用所耗的时间.
-v 输出所有的系统调用.一些调用关于环境变量,状态,输入输出等调用由于使用频繁,默认不输出.
-V 输出strace的版本信息.
-x 以十六进制形式输出非标准字符串
-xx 所有字符串以十六进制形式输出.
-a column
设置返回值的输出位置.默认为40.
-e execve 只记录 execve 这类系统调用
-p 主进程号

Copy after login

2. You can also use the

-c parameter to let strace help summarize, which is very convenient and powerful !

[root@b28-12 log]# strace -cp 9907
Process 9907 attached - interrupt to quit
Process 9907 detached
% time     seconds  usecs/call     calls    errors syscall
------ ----------- ----------- --------- --------- ----------------
56.61    0.016612           5      3121           read
11.11    0.003259           1      2517       715 stat
  8.04    0.002358           7       349           brk
  6.02    0.001767           1      1315           poll
  4.28    0.001255           6       228           recvfrom
  2.71    0.000796           1       671           open
  2.54    0.000745           0      2453           fcntl
  2.37    0.000696           1      1141           write
  1.69    0.000497           1       593        13 access
  1.37    0.000403           0      1816           lseek
  0.89    0.000262           1       451        22 sendto
  0.56    0.000163           1       276       208 lstat
  0.49    0.000145           0       384           getcwd
  0.31    0.000090           0      1222           fstat
  0.28    0.000082           0       173           munmap
  0.26    0.000077           0       174           mmap
  0.24    0.000069           2        41           socket
  0.23    0.000068           0       725           close
  0.00    0.000000           0        13           rt_sigaction
  0.00    0.000000           0        13           rt_sigprocmask
  0.00    0.000000           0         1           rt_sigreturn
  0.00    0.000000           0        78           setitimer
  0.00    0.000000           0        26        26 connect
  0.00    0.000000           0        15         2 accept
  0.00    0.000000           0        39           recvmsg
  0.00    0.000000           0        26           shutdown
  0.00    0.000000           0        13           bind
  0.00    0.000000           0        13           getsockname
  0.00    0.000000           0        65           setsockopt
  0.00    0.000000           0        13           getsockopt
  0.00    0.000000           0         8           getdents
  0.00    0.000000           0        26           chdir
  0.00    0.000000           0         1           futex
------ ----------- ----------- --------- --------- ----------------
100.00    0.029344                 18000       986 total

Copy after login

4. Speed up PHP interpretation and execution

If there is indeed no problem with your program, it just performs too many operations and cannot be optimized anymore. Then consider using PHP accelerators such as

APC or xcache to reduce the time it takes CPU to interpret php files. These
PHP accelerators will generate intermediate code opcode when the php file is first interpreted, so subsequent executions will be much faster and less expensive. CPU operations. Let's take xcache as an example. Let's see how to install and configure it.

Installation

xcacheThe command is as follows. There are many parameters in ./configure and I don’t know what they are used for. There is no explanation on the official website, so I only enable - -enable-xcache: The configuration in

 tar zxvf xcache-3.0.3.tar.gz
     /usr/local/php/bin/phpize
     ./configure --with-php-config=/usr/local/php/bin/php-config --enable-xcache
     make
     make install

Copy after login

php.ini is as follows. The most important ones are the two parameters marked in red. It is generally recommended xcache.sizeDetermined based on the number of php files, xcache.count is the same as CPU core number:

[xcache.admin]
xcache.admin.enable_auth = Off
xcache.admin.user = "xcache"
xcache.admin.pass = ""
 
[xcache]
xcache.shm_scheme ="mmap"
xcache.size=1024M
xcache.count =16
xcache.slots =8K
xcache.ttl=0
xcache.gc_interval =0
xcache.var_size=16M
xcache.var_count =1
xcache.var_slots =8K
xcache.var_ttl=0
xcache.var_maxttl=0
xcache.var_gc_interval =300
xcache.test =Off
xcache.readonly_protection = Off
;xcache.readonly_protection = On
xcache.mmap_path ="/dev/zero"
;xcache.mmap_path ="/tmp/xcache"
xcache.coredump_directory =""
xcache.cacher =On
xcache.stat=On
xcache.optimizer =Off
 
[xcache.coverager]
;;xcache.coverager =On
;;xcache.coveragedump_directory =""

Copy after login

A common problem is starting

php -fpm will report an error:

Cannot open or create file set by xcache.mmap_path, check the path permission or check xcache.size/var_size against system limitation

Copy after login

This is because

/tmp/xcache is a file and cannot be created as a directory.

After restarting the

php-fpm service, use the top command to observe the VIRT( Including the swap area) are all xcache.size size, but REQ becomes very small. Using the above configuration has shortened the peak time of CPU usage, but at the peak time all cores will still reach more than
90%. I don’t know if there is no configuration somewhere. right. In addition, when concurrency is high, /dev/zero this configuration method often leads to
Nginx 502 errors. /tmp/xcache and turning on readonly_protection are very stable.

php程序性能监控

常用的方法就是开启xdebug的性能监控功能，将xdebug输出结果通过WinCacheGrind软件分析。
xdebug的安装和配合IDE调试的方法参见：Vim+XDebug调试PHP

php.ini中配置的这几项是输出性能信息的：

xdebug.auto_trace = on
xdebug.auto_profile = on
xdebug.collect_params = on
xdebug.collect_return = on
xdebug.profiler_enable = on
xdebug.trace_output_dir = "/tmp"
xdebug.profiler_output_dir ="/tmp"

Copy after login

这样XDebug会输出所有执行php函数的性能数据，但产生的文件也会比较大。可以关闭一些选项如collect_params、collect_return，
来减少输出的数据量。或者关闭自动输出，通过在想要监控的函数首尾调用xdebug函数来监控指定的函数。

输出的文件名类似cachegrind.out.1277560600和trace.3495983249.txt，可以拿到Windows平台下用WinCacheGrind进行图形化分析。
WinCacheGrind使用方法网上有很多介绍，这里就不详细说明了,WinCacheGrind for github

Optimization tips for Nginx+PHP-FPM