Board logo

标题: Apache2.4的MPM模式（prefork） [打印本页]
作者: admin    时间: 2020-4-2 10:35     标题: Apache2.4的MPM模式（prefork）

笺注：这是在 LNMP一键安装包（lamp_CentOS6.9) 的基础上进行的。 查看Apache的版本： [root@localhost ~]# find / -name httpd /usr/local/apache/bin/httpd [root@localhost ~]# [root@localhost ~]# /usr/local/apache/bin/httpd -v Server version: Apache/2.4.20 (Unix) Server built: Jun 25 2020 01:46:46 查看Apache编译安装的参数: [root@localhost ~]# find / -name config.nice /usr/local/apache/build/config.nice [root@localhost ~]# [root@localhost ~]# cat /usr/local/apache/build/config.nice #! /bin/sh # # Created by configure "./configure" \ "--prefix=/usr/local/apache" \ "--enable-mods-shared=most" \ "--enable-headers" \ "--enable-mime-magic" \ "--enable-proxy" \ "--enable-so" \ "--enable-rewrite" \ "--with-ssl" \ "--enable-ssl" \ "--enable-deflate" \ "--with-pcre" \ "--with-included-apr" \ "--with-apr-util" \ "--enable-mpms-shared=all" \ "--with-mpm=prefork" \ "--enable-remoteip" \ "$@" 查看Apache现在使用的MPM模式： [root@localhost ~]# /usr/local/apache/bin/httpd -V Server version: Apache/2.4.20 (Unix) Server built: Jun 25 2020 01:46:46 Server's Module Magic Number: 20120211:57 Server loaded: APR 1.5.2, APR-UTIL 1.5.4 Compiled using: APR 1.5.2, APR-UTIL 1.5.4 Architecture: 64-bit Server MPM: prefork threaded: no forked: yes (variable process count) Server compiled with.... -D APR_HAS_SENDFILE -D APR_HAS_MMAP -D APR_HAVE_IPV6 (IPv4-mapped addresses enabled) -D APR_USE_SYSVSEM_SERIALIZE -D APR_USE_PTHREAD_SERIALIZE -D SINGLE_LISTEN_UNSERIALIZED_ACCEPT -D APR_HAS_OTHER_CHILD -D AP_HAVE_RELIABLE_PIPED_LOGS -D DYNAMIC_MODULE_LIMIT=256 -D HTTPD_ROOT="/usr/local/apache" -D SUEXEC_BIN="/usr/local/apache/bin/suexec" -D DEFAULT_PIDLOG="logs/httpd.pid" -D DEFAULT_SCOREBOARD="logs/apache_runtime_status" -D DEFAULT_ERRORLOG="logs/error_log" -D AP_TYPES_CONFIG_FILE="conf/mime.types" -D SERVER_CONFIG_FILE="conf/httpd.conf" 笺注：Apache2.4有三种稳定的MPM（Multi-Processing Module，多进程处理模块）模式，分别是prefork、worker、event。 prefork是比较古老而又稳定的MPM模式，特点是每个进程都是单线程，在一个时间点只能处理一个连接，需要启动大量的进程来处理高并发的请求。 在Apache的主配置文件中指明使用prefork： [root@localhost ~]# cat /usr/local/apache/conf/httpd.conf |grep 'mpm_' #LoadModule mpm_event_module modules/mod_mpm_event.so LoadModule mpm_prefork_module modules/mod_mpm_prefork.so MPM模式的配置文件： [root@localhost ~]# cat /usr/local/apache/conf/httpd.conf |grep -A1 "MPM specific" # Server-pool management (MPM specific) Include conf/extra/httpd-mpm.conf [root@localhost ~]# vi /usr/local/apache/conf/extra/httpd-mpm.conf prefork MPM： 图片1.png 笺注： MaxRequestWorkers 表示的是最大子进程数（即最大并发连接数），任何高于这个数量的连接，都会放入等待队列中； Apache的MPM模式prefork，对高并发的支持稍差； 测试： [root@localhost ~]# ab -bash: ab: command not found [root@localhost ~]# yum -y install httpd-tools [root@localhost ~]# ab -n 1000 -c 250 http://192.168.168.130/phpinfo.php #指定并发请求数为250，总请求数为1000，对http://192.168.168.130/phpinfo.php进行压力测试 This is ApacheBench, Version 2.3 <$Revision: 655654 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking 192.168.168.130 (be patient) Completed 100 requests Completed 200 requests Completed 300 requests Completed 400 requests Completed 500 requests Completed 600 requests Completed 700 requests Completed 800 requests Completed 900 requests Completed 1000 requests Finished 1000 requests Server Software: Apache Server Hostname: 192.168.168.130 Server Port: 80 Document Path: /phpinfo.php Document Length: 87189 bytes Concurrency Level: 250 Time taken for tests: 1.567 seconds Complete requests: 1000 Failed requests: 105 (Connect: 0, Receive: 0, Length: 105, Exceptions: 0) Write errors: 0 Total transferred: 87531723 bytes HTML transferred: 87351828 bytes Requests per second: 638.18 [#/sec] (mean) Time per request: 391.741 [ms] (mean) Time per request: 1.567 [ms] (mean, across all concurrent requests) Transfer rate: 54551.44 [Kbytes/sec] received Connection Times (ms) min mean[+/-sd] median max Connect: 0 9 77.2 0 1000 Processing: 2 276 416.2 145 1550 Waiting: 0 271 416.3 140 1549 Total: 16 285 422.3 146 1561 Percentage of the requests served within a certain time (ms) 50% 146 66% 150 75% 152 80% 153 90% 1443 95% 1512 98% 1543 99% 1553 100% 1561 (longest request) 笺注：可以看到每秒最大处理请求次数为638.18次，每个请求响应时间是1.567毫秒。 ####### 假如MaxRequestWorkers的值超过256： [root@localhost ~]# vi /usr/local/apache/conf/extra/httpd-mpm.conf prefork MPM： 图片2.png [root@localhost ~]# service httpd restart restart apache... AH00180: WARNING: MaxRequestWorkers of 300 exceeds ServerLimit value of 256 servers, decreasing MaxRequestWorkers to 256. To increase, please see the ServerLimit directive. done 修改ServerLimit的值为300： prefork MPM： StartServers 5 MinSpareServers 5 MaxSpareServers 10 ServerLimit 300 MaxRequestWorkers 300 MaxConnectionsPerChild 0 注释： ServerLimit表示Apache允许创建的最大进程数，默认为256； MaxRequestWorkers的值不可以大于ServerLimit的值； ServerLimit、MaxRequestWorkers的最大值为200000，超过20万的话，在重启Apache时会报错； 在服务器资源（CPU、内存等等）充足的情况下，ServerLimit、MaxRequestWorkers的值可以调大一些； [root@localhost ~]# service httpd restart restart apache... done 加大并发请求数，再次测试： [root@localhost ~]# ab -n 1000 -c 300 http://192.168.168.130/phpinfo.php #指定并发请求数为300，总请求数为1000，对http://192.168.168.130/phpinfo.php进行压力测试 This is ApacheBench, Version 2.3 <$Revision: 655654 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking 192.168.168.130 (be patient) Completed 100 requests Completed 200 requests Completed 300 requests Completed 400 requests Completed 500 requests Completed 600 requests Completed 700 requests Completed 800 requests Completed 900 requests Completed 1000 requests Finished 1000 requests Server Software: Apache Server Hostname: 192.168.168.130 Server Port: 80 Document Path: /phpinfo.php Document Length: 87189 bytes Concurrency Level: 300 Time taken for tests: 13.990 seconds Complete requests: 1000 Failed requests: 94 (Connect: 0, Receive: 0, Length: 94, Exceptions: 0) Write errors: 0 Total transferred: 87498966 bytes HTML transferred: 87319250 bytes Requests per second: 71.48 [#/sec] (mean) Time per request: 4197.027 [ms] (mean) Time per request: 13.990 [ms] (mean, across all concurrent requests) Transfer rate: 6107.77 [Kbytes/sec] received Connection Times (ms) min mean[+/-sd] median max Connect: 0 5 63.1 0 1000 Processing: 1 1167 2763.6 144 13980 Waiting: 0 1163 2763.8 139 13979 Total: 15 1173 2763.3 144 13983 Percentage of the requests served within a certain time (ms) 50% 144 66% 149 75% 151 80% 154 90% 4719 95% 4773 98% 13962 99% 13975 100% 13983 (longest request) 笺注：可以看到每秒最大处理请求次数为71.48次，每个请求响应时间是13.990毫秒。 加大并发请求数，再次测试： [root@localhost ~]# ab -n 1000 -c 400 http://192.168.168.130/phpinfo.php #指定并发请求数为400，总请求数为1000，对http://192.168.168.130/phpinfo.php进行压力测试 This is ApacheBench, Version 2.3 <$Revision: 655654 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking 192.168.168.130 (be patient) Completed 100 requests Completed 200 requests Completed 300 requests Completed 400 requests Completed 500 requests Completed 600 requests Completed 700 requests apr_poll: The timeout specified has expired (70007) Total of 759 requests completed 相关文章： Apache2.4的MPM模式（event/prefork） Nginx的并发连接数

图片附件: 图片1.png (2021-3-16 00:40, 144.23 KB) / 下载次数 114
http://blog.zhuohua.store/attachment.php?aid=15938&k=bdf9712d85ebac0f285d9d459c11bcba&t=1714257021&sid=nkAdtT



图片附件: 图片2.png (2021-3-16 00:42, 47.69 KB) / 下载次数 117
http://blog.zhuohua.store/attachment.php?aid=15939&k=cae2a09923c35428496fa22e85c02f0f&t=1714257021&sid=nkAdtT





欢迎光临 blog.zhuohua.store (http://blog.zhuohua.store/) Powered by Discuz! 7.2