php Let Swoole | Pool process pool implement Redis persistent connection
This module implements the process pool based on the Manager process management of Swoole\Server. It can manage multiple work processes. Compared with Process to implement multiple processes, Process\Pool is simpler and has a higher encapsulation level. Developers can implement process management functions without writing too much code. With Co\Server, pure coroutine style can be created. , a server-side program that can utilize multi-core CPUs.
Swoole process pool implements redis data reading
In the following case, the Redis process pool is started through WorkerStart and the Redis list data is read persistently; when WorkerStop disconnects all connections Recycle all child processes.
Step one: Encoding code
File: d10.php
<?php use Swoole\Process; use Swoole\Coroutine; // 指定5个工作进程 $pool = new Process\Pool(5); // 设置启用协程 $pool->set(['enable_coroutine' => true]); /** * onWorkerStart 子进程启动 * @param \Swoole\Process\Pool $pool Pool对象 * @param int $workerId WorkerId当前工作进程的编号,底层会对子进程进行标号 **/ $pool->on("WorkerStart", function (Process\Pool $pool, $workerId) { // 输出当前工作进程 echo "Worker #{$workerId} is started\n"; // 实例化化连接redis $redis = new Redis(); $redis->pconnect('127.0.0.1', 6379); // 指定redis键 $key = "key1"; // 循环读取列表数据 while (true) { // 弹出列表最后一个元素 $msgs = $redis->brpop($key, 2); // 元素值为空则跳过 if ( $msgs == null) { continue; } // 打印获取的值 var_dump($msgs); echo "Processed by Worker#{$workerId}\n"; } }); // 子进程结束 $pool->on("WorkerStop", function ($pool, $workerId) { echo "Worker#{$workerId} is stopped\n"; }); // 启动工作进程 $pool->start();
Step two: Start the Redis service and write list data through the client
This case requires PHP to install the redis extension
# 通过redis客户端连接 ./redis-cli 127.0.0.1:6379> lpush key1 'world'
Copy code
Step 3: Run d10.php
php d10.php
Step 4: View the process
ps aux | grep php root 938 0.0 1.2 129164 12412 ? Ss Apr21 0:00 php-fpm: master process (/usr/local/php-8.0.1/etc/php-fpm.conf) www 951 0.0 0.6 129164 6636 ? S Apr21 0:00 php-fpm: pool www www 952 0.0 0.6 129164 6640 ? S Apr21 0:00 php-fpm: pool www root 12327 0.0 1.2 126992 12800 pts/2 S+ 00:02 0:00 php d10.php root 12328 0.0 0.7 131096 7444 pts/2 S+ 00:02 0:00 php d10.php root 12329 0.0 0.7 131096 7448 pts/2 S+ 00:02 0:00 php d10.php root 12330 0.0 0.7 131096 7448 pts/2 S+ 00:02 0:00 php d10.php root 12331 0.0 0.7 131096 7448 pts/2 S+ 00:02 0:00 php d10.php root 12332 0.0 0.7 131096 7448 pts/2 S+ 00:02 0:00 php d10.php root 12355 0.0 0.0 112812 976 pts/3 R+ 00:09 0:00 grep --color=auto php
Step 5: Output the result
php d10.php Worker #1 is started Worker #2 is started Worker #3 is started Worker #4 is started Worker #0 is started array(2) { [0]=> string(4) "key1" [1]=> string(5) "world" } Processed by Worker#1
d10.php file will always be blocked after running and keep reading the redis list data. Once the Redis list is output, it will be immediately popped up and printed on the screen.
The above is the detailed content of How does php enable the Swoole/Pool process pool to implement Redis persistent connections?. For more information, please follow other related articles on the PHP Chinese website!