Details about using redis to store session instances in php

Phpinfo, you can see the session storage, you can use files user memcache redis, the advantage of using the database to store sessions is that compared with the file storage, the speed is faster and the performance is better in a large number of users, in addition, for a distributed system, you must use a database to store sessions. Here we will summarize the two methods for storing session in redis.
We should have used mysql to store Sessions. The principle of the session lifecycle is actually determined by the two parameters session. gcprobability, session. gcdivisor and the maximum life time. Each php request has a certain probability of triggering the session detection mechanism. Session_setsavehandler can be used to redefine session behavior. There are two ways to implement redis storage session

Code Example 1:

The code is as follows:

Copy code

<? Php Ini_set ('session. Gc_maxlifetime', 3600 ); Ini_set ("session. save_handler", "redis "); Ini_set ("session. save_path", "tcp: // fig: 6379? Auth = authpwd "); Session_start (); // $ _ SESSION ['session'] = 'This is session content! '; Echo $ _ SESSION ['session ']; Echo session_id (). '<br/> ';   $ Redis = new redis (); $ Redis-> connect ('2017. 0.0.1 ', 127 ); // Redis uses session_id as the key and stores it as a string Echo $ redis-> get ('phpredis _ SESSION: '. session_id ()); ?>

Code Example 2:

The code is as follows:

Copy code

<? Php   Class RedisSessionHandler { Public $ ttl = 1800; // 30 minutes default Protected $ db; Protected $ prefix; Public function _ construct ($ db, $ prefix = 'phpsessid1: ', $ time = 1800 ){ $ This-> db = $ db; $ This-> prefix = $ prefix; $ This-> ttl = $ time;     } Function _ open ()     { //     }   Function _ close ()     { $ This-> db = null; Unset ($ this-> db );     }   Function _ read ($ id)     { $ Id = $ this-> prefix. $ id; $ SessData = $ this-> db-> get ($ id ); $ This-> db-> expire ($ id, $ this-> ttl ); Return $ sessData;     }   Function _ write ($ id, $ data)     { $ Id = $ this-> prefix. $ id; $ This-> db-> set ($ id, $ data ); $ This-> db-> expire ($ id, $ this-> ttl );     }   Function _ destroy ($ id)     { $ This-> db-> del ($ this-> prefix. $ id );     }   Function _ clean ($ max)     { //     } } $ Redis = new Redis (); $ Redis-> connect ('2017. 0.0.1 ', 127 ); $ Redis-> select (1 ); $ SESSION_ID_PREFIX = 'rsid :'; $ SESSION_MAX_TIME = 1440; Ini_set ('session. gc_maxlifetime ', $ SESSION_MAX_TIME ); $ SessHandler = new RedisSessionHandler ($ redis, $ SESSION_ID_PREFIX, $ SESSION_MAX_TIME ); Session_set_save_handler ( Array ($ sessHandler, '_ open '), Array ($ sessHandler, '_ close '), Array ($ sessHandler, '_ READ '), Array ($ sessHandler, '_ write '), Array ($ sessHandler, '_ destroy '), Array ($ sessHandler, '_ clean ') );   Session_start (); Echo session_id (); Echo $ _ SESSION ['name'] = 'hangzhou ';

> The first method is relatively simple, but the database number cannot be selected, and the prefix information of session_id is difficult to control. However, this method can also directly modify the php configuration file so that sessions of all programs use redis.
> The second method is relatively troublesome, but it is easy to use.
> When we talk about concurrency consistency on the Internet, all hash algorithms will certainly have this problem, but it can be solved to some extent through the ip Prefix hash combination.