首页 > 资讯列表 > 编程/数据库 >> PHP

Windows下Redis的安装及PHP扩展使用

PHP 2014-11-30 00:15:57 转载来源: 网络整理/侵权必删

本文为大家讲解的是在Windows下安装Redis和PHP扩展及简单使用方法,感兴趣的同学参考下。 1、下载redis的windows应用程序,支持32位和64位,根据实际情况下载 下载地址:https://github.com/dmajkic/redis/downloads 2、将相应的程序copy到你所需要的目录中,在这里我使用的64位,放到E:\redis目录 3、启动redis服务端: 打开一个cmd窗口,先切换到redis所放目录(E:\redis),运行 redis-server.exe redis.conf  注意redis.conf为配置文件,主要配置了redis所使用的端口等信息(如果不写则默认redis.conf) 有的下载的redis压缩包里没有redis.conf,我把默认的redis.conf的文件内容放在文章最后

本文为大家讲解的是在Windows安装RedisPHP扩展及简单使用方法,感兴趣的同学参考下。

1、下载redis的windows应用程序,支持32位和64位,根据实际情况下载

下载地址:https://github.com/dmajkic/redis/downloads

2、将相应的程序copy到你所需要的目录中,在这里我使用的64位,放到E:redis目录

3、启动redis服务端

打开一个cmd窗口,先切换到redis所放目录(E:redis),运行 redis-server.exe redis.conf 

注意redis.conf为配置文件,主要配置了redis所使用的端口等信息(如果不写则默认redis.conf)

有的下载的redis压缩包里没有redis.conf,我把默认的redis.conf的文件内容放在文章最后。

注意:此窗口为redis服务端运行窗口,关闭后则redis关闭。

4、启动redis客户端:另开一个cmd窗口,进入目录之后运行命令redis-cli.exe -h 127.0.0.1 -p 6379,然后就可以进行操作了

5、下载redis的php扩展

下载地址:https://github.com/nicolasff/phpredis/downloads

根据php的版本来下载相应的扩展,版本必须对应

6、将php_redis.dll放入php的ext文件夹中,然后再php.ini添加代码extension=php_redis.dll

7、重启web服务器

8、php测试

<?php  
    $redis = new Redis();  
    $redis->connect('127.0.0.1',6379);  
    $redis->set('test','hello redis');  
    echo $redis->get('test');  
?>

9、附:默认redis.conf文件内容

# Redis configuration file example  
 
# Note on units: when memory size is needed, it is possible to specifiy  
# it in the usual form of 1k 5GB 4M and so forth:  
#  
# 1k => 1000 bytes  
# 1kb => 1024 bytes  
# 1m => 1000000 bytes  
# 1mb => 1024*1024 bytes  
# 1g => 1000000000 bytes  
# 1gb => 1024*1024*1024 bytes  
#  
# units are case insensitive so 1GB 1Gb 1gB are all the same.  
 
# By default Redis does not run as a daemon. Use 'yes' if you need it.  
# Note that Redis will write a pid file in /var/run/redis.pid when daemonized.  
daemonize no  
 
# When running daemonized, Redis writes a pid file in /var/run/redis.pid by  
# default. You can specify a custom pid file location here.  
pidfile /var/run/redis.pid  
 
# Accept connections on the specified port, default is 6379.  
# If port 0 is specified Redis will not listen on a TCP socket.  
port 6379  
 
# If you want you can bind a single interface, if the bind option is not  
# specified all the interfaces will listen for incoming connections.  
#  
# bind 127.0.0.1  
 
# Specify the path for the unix socket that will be used to listen for  
# incoming connections. There is no default, so Redis will not listen  
# on a unix socket when not specified.  
#  
# unixsocket /tmp/redis.sock  
# unixsocketperm 755  
 
# Close the connection after a client is idle for N seconds (0 to disable)  
timeout 0  
 
# Set server verbosity to 'debug'  
# it can be one of:  
# debug (a lot of information, useful for development/testing)  
# verbose (many rarely useful info, but not a mess like the debug level)  
# notice (moderately verbose, what you want in production probably)  
# warning (only very important / critical messages are logged)  
loglevel verbose  
 
# Specify the log file name. Also 'stdout' can be used to force  
# Redis to log on the standard output. Note that if you use standard  
# output for logging but daemonize, logs will be sent to /dev/null  
logfile stdout  
 
# To enable logging to the system logger, just set 'syslog-enabled' to yes,  
# and optionally update the other syslog parameters to suit your needs.  
# syslog-enabled no  
 
# Specify the syslog identity.  
# syslog-ident redis  
 
# Specify the syslog facility.  Must be USER or between LOCAL0-LOCAL7.  
# syslog-facility local0  
 
# Set the number of databases. The default database is DB 0, you can select  
# a different one on a per-connection basis using SELECT <dbid> where  
# dbid is a number between 0 and 'databases'-1  
databases 16  
 
################################ SNAPSHOTTING  #################################  
#  
# Save the DB on disk:  
#  
#   save <seconds> <changes>  
#  
#   Will save the DB if both the given number of seconds and the given  
#   number of write operations against the DB occurred.  
#  
#   In the example below the behaviour will be to save:  
#   after 900 sec (15 min) if at least 1 key changed  
#   after 300 sec (5 min) if at least 10 keys changed  
#   after 60 sec if at least 10000 keys changed  
#  
#   Note: you can disable saving at all commenting all the "save" lines.  
 
save 900 1  
save 300 10  
save 60 10000  
 
# Compress string objects using LZF when dump .rdb databases?  
# For default that's set to 'yes' as it's almost always a win.  
# If you want to save some CPU in the saving child set it to 'no' but  
# the dataset will likely be bigger if you have compressible values or keys.  
rdbcompression yes  
 
# The filename where to dump the DB  
dbfilename dump.rdb  
 
# The working directory.  
#  
# The DB will be written inside this directory, with the filename specified  
# above using the 'dbfilename' configuration directive.  
#   
# Also the Append Only File will be created inside this directory.  
#   
# Note that you must specify a directory here, not a file name.  
dir ./  
 
################################# REPLICATION #################################  
 
# Master-Slave replication. Use slaveof to make a Redis instance a copy of  
# another Redis server. Note that the configuration is local to the slave  
# so for example it is possible to configure the slave to save the DB with a  
# different interval, or to listen to another port, and so on.  
#  
# slaveof <masterip> <masterport>  
 
# If the master is password protected (using the "requirepass" configuration  
# directive below) it is possible to tell the slave to authenticate before  
# starting the replication synchronization process, otherwise the master will  
# refuse the slave request.  
#  
# masterauth <master-password>  
 
# When a slave lost the connection with the master, or when the replication  
# is still in progress, the slave can act in two different ways:  
#  
# 1) if slave-serve-stale-data is set to 'yes' (the default) the slave will  
#    still reply to client requests, possibly with out of data data, or the  
#    data set may just be empty if this is the first synchronization.  
#  
# 2) if slave-serve-stale data is set to 'no' the slave will reply with  
#    an error "SYNC with master in progress" to all the kind of commands  
#    but to INFO and SLAVEOF.  
#  
slave-serve-stale-data yes  
 
# Slaves send PINGs to server in a predefined interval. It's possible to change  
# this interval with the repl_ping_slave_period option. The default value is 10  
# seconds.  
#  
# repl-ping-slave-period 10  
 
# The following option sets a timeout for both Bulk transfer I/O timeout and  
# master data or ping response timeout. The default value is 60 seconds.  
#  
# It is important to make sure that this value is greater than the value  
# specified for repl-ping-slave-period otherwise a timeout will be detected  
# every time there is low traffic between the master and the slave.  
#  
# repl-timeout 60  
 
################################## SECURITY ###################################  
 
# Require clients to issue AUTH <PASSWORD> before processing any other  
# commands.  This might be useful in environments in which you do not trust  
# others with access to the host running redis-server.  
#  
# This should stay commented out for backward compatibility and because most  
# people do not need auth (e.g. they run their own servers).  
#   
# Warning: since Redis is pretty fast an outside user can try up to  
# 150k passwords per second against a good box. This means that you should  
# use a very strong password otherwise it will be very easy to break.  
#  
# requirepass foobared  
 
# Command renaming.  
#  
# It is possilbe to change the name of dangerous commands in a shared  
# environment. For instance the CONFIG command may be renamed into something  
# of hard to guess so that it will be still available for internal-use  
# tools but not available for general clients.  
#  
# Example:  
#  
# rename-command CONFIG b840fc02d524045429941cc15f59e41cb7be6c52  
#  
# It is also possilbe to completely kill a command renaming it into  
# an empty string:  
#  
# rename-command CONFIG ""  
 
################################### LIMITS ####################################  
 
# Set the max number of connected clients at the same time. By default there  
# is no limit, and it's up to the number of file descriptors the Redis process  
# is able to open. The special value '0' means no limits.  
# Once the limit is reached Redis will close all the new connections sending  
# an error 'max number of clients reached'.  
#  
# maxclients 128  
 
# Don't use more memory than the specified amount of bytes.  
# When the memory limit is reached Redis will try to remove keys with an  
# EXPIRE set. It will try to start freeing keys that are going to expire  
# in little time and preserve keys with a longer time to live.  
# Redis will also try to remove objects from free lists if possible.  
#  
# If all this fails, Redis will start to reply with errors to commands  
# that will use more memory, like SET, LPUSH, and so on, and will continue  
# to reply to most read-only commands like GET.  
#  
# WARNING: maxmemory can be a good idea mainly if you want to use Redis as a  
# 'state' server or cache, not as a real DB. When Redis is used as a real  
# database the memory usage will grow over the weeks, it will be obvious if  
# it is going to use too much memory in the long run, and you'll have the time  
# to upgrade. With maxmemory after the limit is reached you'll start to get  
# errors for write operations, and this may even lead to DB inconsistency.  
#  
# maxmemory <bytes>  
 
# MAXMEMORY POLICY: how Redis will select what to remove when maxmemory  
# is reached? You can select among five behavior:  
#   
# volatile-lru -> remove the key with an expire set using an LRU algorithm  
# allkeys-lru -> remove any key accordingly to the LRU algorithm  
# volatile-random -> remove a random key with an expire set  
# allkeys->random -> remove a random key, any key  
# volatile-ttl -> remove the key with the nearest expire time (minor TTL)  
# noeviction -> don't expire at all, just return an error on write operations  
#   
# Note: with all the kind of policies, Redis will return an error on write  
#       operations, when there are not suitable keys for eviction.  
#  
#       At the date of writing this commands are: set setnx setex append  
#       incr decr rpush lpush rpushx lpushx linsert lset rpoplpush sadd  
#       sinter sinterstore sunion sunionstore sdiff sdiffstore zadd zincrby  
#       zunionstore zinterstore hset hsetnx hmset hincrby incrby decrby  
#       getset mset msetnx exec sort  
#  
# The default is:  
#  
# maxmemory-policy volatile-lru  
 
# LRU and minimal TTL algorithms are not precise algorithms but approximated  
# algorithms (in order to save memory), so you can select as well the sample  
# size to check. For instance for default Redis will check three keys and  
# pick the one that was used less recently, you can change the sample size  
# using the following configuration directive.  
#  
# maxmemory-samples 3  
 
############################## APPEND ONLY MODE ###############################  
 
# By default Redis asynchronously dumps the dataset on disk. If you can live  
# with the idea that the latest records will be lost if something like a crash  
# happens this is the preferred way to run Redis. If instead you care a lot  
# about your data and don't want to that a single record can get lost you should  
# enable the append only mode: when this mode is enabled Redis will append  
# every write operation received in the file appendonly.aof. This file will  
# be read on startup in order to rebuild the full dataset in memory.  
#  
# Note that you can have both the async dumps and the append only file if you  
# like (you have to comment the "save" statements above to disable the dumps).  
# Still if append only mode is enabled Redis will load the data from the  
# log file at startup ignoring the dump.rdb file.  
#  
# IMPORTANT: Check the BGREWRITEAOF to check how to rewrite the append  
# log file in background when it gets too big.  
 
appendonly no  
 
# The name of the append only file (default: "appendonly.aof")  
# appendfilename appendonly.aof  
 
# The fsync() call tells the Operating System to actually write data on disk  
# instead to wait for more data in the output buffer. Some OS will really flush   
# data on disk, some other OS will just try to do it ASAP.  
#  
# Redis supports three different modes:  
#  
# no: don't fsync, just let the OS flush the data when it wants. Faster.  
# always: fsync after every write to the append only log . Slow, Safest.  
# everysec: fsync only if one second passed since the last fsync. Compromise.  
#  
# The default is "everysec" that's usually the right compromise between  
# speed and data safety. It's up to you to understand if you can relax this to  
# "no" that will will let the operating system flush the output buffer when  
# it wants, for better performances (but if you can live with the idea of  
# some data loss consider the default persistence mode that's snapshotting),  
# or on the contrary, use "always" that's very slow but a bit safer than  
# everysec.  
#  
# If unsure, use "everysec".  
 
# appendfsync always  
appendfsync everysec  
# appendfsync no  
 
# When the AOF fsync policy is set to always or everysec, and a background  
# saving process (a background save or AOF log background rewriting) is  
# performing a lot of I/O against the disk, in some Linux configurations  
# Redis may block too long on the fsync() call. Note that there is no fix for  
# this currently, as even performing fsync in a different thread will block  
# our synchronous write(2) call.  
#  
# In order to mitigate this problem it's possible to use the following option  
# that will prevent fsync() from being called in the main process while a  
# BGSAVE or BGREWRITEAOF is in progress.  
#  
# This means that while another child is saving the durability of Redis is  
# the same as "appendfsync none", that in pratical terms means that it is  
# possible to lost up to 30 seconds of log in the worst scenario (with the  
# default Linux settings).  
#   
# If you have latency problems turn this to "yes". Otherwise leave it as  
# "no" that is the safest pick from the point of view of durability.  
no-appendfsync-on-rewrite no  
 
# Automatic rewrite of the append only file.  
# Redis is able to automatically rewrite the log file implicitly calling  
# BGREWRITEAOF when the AOF log size will growth by the specified percentage.  
#   
# This is how it works: Redis remembers the size of the AOF file after the  
# latest rewrite (or if no rewrite happened since the restart, the size of  
# the AOF at startup is used).  
#  
# This base size is compared to the current size. If the current size is  
# bigger than the specified percentage, the rewrite is triggered. Also  
# you need to specify a minimal size for the AOF file to be rewritten, this  
# is useful to avoid rewriting the AOF file even if the percentage increase  
# is reached but it is still pretty small.  
#  
# Specify a precentage of zero in order to disable the automatic AOF  
# rewrite feature.  
 
auto-aof-rewrite-percentage 100  
auto-aof-rewrite-min-size 64mb  
 
################################## SLOW LOG ###################################  
 
# The Redis Slow Log is a system to log queries that exceeded a specified  
# execution time. The execution time does not include the I/O operations  
# like talking with the client, sending the reply and so forth,  
# but just the time needed to actually execute the command (this is the only  
# stage of command execution where the thread is blocked and can not serve  
# other requests in the meantime).  
#   
# You can configure the slow log with two parameters: one tells Redis  
# what is the execution time, in microseconds, to exceed in order for the  
# command to get logged, and the other parameter is the length of the  
# slow log. When a new command is logged the oldest one is removed from the  
# queue of logged commands.  
 
# The following time is expressed in microseconds, so 1000000 is equivalent  
# to one second. Note that a negative number disables the slow log, while  
# a value of zero forces the logging of every command.  
slowlog-log-slower-than 10000  
 
# There is no limit to this length. Just be aware that it will consume memory.  
# You can reclaim memory used by the slow log with SLOWLOG RESET.  
slowlog-max-len 1024  
 
################################ VIRTUAL MEMORY ###############################  
 
### WARNING! Virtual Memory is deprecated in Redis 2.4  
### The use of Virtual Memory is strongly discouraged.  
 
### WARNING! Virtual Memory is deprecated in Redis 2.4  
### The use of Virtual Memory is strongly discouraged.  
 
# Virtual Memory allows Redis to work with datasets bigger than the actual  
# amount of RAM needed to hold the whole dataset in memory.  
# In order to do so very used keys are taken in memory while the other keys  
# are swapped into a swap file, similarly to what operating systems do  
# with memory pages.  
#  
# To enable VM just set 'vm-enabled' to yes, and set the following three  
# VM parameters accordingly to your needs.  
 
vm-enabled no  
# vm-enabled yes  
 
# This is the path of the Redis swap file. As you can guess, swap files  
# can't be shared by different Redis instances, so make sure to use a swap  
# file for every redis process you are running. Redis will complain if the  
# swap file is already in use.  
#  
# The best kind of storage for the Redis swap file (that's accessed at random)   
# is a Solid State Disk (SSD).  
#  
# *** WARNING *** if you are using a shared hosting the default of putting  
# the swap file under /tmp is not secure. Create a dir with access granted  
# only to Redis user and configure Redis to create the swap file there.  
vm-swap-file /tmp/redis.swap  
 
# vm-max-memory configures the VM to use at max the specified amount of  
# RAM. Everything that deos not fit will be swapped on disk *if* possible, that  
# is, if there is still enough contiguous space in the swap file.  
#  
# With vm-max-memory 0 the system will swap everything it can. Not a good  
# default, just specify the max amount of RAM you can in bytes, but it's  
# better to leave some margin. For instance specify an amount of RAM  
# that's more or less between 60 and 80% of your free RAM.  
vm-max-memory 0  
 
# Redis swap files is split into pages. An object can be saved using multiple  
# contiguous pages, but pages can't be shared between different objects.  
# So if your page is too big, small objects swapped out on disk will waste  
# a lot of space. If you page is too small, there is less space in the swap  
# file (assuming you configured the same number of total swap file pages).  
#  
# If you use a lot of small objects, use a page size of 64 or 32 bytes.  
# If you use a lot of big objects, use a bigger page size.  
# If unsure, use the default :)  
vm-page-size 32  
 
# Number of total memory pages in the swap file.  
# Given that the page table (a bitmap of free/used pages) is taken in memory,  
# every 8 pages on disk will consume 1 byte of RAM.  
#  
# The total swap size is vm-page-size * vm-pages  
#  
# With the default of 32-bytes memory pages and 134217728 pages Redis will  
# use a 4 GB swap file, that will use 16 MB of RAM for the page table.  
#  
# It's better to use the smallest acceptable value for your application,  
# but the default is large in order to work in most conditions.  
vm-pages 134217728  
 
# Max number of VM I/O threads running at the same time.  
# This threads are used to read/write data from/to swap file, since they  
# also encode and decode objects from disk to memory or the reverse, a bigger  
# number of threads can help with big objects even if they can't help with  
# I/O itself as the physical device may not be able to couple with many  
# reads/writes operations at the same time.  
#  
# The special value of 0 turn off threaded I/O and enables the blocking  
# Virtual Memory implementation.  
vm-max-threads 4  
 
############################### ADVANCED CONFIG ###############################  
 
# Hashes are encoded in a special way (much more memory efficient) when they  
# have at max a given numer of elements, and the biggest element does not  
# exceed a given threshold. You can configure this limits with the following  
# configuration directives.  
hash-max-zipmap-entries 512  
hash-max-zipmap-value 64  
 
# Similarly to hashes, small lists are also encoded in a special way in order  
# to save a lot of space. The special representation is only used when  
# you are under the following limits:  
list-max-ziplist-entries 512  
list-max-ziplist-value 64  
 
# Sets have a special encoding in just one case: when a set is composed  
# of just strings that happens to be integers in radix 10 in the range  
# of 64 bit signed integers.  
# The following configuration setting sets the limit in the size of the  
# set in order to use this special memory saving encoding.  
set-max-intset-entries 512  
 
# Similarly to hashes and lists, sorted sets are also specially encoded in  
# order to save a lot of space. This encoding is only used when the length and  
# elements of a sorted set are below the following limits:  
zset-max-ziplist-entries 128  
zset-max-ziplist-value 64  
 
# Active rehashing uses 1 millisecond every 100 milliseconds of CPU time in  
# order to help rehashing the main Redis hash table (the one mapping top-level  
# keys to values). The hash table implementation redis uses (see dict.c)  
# performs a lazy rehashing: the more operation you run into an hash table  
# that is rhashing, the more rehashing "steps" are performed, so if the  
# server is idle the rehashing is never complete and some more memory is used  
# by the hash table.  
#   
# The default is to use this millisecond 10 times every second in order to  
# active rehashing the main dictionaries, freeing memory when possible.  
#  
# If unsure:  
# use "activerehashing no" if you have hard latency requirements and it is  
# not a good thing in your environment that Redis can reply form time to time  
# to queries with 2 milliseconds delay.  
#  
# use "activerehashing yes" if you don't have such hard requirements but  
# want to free memory asap when possible.  
activerehashing yes  
 
################################## INCLUDES ###################################  
 
# Include one or more other config files here.  This is useful if you  
# have a standard template that goes to all redis server but also need  
# to customize a few per-server settings.  Include files can include  
# other files, so use this wisely.  
#  
# include /path/to/local.conf  
# include /path/to/other.conf


标签: Windows Redis 安装 PHP 扩展 使用


声明:本文内容来源自网络,文字、图片等素材版权属于原作者,平台转载素材出于传递更多信息,文章内容仅供参考与学习,切勿作为商业目的使用。如果侵害了您的合法权益,请您及时与我们联系,我们会在第一时间进行处理!我们尊重版权,也致力于保护版权,站搜网感谢您的分享!

站长搜索

http://www.adminso.com

Copyright @ 2007~2024 All Rights Reserved.

Powered By 站长搜索

打开手机扫描上面的二维码打开手机版


使用手机软件扫描微信二维码

关注我们可获取更多热点资讯

站长搜索目录系统技术支持