【转】Redis安装整理(window平台和Linux平台)

原文连接:http://zheng12tian.iteye.com/blog/1471726

原文作者:zheng12tian

转载注明以上信息!

window平台Redis安装

redis windows安装文件下载地址:http://code.google.com/p/servicestack/wiki/RedisWindowsDownload#Download_32bit_Cygwin_builds_for_Windows
我选择的redis为最新版的安装文件，见下图：


Redis安装文件解压后，有以下几个文件。见下图

redis-server.exe：服务程序
redis-check-dump.exe：本地数据库检查
redis-check-aof.exe：更新日志检查
redis-benchmark.exe：性能测试，用以模拟同时由N个客户端发送M个 SETs/GETs 查询 (类似于 Apache 的ab 工具).

在解压好redis的安装文件到E:\根目录后，还需要在redis根目录增加一个redis的配置文件redis.conf，文件具体内容附件中有，不过这里我仍然把配置文件的内容贴上来：

Java代码 收藏代码
  1. # Redis configuration file example 
  2.  
  3. # By default Redis does not run as a daemon. Use 'yes'if you need it. 
  4. # Note that Redis will write a pid file in /var/run/redis.pid when daemonized. 
  5. daemonize no 
  6.  
  7. # When run as a daemon, Redis write a pid file in /var/run/redis.pid by default. 
  8. # You can specify a custom pid file location here. 
  9. pidfile /var/run/redis.pid 
  10.  
  11. # Accept connections on the specified port, default is 6379 
  12. port 6379 
  13.  
  14. # If you want you can bind a single interface, if the bind option is not 
  15. # specified all the interfaces will listen for connections. 
  16. # 
  17. # bind 127.0.0.1 
  18.  
  19. # Close the connection after a client is idle for N seconds (0 to disable) 
  20. timeout 300 
  21.  
  22. # Set server verbosity to 'debug' 
  23. # it can be one of: 
  24. # debug (a lot of information, useful for development/testing) 
  25. # notice (moderately verbose, what you want in production probably) 
  26. # warning (only very important / critical messages are logged) 
  27. loglevel debug 
  28.  
  29. # Specify the log file name. Also 'stdout' can be used to force 
  30. # the demon to log on the standard output. Note that if you use standard 
  31. # output for logging but daemonize, logs will be sent to /dev/null 
  32. logfile stdout 
  33.  
  34. # Set the number of databases. The default database is DB 0, you can select 
  35. # a different one on a per-connection basis using SELECT <dbid> where 
  36. # dbid is a number between 0 and 'databases'-1 
  37. databases 16 
  38.  
  39. ################################ SNAPSHOTTING  ################################# 
  40. # 
  41. # Save the DB on disk: 
  42. # 
  43. #   save <seconds> <changes> 
  44. # 
  45. #   Will save the DB if both the given number of seconds and the given 
  46. #   number of write operations against the DB occurred. 
  47. # 
  48. #   In the example below the behaviour will be to save: 
  49. #   after 900 sec (15 min) if at least 1 key changed 
  50. #   after 300 sec (5 min) if at least 10 keys changed 
  51. #   after 60 sec if at least 10000 keys changed 
  52. save 9001 
  53. save 30010 
  54. save 6010000 
  55.  
  56. # Compress string objects using LZF when dump .rdb databases? 
  57. # For default that's set to 'yes' as it's almost always a win. 
  58. # If you want to save some CPU in the saving child set it to 'no' but 
  59. # the dataset will likely be bigger if you have compressible values or keys. 
  60. rdbcompression yes 
  61.  
  62. # The filename where to dump the DB 
  63. dbfilename dump.rdb 
  64.  
  65. # For default save/load DB in/from the working directory 
  66. # Note that you must specify a directory not a file name. 
  67. dir ./ 
  68.  
  69. ################################# REPLICATION ################################# 
  70.  
  71. # Master-Slave replication. Use slaveof to make a Redis instance a copy of 
  72. # another Redis server. Note that the configuration is local to the slave 
  73. # so for example it is possible to configure the slave to save the DB with a 
  74. # different interval, or to listen to another port, and so on. 
  75. # 
  76. # slaveof <masterip> <masterport> 
  77.  
  78. # If the master is password protected (using the "requirepass" configuration 
  79. # directive below) it is possible to tell the slave to authenticate before 
  80. # starting the replication synchronization process, otherwise the master will 
  81. # refuse the slave request. 
  82. # 
  83. # masterauth <master-password> 
  84.  
  85. ################################## SECURITY ################################### 
  86.  
  87. # Require clients to issue AUTH <PASSWORD> before processing any other 
  88. # commands.  This might be useful in environments in which you do not trust 
  89. # others with access to the host running redis-server. 
  90. # 
  91. # This should stay commented out for backward compatibility and because most 
  92. # people do not need auth (e.g. they run their own servers). 
  93. # 
  94. # requirepass foobared 
  95.  
  96. ################################### LIMITS #################################### 
  97.  
  98. # Set the max number of connected clients at the same time. By default there 
  99. # is no limit, and it's up to the number of file descriptors the Redis process 
  100. # is able to open. The special value '0' means no limts. 
  101. # Once the limit is reached Redis will close all the new connections sending 
  102. # an error 'max number of clients reached'. 
  103. # 
  104. # maxclients 128 
  105.  
  106. # Don't use more memory than the specified amount of bytes. 
  107. # When the memory limit is reached Redis will try to remove keys with an 
  108. # EXPIRE set. It will try to start freeing keys that are going to expire 
  109. # in little time and preserve keys with a longer time to live. 
  110. # Redis will also try to remove objects from free lists if possible. 
  111. # 
  112. # If all this fails, Redis will start to reply with errors to commands 
  113. # that will use more memory, like SET, LPUSH, and so on, and will continue 
  114. # to reply to most read-only commands like GET. 
  115. # 
  116. # WARNING: maxmemory can be a good idea mainly if you want to use Redis as a 
  117. # 'state' server or cache, not as a real DB. When Redis is used as a real 
  118. # database the memory usage will grow over the weeks, it will be obvious if 
  119. # it is going to use too much memory in the long run, and you'll have the time 
  120. # to upgrade. With maxmemory after the limit is reached you'll start to get 
  121. # errors for write operations, and this may even lead to DB inconsistency. 
  122. # 
  123. # maxmemory <bytes> 
  124.  
  125. ############################## APPEND ONLY MODE ############################### 
  126.  
  127. # By default Redis asynchronously dumps the dataset on disk. If you can live 
  128. # with the idea that the latest records will be lost if something like a crash 
  129. # happens this is the preferred way to run Redis. If instead you care a lot 
  130. # about your data and don't want to that a single record can get lost you should 
  131. # enable the append only mode: when this mode is enabled Redis will append 
  132. # every write operation received in the file appendonly.log. This file will 
  133. # be read on startup in order to rebuild the full dataset in memory. 
  134. # 
  135. # Note that you can have both the async dumps and the append only file if you 
  136. # like (you have to comment the "save" statements above to disable the dumps). 
  137. # Still if append only mode is enabled Redis will load the data from the 
  138. # log file at startup ignoring the dump.rdb file. 
  139. # 
  140. # The name of the append only file is "appendonly.log" 
  141. # 
  142. # IMPORTANT: Check the BGREWRITEAOF to check how to rewrite the append 
  143. # log file in background when it gets too big. 
  144.  
  145. appendonly no 
  146.  
  147. # The fsync() call tells the Operating System to actually write data on disk 
  148. # instead to wait for more data in the output buffer. Some OS will really flush 
  149. # data on disk, some other OS will just try to do it ASAP. 
  150. # 
  151. # Redis supports three different modes: 
  152. # 
  153. # no: don't fsync, just let the OS flush the data when it wants. Faster. 
  154. # always: fsync after every write to the append only log . Slow, Safest. 
  155. # everysec: fsync only if one second passed since the last fsync. Compromise. 
  156. # 
  157. # The default is "always" that's the safer of the options. It's up to you to 
  158. # understand if you can relax this to "everysec" that will fsync every second 
  159. # or to "no" that will let the operating system flush the output buffer when 
  160. # it want, for better performances (but if you can live with the idea of 
  161. # some data loss consider the default persistence mode that's snapshotting). 
  162.  
  163. appendfsync always 
  164. # appendfsync everysec 
  165. # appendfsync no 
  166.  
  167. ############################### ADVANCED CONFIG ############################### 
  168.  
  169. # Glue small output buffers together in order to send small replies in a 
  170. # single TCP packet. Uses a bit more CPU but most of the times it is a win 
  171. # in terms of number of queries per second. Use 'yes'if unsure. 
  172. glueoutputbuf yes 
  173.  
  174. # Use object sharing. Can save a lot of memory if you have many common 
  175. # string in your dataset, but performs lookups against the shared objects 
  176. # pool so it uses more CPU and can be a bit slower. Usually it's a good 
  177. # idea. 
  178. # 
  179. # When object sharing is enabled (shareobjects yes) you can use 
  180. # shareobjectspoolsize to control the size of the pool used in order to try 
  181. # object sharing. A bigger pool size will lead to better sharing capabilities. 
  182. # In general you want this value to be at least the double of the number of 
  183. # very common strings you have in your dataset. 
  184. # 
  185. # WARNING: object sharing is experimental, don't enable this feature 
  186. # in production before of Redis 1.0-stable. Still please trythis feature in 
  187. # your development environment so that we can test it better. 
  188. # shareobjects no 
  189. # shareobjectspoolsize 1024 



将附件中的redis_conf.rar解压下来放到redis的根目录中即可。到此，redis的安装已经完毕。下面开始使用redis数据库。

启动redis：
输入命令：redis-server.exe redis.conf
启动后如下图所示：


启动cmd窗口要一直开着，关闭后则Redis服务关闭。
这时服务开启着，另外开一个窗口进行，设置客户端：
输入命令：redis-cli.exe -h 202.117.16.133 -p 6379
输入后如下图所示：

然后可以开始玩了：

设置一个Key并获取返回的值:

Java代码 收藏代码
  1. $ ./redis-cli set mykey somevalue 
  2. OK 
  3. $ ./redis-cli get mykey 
  4. Somevalue 



如何添加值到list: 

Java代码 收藏代码
  1. $ ./redis-cli lpush mylist firstvalue 
  2. OK 
  3. $ ./redis-cli lpush mylist secondvalue 
  4. OK 
  5. $ ./redis-cli lpush mylist thirdvalue 
  6. OK 
  7. $ ./redis-cli lrange mylist 0 -1 
  8. . thirdvalue 
  9. . secondvalue 
  10. . firstvalue 
  11. $ ./redis-cli rpop mylist 
  12. firstvalue 
  13. $ ./redis-cli lrange mylist 0 -1 
  14. . thirdvalue 
  15. . secondvalue 



redis-benchmark.exe：性能测试，用以模拟同时由N个客户端发送M个 SETs/GETs 查询 (类似于 Apache 的 ab 工具).

Java代码 收藏代码
  1. ./redis-benchmark -n 100000 –c 50 
  2.     ====== SET ====== 
  3.     100007 requests completed in 0.88 seconds （译者注：100004 查询完成于 1.14 秒 ） 
  4. 50 parallel clients （译者注：50个并发客户端） 
  5. 3 bytes payload （译者注：3字节有效载荷) 
  6. keep alive: 1 （译者注：保持1个连接) 
  7. 58.50% <= 0 milliseconds（译者注：毫秒） 
  8. 99.17% <= 1 milliseconds 
  9. 99.58% <= 2 milliseconds 
  10. 99.85% <= 3 milliseconds 
  11. 99.90% <= 6 milliseconds 
  12. 100.00% <= 9 milliseconds 
  13. 114293.71 requests per second（译者注：每秒 114293.71 次查询） 


Windows下测试并发客户端极限为60

========================================================================

linux平台Redis安装：

Java代码 收藏代码
  1. wget http://code.google.com/p/redis/downloads/detail?name=redis-2.0.4.tar.gz 
  2. tar xvzf redis-2.0.4.tar.gz 
  3. cd  redis-2.0.4 
  4. make 
  5. mkdir /home/redis 
  6. cp redis-server  /home/redis 
  7. cp redis-benchmark  /home/redis 
  8. cp redis-cli  /home/redis 
  9. cp redis.conf  /home/redis 
  10. cd  /home/redis 



在安装过程中可能需要用到sudo命令，可能新装的redhat虚拟机中新用户还不能使用sudo命令，因此需要手动的修改/etc/sudoers文件，命令如下：

Java代码 收藏代码
  1. cd /etc 
  2. su root ##切换为root用户，同时输入密码 
  3. chmod u+w sudoers ##放开sudoers文件的写权限 
  4. ##在root ALL = (ALL) ALL下面一行增加 "你的用户名" ALL = (ALL) ALL 
  5. :wq ##保存退出 
  6. chmod u-w sudoers ##取消修改权限 



启动
./redis-server redis.conf
进入命令交互模式，两种：
1:   ./redis-cli
2:   telnet 127.0.0.1 6379       (ip接端口)

=============================================================
配置文件参数说明:

1. Redis默认不是以守护进程的方式运行，可以通过该配置项修改，使用yes启用守护进程
   daemonize no
2. 当Redis以守护进程方式运行时，Redis默认会把pid写入/var/run/redis.pid文件，可以通过pidfile指定
   pidfile /var/run/redis.pid
3. 指定Redis监听端口，默认端口为6379，作者在自己的一篇博文中解释了为什么选用6379作为默认端口，因为6379在手机按键上MERZ对应的号码，而MERZ取自意大利歌女Alessia Merz的名字
   port 6379
4. 绑定的主机地址
   bind 127.0.0.1
5.当 客户端闲置多长时间后关闭连接，如果指定为0，表示关闭该功能
   timeout 300
6. 指定日志记录级别，Redis总共支持四个级别：debug、verbose、notice、warning，默认为verbose
   loglevel verbose
7. 日志记录方式，默认为标准输出，如果配置Redis为守护进程方式运行，而这里又配置为日志记录方式为标准输出，则日志将会发送给/dev/null
   logfile stdout
8. 设置数据库的数量，默认数据库为0，可以使用SELECT <dbid>命令在连接上指定数据库id
   databases 16
9. 指定在多长时间内，有多少次更新操作，就将数据同步到数据文件，可以多个条件配合
    save <seconds> <changes>
    Redis默认配置文件中提供了三个条件：
   save 900 1     save 300 10     save 60 10000
    分别表示900秒（15分钟）内有1个更改，300秒（5分钟）内有10个更改以及60秒内有10000个更改。

10. 指定存储至本地数据库时是否压缩数据，默认为yes，Redis采用LZF压缩，如果为了节省CPU时间，可以关闭该选项，但会导致数据库文件变的巨大
   rdbcompression yes
11. 指定本地数据库文件名，默认值为dump.rdb
   dbfilename dump.rdb
12. 指定本地数据库存放目录
   dir ./
13. 设置当本机为slav服务时，设置master服务的IP地址及端口，在Redis启动时，它会自动从master进行数据同步
   slaveof <masterip> <masterport>
14. 当master服务设置了密码保护时，slav服务连接master的密码
   masterauth <master-password>
15. 设置Redis连接密码，如果配置了连接密码，客户端在连接Redis时需要通过AUTH <password>命令提供密码，默认关闭
   requirepass foobared
16. 设置同一时间最大客户端连接数，默认无限制，Redis可以同时打开的客户端连接数为Redis进程可以打开的最大文件描述符数，如果设置 maxclients 0，表示不作限制。当客户端连接数到达限制时，Redis会关闭新的连接并向客户端返回max number of clients reached错误信息
   maxclients 128
17. 指定Redis最大内存限制，Redis在启动时会把数据加载到内存中，达到最大内存后，Redis会先尝试清除已到期或即将到期的Key，当此方法处理 后，仍然到达最大内存设置，将无法再进行写入操作，但仍然可以进行读取操作。Redis新的vm机制，会把Key存放内存，Value会存放在swap区
   maxmemory <bytes>
18. 指定是否在每次更新操作后进行日志记录，Redis在默认情况下是异步的把数据写入磁盘，如果不开启，可能会在断电时导致一段时间内的数据丢失。因为 redis本身同步数据文件是按上面save条件来同步的，所以有的数据会在一段时间内只存在于内存中。默认为no
   appendonly no
19. 指定更新日志文件名，默认为appendonly.aof
    appendfilename appendonly.aof
20. 指定更新日志条件，共有3个可选值：
   no：表示等操作系统进行数据缓存同步到磁盘（快）
   always：表示每次更新操作后手动调用fsync()将数据写到磁盘（慢，安全）
   everysec：表示每秒同步一次（折衷，默认值）
   appendfsync everysec

21. 指定是否启用虚拟内存机制，默认值为no，简单的介绍一下，VM机制将数据分页存放，由Redis将访问量较少的页即冷数据swap到磁盘上，访问多的页面由磁盘自动换出到内存中（在后面的文章我会仔细分析Redis的VM机制）
    vm-enabled no
22. 虚拟内存文件路径，默认值为/tmp/redis.swap，不可多个Redis实例共享
    vm-swap-file /tmp/redis.swap
23. 将所有大于vm-max-memory的数据存入虚拟内存,无论vm-max-memory设置多小,所有索引数据都是内存存储的(Redis的索引数据 就是keys),也就是说,当vm-max-memory设置为0的时候,其实是所有value都存在于磁盘。默认值为0
    vm-max-memory 0
24. Redis swap文件分成了很多的page，一个对象可以保存在多个page上面，但一个page上不能被多个对象共享，vm-page-size是要根据存储的 数据大小来设定的，作者建议如果存储很多小对象，page大小最好设置为32或者64bytes；如果存储很大大对象，则可以使用更大的page，如果不 确定，就使用默认值
    vm-page-size 32
25. 设置swap文件中的page数量，由于页表（一种表示页面空闲或使用的bitmap）是在放在内存中的，，在磁盘上每8个pages将消耗1byte的内存。
    vm-pages 134217728
26. 设置访问swap文件的线程数,最好不要超过机器的核数,如果设置为0,那么所有对swap文件的操作都是串行的，可能会造成比较长时间的延迟。默认值为4
    vm-max-threads 4
27. 设置在向客户端应答时，是否把较小的包合并为一个包发送，默认为开启
   glueoutputbuf yes
28. 指定在超过一定的数量或者最大的元素超过某一临界值时，采用一种特殊的哈希算法
   hash-max-zipmap-entries 64     hash-max-zipmap-value 512
29. 指定是否激活重置哈希，默认为开启（后面在介绍Redis的哈希算法时具体介绍）
   activerehashing yes
30. 指定包含其它的配置文件，可以在同一主机上多个Redis实例之间使用同一份配置文件，而同时各个实例又拥有自己的特定配置文件
   include /path/to/local.conf

posted @ 2014-03-26 13:08  星空刺  阅读(174)  评论(0)  编辑  收藏  举报