Ubuntu16.04安装Redis并配置

Ubuntu16.04安装Redis并配置

2018年05月22日 10:40:35 Hello_刘 阅读数：29146

 

Ubuntu16.04安装Redis并配置
1):安装:

1:终端命令下载redis-4.0.9.tar.gz包

wget http://download.redis.io/releases/redis-4.0.9.tar.gz包

2:解压

tar xzf redis-4.0.9.tar.gz

3:移动，放到usr/local⽬录下

sudo mv ./redis-4.0.9 /usr/local/redis/

4:进⼊redis⽬录

cd /usr/local/redis/

5:生成

sudo make

6:测试,这段运⾏时间会较⻓

sudo make test

7:安装,将redis的命令安装到/usr/local/bin/⽬录

sudo make install

8:安装完成后，进入目录/usr/local/bin中查看

1.  
   cd /usr/local/bin
2.  
   ls -all

• redis-server redis服务器
• redis-cli redis命令行客户端
• redis-benchmark redis性能测试工具
• redis-check-aof AOF文件修复工具
• redis-check-rdb RDB文件检索工具

9:把配置⽂件移动到/etc/redis⽬录下
配置⽂件⽬录为/usr/local/redis/redis.conf

在/etc/目录下创建redis目录,然后移动配置文件

sudo cp /usr/local/redis/redis.conf /etc/redis/

2):配置
1:查看
Redis的配置信息在/etc/redis/redis.conf下

sudo vim /etc/redis/redis.conf

2:核心配置

绑定ip：如果需要远程访问，可将此⾏注释，或绑定⼀个真实ip
bind 127.0.0.1

端⼝，默认为6379
port 6379

是否以守护进程运⾏

• 如果以守护进程运⾏，则不会在命令⾏阻塞，类似于服务
• 如果以⾮守护进程运⾏，则当前终端被阻塞
• 设置为yes表示守护进程，设置为no表示⾮守护进程
• 推荐设置为yes

daemonize yes

数据⽂件
dbfilename dump.rdb

数据⽂件存储路径
dir /var/lib/redis

⽇志⽂件
logfile "/var/log/redis/redis-server.log"

数据库，默认有16个
database 16

主从复制，类似于双机备份。

slaveof

详细配置参数:

1.  
   # Redis configuration file example
2.  
    
3.  
   # Note on units: when memory size is needed, it is possible to specify
4.  
   # it in the usual form of 1k 5GB 4M and so forth:
5.  
   # 内存大小的配置，下面是内存大小配置的转换方式
6.  
   #
7.  
   # 1k => 1000 bytes
8.  
   # 1kb => 1024 bytes
9.  
   # 1m => 1000000 bytes
10.  
    # 1mb => 1024*1024 bytes
11.  
    # 1g => 1000000000 bytes
12.  
    # 1gb => 1024*1024*1024 bytes
13.  
    #
14.  
    # units are case insensitive so 1GB 1Gb 1gB are all the same.
15.  
    # 内存大小的配置，不区分大小写
16.  
     
17.  
    ################################## INCLUDES ###################################
18.  
     
19.  
    # Include one or more other config files here. This is useful if you
20.  
    # have a standard template that goes to all Redis server but also need
21.  
    # to customize a few per-server settings. Include files can include
22.  
    # other files, so use this wisely.
23.  
    #
24.  
    # Notice option "include" won't be rewritten by command "CONFIG REWRITE"
25.  
    # from admin or Redis Sentinel. Since Redis always uses the last processed
26.  
    # line as value of a configuration directive, you'd better put includes
27.  
    # at the beginning of this file to avoid overwriting config change at runtime.
28.  
    #
29.  
    # If instead you are interested in using includes to override configuration
30.  
    # options, it is better to use include as the last line.
31.  
    #
32.  
    # include /path/to/local.conf
33.  
    # include /path/to/other.conf
34.  
    # 当配置多个redis时，可能大部分配置一样，而对于不同的redis，只有少部分配置需要定制
35.  
    # 就可以配置一个公共的模板配置。
36.  
    # 对于具体的reids，只需设置少量的配置，并用include把模板配置包含进来即可。
37.  
    #
38.  
    # 值得注意的是，对于同一个配置项，redis只对最后一行的有效
39.  
    # 所以为避免模板配置覆盖当前配置，应在配置文件第一行使用include
40.  
    # 当然，如果模板配置的优先级比较高，就在配置文件最后一行使用include
41.  
     
42.  
    ################################ GENERAL #####################################
43.  
     
44.  
    # By default Redis does not run as a daemon. Use 'yes' if you need it.
45.  
    # Note that Redis will write a pid file in /var/run/redis.pid when daemonized.
46.  
    # yes为使用守护进程，此时redis的进程ID会被写进 pidfile的配置中
47.  
    daemonize yes
48.  
     
49.  
    # When running daemonized, Redis writes a pid file in /var/run/redis.pid by
50.  
    # default. You can specify a custom pid file location here.
51.  
    # 当redis以守护进程的方式启动时，redis的进程ID将会写在这个文件中
52.  
    pidfile /var/run/redis.pid
53.  
     
54.  
    # Accept connections on the specified port, default is 6379.
55.  
    # If port 0 is specified Redis will not listen on a TCP socket.
56.  
    # redis 启动的端口。【应该知道redis是服务端吧】
57.  
    port 6379
58.  
     
59.  
    # TCP listen() backlog.
60.  
    #
61.  
    # In high requests-per-second environments you need an high backlog in order
62.  
    # to avoid slow clients connections issues. Note that the Linux kernel
63.  
    # will silently truncate it to the value of /proc/sys/net/core/somaxconn so
64.  
    # make sure to raise both the value of somaxconn and tcp_max_syn_backlog
65.  
    # in order to get the desired effect.
66.  
    # 最大链接缓冲池的大小，这里应该是指的未完成链接请求的数量
67.  
    #（测试值为1时，仍可以有多个链接）
68.  
    # 但该值与listen函数中的backlog意义应该是相同的，源码中该值就是被用在了listen函数中
69.  
    # 该值同时受/proc/sys/net/core/somaxconn 和 tcp_max_syn_backlog（/etc/sysctl.conf中配置）的限制
70.  
    # tcp_max_syn_backlog 指的是未完成链接的数量
71.  
    tcp-backlog 511
72.  
     
73.  
    # By default Redis listens for connections from all the network interfaces
74.  
    # available on the server. It is possible to listen to just one or multiple
75.  
    # interfaces using the "bind" configuration directive, followed by one or
76.  
    # more IP addresses.
77.  
    # 绑定ip，指定ip可以连接到redis
78.  
    #
79.  
    # Examples:
80.  
    #
81.  
    # bind 192.168.1.100 10.0.0.1
82.  
    # bind 127.0.0.1
83.  
     
84.  
    # Specify the path for the Unix socket that will be used to listen for
85.  
    # incoming connections. There is no default, so Redis will not listen
86.  
    # on a unix socket when not specified.
87.  
    #
88.  
    # 这个应该就是以文件形式创建的socket
89.  
    # unixsocket /tmp/redis.sock
90.  
    # unixsocketperm 755
91.  
     
92.  
    # Close the connection after a client is idle for N seconds (0 to disable)
93.  
    # 超时断链机制，如果一个链接在N秒内没有任何操作，则断开该链接
94.  
    # N为0时，该机制失效
95.  
    timeout 0
96.  
     
97.  
    # TCP keepalive.
98.  
    #
99.  
    # If non-zero, use SO_KEEPALIVE to send TCP ACKs to clients in absence
100.  
     # of communication. This is useful for two reasons:
101.  
     #
102.  
     # 1) Detect dead peers.
103.  
     # 2) Take the connection alive from the point of view of network
104.  
     # equipment in the middle.
105.  
     #
106.  
     # On Linux, the specified value (in seconds) is the period used to send ACKs.
107.  
     # Note that to close the connection the double of the time is needed.
108.  
     # On other kernels the period depends on the kernel configuration.
109.  
     # 就像心跳检测一样，检查链接是否保持正常，同时也可以保持正常链接的通信
110.  
     # 建议值为60
111.  
     #
112.  
     # A reasonable value for this option is 60 seconds.
113.  
     tcp-keepalive 0
114.  
      
115.  
     # Specify the server verbosity level.
116.  
     # This can be one of:
117.  
     # debug (a lot of information, useful for development/testing)
118.  
     # verbose (many rarely useful info, but not a mess like the debug level)
119.  
     # notice (moderately verbose, what you want in production probably)
120.  
     # warning (only very important / critical messages are logged)
121.  
     # 日志级别
122.  
     loglevel notice
123.  
      
124.  
     # Specify the log file name. Also the empty string can be used to force
125.  
     # Redis to log on the standard output. Note that if you use standard
126.  
     # output for logging but daemonize, logs will be sent to /dev/null
127.  
     # 日志存放路径，默认是输出到标准输出，但当以守护进程方式启动时，默认输出到/dev/null(传说中的linux黑洞)
128.  
     logfile ""
129.  
      
130.  
     # To enable logging to the system logger, just set 'syslog-enabled' to yes,
131.  
     # and optionally update the other syslog parameters to suit your needs.
132.  
     # yes 表示将日志写到系统日志中
133.  
     # syslog-enabled no
134.  
      
135.  
     # Specify the syslog identity.
136.  
     # 当syslog-enabled为yes时，指定系统日志的标示为 redis
137.  
     # syslog-ident redis
138.  
      
139.  
     # Specify the syslog facility. Must be USER or between LOCAL0-LOCAL7.
140.  
     # 指定系统日志的设备
141.  
     # syslog-facility local0
142.  
      
143.  
     # Set the number of databases. The default database is DB 0, you can select
144.  
     # a different one on a per-connection basis using SELECT <dbid> where
145.  
     # dbid is a number between 0 and 'databases'-1
146.  
     # redis的数据库格式，默认16个(0~15)，默认使用第0个。
147.  
     databases 16
148.  
      
149.  
     ################################ SNAPSHOTTING ################################
150.  
     #
151.  
     # Save the DB on disk:
152.  
     #
153.  
     # save <seconds> <changes>
154.  
     #
155.  
     # Will save the DB if both the given number of seconds and the given
156.  
     # number of write operations against the DB occurred.
157.  
     # 快照，即将数据写到硬盘上，在<seconds>秒内，至少有<changes>次写入数据库操作
158.  
     # 则会将数据写入硬盘一次。
159.  
     # 将save行注释掉则永远不会写入硬盘
160.  
     # save "" 表示删除所有的快照点
161.  
     #
162.  
     # In the example below the behaviour will be to save:
163.  
     # after 900 sec (15 min) if at least 1 key changed
164.  
     # after 300 sec (5 min) if at least 10 keys changed
165.  
     # after 60 sec if at least 10000 keys changed
166.  
     #
167.  
     # Note: you can disable saving at all commenting all the "save" lines.
168.  
     #
169.  
     # It is also possible to remove all the previously configured save
170.  
     # points by adding a save directive with a single empty string argument
171.  
     # like in the following example:
172.  
     #
173.  
     # save ""
174.  
      
175.  
     save 900 1
176.  
     save 300 10
177.  
     save 60 10000
178.  
      
179.  
     # By default Redis will stop accepting writes if RDB snapshots are enabled
180.  
     # (at least one save point) and the latest background save failed.
181.  
     # This will make the user aware (in a hard way) that data is not persisting
182.  
     # on disk properly, otherwise chances are that no one will notice and some
183.  
     # disaster will happen.
184.  
     #
185.  
     # If the background saving process will start working again Redis will
186.  
     # automatically allow writes again.
187.  
     #
188.  
     # However if you have setup your proper monitoring of the Redis server
189.  
     # and persistence, you may want to disable this feature so that Redis will
190.  
     # continue to work as usual even if there are problems with disk,
191.  
     # permissions, and so forth.
192.  
     # 当做快照失败的时候，redis会停止继续向其写入数据，保证第一时间发现redis快照出现问题
193.  
     # 当然，通过下面配置为 no，即使redis快照失败，也能继续向redis写入数据
194.  
     stop-writes-on-bgsave-error yes
195.  
      
196.  
     # Compress string objects using LZF when dump .rdb databases?
197.  
     # For default that's set to 'yes' as it's almost always a win.
198.  
     # If you want to save some CPU in the saving child set it to 'no' but
199.  
     # the dataset will likely be bigger if you have compressible values or keys.
200.  
     # 快照的时候，是否用LZF压缩，使用压缩会占一定的cpu，但不使用压缩，快照会很大
201.  
     rdbcompression yes
202.  
      
203.  
     # Since version 5 of RDB a CRC64 checksum is placed at the end of the file.
204.  
     # This makes the format more resistant to corruption but there is a performance
205.  
     # hit to pay (around 10%) when saving and loading RDB files, so you can disable it
206.  
     # for maximum performances.
207.  
     #
208.  
     # RDB files created with checksum disabled have a checksum of zero that will
209.  
     # tell the loading code to skip the check.
210.  
     # 数据校验，快照末尾会存放一个校验值，保证数据的准确性
211.  
     # 但数据校验会使性能下降约10%，默认开启校验
212.  
     rdbchecksum yes
213.  
      
214.  
     # The filename where to dump the DB
215.  
     # 快照的名字
216.  
     dbfilename dump.rdb
217.  
      
218.  
     # The working directory.
219.  
     #
220.  
     # The DB will be written inside this directory, with the filename specified
221.  
     # above using the 'dbfilename' configuration directive.
222.  
     #
223.  
     # The Append Only File will also be created inside this directory.
224.  
     #
225.  
     # Note that you must specify a directory here, not a file name.
226.  
     #
227.  
     # 快照存放的目录
228.  
     # linux root下测试，会发现该进程会在当前目录下创建一个dump.rdb
229.  
     # 但快照却放在了根目录/下，重启的时候，是不会从快照中恢复数据的
230.  
     # 当把根目录下的dump.rdb文件拷贝到当前目录的时候，再次启动，就会从快照中恢复数据
231.  
     # 而且以后的快照也都在当前目录的dump.rdb中做操作
232.  
     #
233.  
     # 值得一提的是，快照是异步方式的，如果在还未达到快照的时候，修改了数据，而且redis发生问题crash了
234.  
     # 那么中间的修改数据是不会被保存到dump.rdb快照中的
235.  
     # 解决办法就是用Append Only Mode的同步模式(下面将会有该配置项)
236.  
     # 将会把每个操作写到Append Only File中，该文件也存放于当前配置的目录
237.  
     # 建议使用绝对路径！！！
238.  
     #
239.  
     dir ./
240.  
      
241.  
     ################################# REPLICATION #################################
242.  
      
243.  
     # Master-Slave replication. Use slaveof to make a Redis instance a copy of
244.  
     # another Redis server. Note that the configuration is local to the slave
245.  
     # so for example it is possible to configure the slave to save the DB with a
246.  
     # different interval, or to listen to another port, and so on.
247.  
     #
248.  
     # 主从复制，类似于双机备份。
249.  
     # 配置需指定主机的ip 和port
250.  
     # slaveof <masterip> <masterport>
251.  
      
252.  
     # If the master is password protected (using the "requirepass" configuration
253.  
     # directive below) it is possible to tell the slave to authenticate before
254.  
     # starting the replication synchronization process, otherwise the master will
255.  
     # refuse the slave request.
256.  
     #
257.  
     # 如果主机redis需要密码，则指定密码
258.  
     # 密码配置在下面安全配置中
259.  
     # masterauth <master-password>
260.  
      
261.  
     # When a slave loses its connection with the master, or when the replication
262.  
     # is still in progress, the slave can act in two different ways:
263.  
     #
264.  
     # 1) if slave-serve-stale-data is set to 'yes' (the default) the slave will
265.  
     # still reply to client requests, possibly with out of date data, or the
266.  
     # data set may just be empty if this is the first synchronization.
267.  
     #
268.  
     # 2) if slave-serve-stale-data is set to 'no' the slave will reply with
269.  
     # an error "SYNC with master in progress" to all the kind of commands
270.  
     # but to INFO and SLAVEOF.
271.  
     #
272.  
     # 当从机与主机断开时，即同步出现问题的时候，从机有两种处理方式
273.  
     # yes, 继续响应客户端请求，但可能有脏数据（过期数据、空数据等）
274.  
     # no，对客户端的请求统一回复为“SYNC with master in progress”，除了INFO和SLAVEOF命令
275.  
     slave-serve-stale-data yes
276.  
      
277.  
     # You can configure a slave instance to accept writes or not. Writing against
278.  
     # a slave instance may be useful to store some ephemeral data (because data
279.  
     # written on a slave will be easily deleted after resync with the master) but
280.  
     # may also cause problems if clients are writing to it because of a
281.  
     # misconfiguration.
282.  
     #
283.  
     # Since Redis 2.6 by default slaves are read-only.
284.  
     #
285.  
     # Note: read only slaves are not designed to be exposed to untrusted clients
286.  
     # on the internet. It's just a protection layer against misuse of the instance.
287.  
     # Still a read only slave exports by default all the administrative commands
288.  
     # such as CONFIG, DEBUG, and so forth. To a limited extent you can improve
289.  
     # security of read only slaves using 'rename-command' to shadow all the
290.  
     # administrative / dangerous commands.
291.  
     # slave只读选项，设置从机只读（默认）。
292.  
     # 即使设置可写，当下一次从主机上同步数据，仍然会删除当前从机上写入的数据
293.  
     # 【待测试】：主机与从机互为slave会出现什么情况？
294.  
     # 【预期三种结果】：1. 提示报错 2. 主从服务器数据不可控 3. 一切正常
295.  
     slave-read-only yes
296.  
      
297.  
     # Slaves send PINGs to server in a predefined interval. It's possible to change
298.  
     # this interval with the repl_ping_slave_period option. The default value is 10
299.  
     # seconds.
300.  
     #
301.  
     # 从服务器向主服务器发送心跳包，默认10发送一次
302.  
     # repl-ping-slave-period 10
303.  
      
304.  
     # The following option sets the replication timeout for:
305.  
     #
306.  
     # 1) Bulk transfer I/O during SYNC, from the point of view of slave.
307.  
     # 2) Master timeout from the point of view of slaves (data, pings).
308.  
     # 3) Slave timeout from the point of view of masters (REPLCONF ACK pings).
309.  
     #
310.  
     # It is important to make sure that this value is greater than the value
311.  
     # specified for repl-ping-slave-period otherwise a timeout will be detected
312.  
     # every time there is low traffic between the master and the slave.
313.  
     #
314.  
     # 超时响应时间，值必须比repl-ping-slave-period大
315.  
     # 批量数据传输超时、ping超时
316.  
     # repl-timeout 60
317.  
      
318.  
     # Disable TCP_NODELAY on the slave socket after SYNC?
319.  
     #
320.  
     # If you select "yes" Redis will use a smaller number of TCP packets and
321.  
     # less bandwidth to send data to slaves. But this can add a delay for
322.  
     # the data to appear on the slave side, up to 40 milliseconds with
323.  
     # Linux kernels using a default configuration.
324.  
     #
325.  
     # If you select "no" the delay for data to appear on the slave side will
326.  
     # be reduced but more bandwidth will be used for replication.
327.  
     #
328.  
     # By default we optimize for low latency, but in very high traffic conditions
329.  
     # or when the master and slaves are many hops away, turning this to "yes" may
330.  
     # be a good idea.
331.  
     # 主从同步是否延迟
332.  
     # yes 有延迟，约40毫秒（linux kernel的默认配置），使用较少的数据包，较小的带宽
333.  
     # no 无延迟（减少延迟），但需要更大的带宽
334.  
     repl-disable-tcp-nodelay no
335.  
      
336.  
     # Set the replication backlog size. The backlog is a buffer that accumulates
337.  
     # slave data when slaves are disconnected for some time, so that when a slave
338.  
     # wants to reconnect again, often a full resync is not needed, but a partial
339.  
     # resync is enough, just passing the portion of data the slave missed while
340.  
     # disconnected.
341.  
     #
342.  
     # The biggest the replication backlog, the longer the time the slave can be
343.  
     # disconnected and later be able to perform a partial resynchronization.
344.  
     #
345.  
     # The backlog is only allocated once there is at least a slave connected.
346.  
     #
347.  
     # 默认情况下，当slave重连的时候，会进行全量数据同步
348.  
     # 但实际上slave只需要部分同步即可，这个选项设置部分同步的大小
349.  
     # 设置值越大，同步的时间就越长
350.  
     # repl-backlog-size 1mb
351.  
      
352.  
     # After a master has no longer connected slaves for some time, the backlog
353.  
     # will be freed. The following option configures the amount of seconds that
354.  
     # need to elapse, starting from the time the last slave disconnected, for
355.  
     # the backlog buffer to be freed.
356.  
     #
357.  
     # A value of 0 means to never release the backlog.
358.  
     #
359.  
     # 主机的后台日志释放时间，即当没有slave连接时，过多久释放后台日志
360.  
     # 0表示不释放
361.  
     # repl-backlog-ttl 3600
362.  
      
363.  
     # The slave priority is an integer number published by Redis in the INFO output.
364.  
     # It is used by Redis Sentinel in order to select a slave to promote into a
365.  
     # master if the master is no longer working correctly.
366.  
     #
367.  
     # A slave with a low priority number is considered better for promotion, so
368.  
     # for instance if there are three slaves with priority 10, 100, 25 Sentinel will
369.  
     # pick the one with priority 10, that is the lowest.
370.  
     #
371.  
     # However a special priority of 0 marks the slave as not able to perform the
372.  
     # role of master, so a slave with priority of 0 will never be selected by
373.  
     # Redis Sentinel for promotion.
374.  
     #
375.  
     # By default the priority is 100.
376.  
     # 当主机crash的时候，在从机中选择一台作为主机，数字越小，优先级越高
377.  
     # 0 表示永远不作为主机，默认值是100
378.  
     slave-priority 100
379.  
      
380.  
     # It is possible for a master to stop accepting writes if there are less than
381.  
     # N slaves connected, having a lag less or equal than M seconds.
382.  
     #
383.  
     # The N slaves need to be in "online" state.
384.  
     #
385.  
     # The lag in seconds, that must be <= the specified value, is calculated from
386.  
     # the last ping received from the slave, that is usually sent every second.
387.  
     #
388.  
     # This option does not GUARANTEES that N replicas will accept the write, but
389.  
     # will limit the window of exposure for lost writes in case not enough slaves
390.  
     # are available, to the specified number of seconds.
391.  
     #
392.  
     # For example to require at least 3 slaves with a lag <= 10 seconds use:
393.  
     #
394.  
     # 当slave数量小于min-slaves-to-write，且延迟小于等于min-slaves-max-lag时，
395.  
     # 主机停止写入操作
396.  
     # 0表示禁用
397.  
     # 默认min-slaves-to-write为0，即禁用。min-slaves-max-lag为10
398.  
     # min-slaves-to-write 3
399.  
     # min-slaves-max-lag 10
400.  
     #
401.  
     # Setting one or the other to 0 disables the feature.
402.  
     #
403.  
     # By default min-slaves-to-write is set to 0 (feature disabled) and
404.  
     # min-slaves-max-lag is set to 10.
405.  
      
406.  
     ################################## SECURITY ###################################
407.  
      
408.  
     # Require clients to issue AUTH <PASSWORD> before processing any other
409.  
     # commands. This might be useful in environments in which you do not trust
410.  
     # others with access to the host running redis-server.
411.  
     #
412.  
     # This should stay commented out for backward compatibility and because most
413.  
     # people do not need auth (e.g. they run their own servers).
414.  
     #
415.  
     # Warning: since Redis is pretty fast an outside user can try up to
416.  
     # 150k passwords per second against a good box. This means that you should
417.  
     # use a very strong password otherwise it will be very easy to break.
418.  
     #
419.  
     # redis密码，默认不配置，即无密码
420.  
     # 这里注意，如果设置了密码，应该设置一个复杂度比较高的密码
421.  
     # 因为redis的速度很快，每秒可以尝试150k次的密码测试，很容易对其进行暴力破解（跑码）。
422.  
     # 疑问：这里为什么不设置一个针对主机的测试次数限制的，例如每10次，则禁止建立连接1个小时！
423.  
     # requirepass foobared
424.  
      
425.  
     # Command renaming.
426.  
     #
427.  
     # It is possible to change the name of dangerous commands in a shared
428.  
     # environment. For instance the CONFIG command may be renamed into something
429.  
     # hard to guess so that it will still be available for internal-use tools
430.  
     # but not available for general clients.
431.  
     #
432.  
     # 命令重命名，将命令重命名为另一个字符串标识
433.  
     # 如果命令为空串("")，则会彻底禁用该命令
434.  
     # 命令重命名，会对写AOF(Append of file)文件、slave从机造成一些问题
435.  
     # Example:
436.  
     #
437.  
     # rename-command CONFIG b840fc02d524045429941cc15f59e41cb7be6c52
438.  
     #
439.  
     # It is also possible to completely kill a command by renaming it into
440.  
     # an empty string:
441.  
     #
442.  
     # rename-command CONFIG ""
443.  
     #
444.  
     # Please note that changing the name of commands that are logged into the
445.  
     # AOF file or transmitted to slaves may cause problems.
446.  
      
447.  
     ################################### LIMITS ####################################
448.  
      
449.  
     # Set the max number of connected clients at the same time. By default
450.  
     # this limit is set to 10000 clients, however if the Redis server is not
451.  
     # able to configure the process file limit to allow for the specified limit
452.  
     # the max number of allowed clients is set to the current file limit
453.  
     # minus 32 (as Redis reserves a few file descriptors for internal uses).
454.  
     #
455.  
     # Once the limit is reached Redis will close all the new connections sending
456.  
     # an error 'max number of clients reached'.
457.  
     #
458.  
     # 这只redis的最大连接数目，默认设置为10000个客户端
459.  
     # 当超过限制时，将段开新的连接，并响应“max number of clients reached”
460.  
     # maxclients 10000
461.  
      
462.  
     # Don't use more memory than the specified amount of bytes.
463.  
     # When the memory limit is reached Redis will try to remove keys
464.  
     # according to the eviction policy selected (see maxmemory-policy).
465.  
     #
466.  
     # If Redis can't remove keys according to the policy, or if the policy is
467.  
     # set to 'noeviction', Redis will start to reply with errors to commands
468.  
     # that would use more memory, like SET, LPUSH, and so on, and will continue
469.  
     # to reply to read-only commands like GET.
470.  
     #
471.  
     # This option is usually useful when using Redis as an LRU cache, or to set
472.  
     # a hard memory limit for an instance (using the 'noeviction' policy).
473.  
     #
474.  
     # WARNING: If you have slaves attached to an instance with maxmemory on,
475.  
     # the size of the output buffers needed to feed the slaves are subtracted
476.  
     # from the used memory count, so that network problems / resyncs will
477.  
     # not trigger a loop where keys are evicted, and in turn the output
478.  
     # buffer of slaves is full with DELs of keys evicted triggering the deletion
479.  
     # of more keys, and so forth until the database is completely emptied.
480.  
     #
481.  
     # In short... if you have slaves attached it is suggested that you set a lower
482.  
     # limit for maxmemory so that there is some free RAM on the system for slave
483.  
     # output buffers (but this is not needed if the policy is 'noeviction').
484.  
     #
485.  
     # redis的最大内存限制，如果达到最大内存，会按照下面的maxmemory-policy进行清除
486.  
     # 如果不能再清除或者maxmemory-policy为noeviction，则对于需要增加空间的操作，将会返回错误
487.  
     maxmemory <1024*1024*1024>
488.  
      
489.  
     # MAXMEMORY POLICY: how Redis will select what to remove when maxmemory
490.  
     # is reached. You can select among five behaviors:
491.  
     #
492.  
     # volatile-lru -> remove the key with an expire set using an LRU algorithm
493.  
     # allkeys-lru -> remove any key accordingly to the LRU algorithm
494.  
     # volatile-random -> remove a random key with an expire set
495.  
     # allkeys-random -> remove a random key, any key
496.  
     # volatile-ttl -> remove the key with the nearest expire time (minor TTL)
497.  
     # noeviction -> don't expire at all, just return an error on write operations
498.  
     #
499.  
     # Note: with any of the above policies, Redis will return an error on write
500.  
     # operations, when there are not suitable keys for eviction.
501.  
     #
502.  
     # At the date of writing this commands are: set setnx setex append
503.  
     # incr decr rpush lpush rpushx lpushx linsert lset rpoplpush sadd
504.  
     # sinter sinterstore sunion sunionstore sdiff sdiffstore zadd zincrby
505.  
     # zunionstore zinterstore hset hsetnx hmset hincrby incrby decrby
506.  
     # getset mset msetnx exec sort
507.  
     #
508.  
     # The default is:
509.  
     #
510.  
     # 内存删除策略，默认volatile-lru，利用LRU算法，删除过期的key
511.  
     maxmemory-policy volatile-lru
512.  
      
513.  
     # LRU and minimal TTL algorithms are not precise algorithms but approximated
514.  
     # algorithms (in order to save memory), so you can select as well the sample
515.  
     # size to check. For instance for default Redis will check three keys and
516.  
     # pick the one that was used less recently, you can change the sample size
517.  
     # using the following configuration directive.
518.  
     #
519.  
     # LRU算法与最小TTL算法只是相对精确的算法，并不是绝对精确的算法
520.  
     # 为了更精确，可以设置样本个数
521.  
     # 比如设置3个样本，redis会选取三个key，并选择删除那个上次使用时间最远的
522.  
     # maxmemory-samples 3
523.  
      
524.  
     ############################## APPEND ONLY MODE ###############################
525.  
      
526.  
     # By default Redis asynchronously dumps the dataset on disk. This mode is
527.  
     # good enough in many applications, but an issue with the Redis process or
528.  
     # a power outage may result into a few minutes of writes lost (depending on
529.  
     # the configured save points).
530.  
     #
531.  
     # The Append Only File is an alternative persistence mode that provides
532.  
     # much better durability. For instance using the default data fsync policy
533.  
     # (see later in the config file) Redis can lose just one second of writes in a
534.  
     # dramatic event like a server power outage, or a single write if something
535.  
     # wrong with the Redis process itself happens, but the operating system is
536.  
     # still running correctly.
537.  
     #
538.  
     # AOF and RDB persistence can be enabled at the same time without problems.
539.  
     # If the AOF is enabled on startup Redis will load the AOF, that is the file
540.  
     # with the better durability guarantees.
541.  
     #
542.  
     # Please check http://redis.io/topics/persistence for more information.
543.  
     # 将对redis所有的操作都保存到AOF文件中
544.  
     # 因为dump.rdb是异步的，在下次快照到达之前，如果出现crash等问题，会造成数据丢失
545.  
     # 而AOF文件时同步记录的，所以会完整的恢复数据
546.  
      
547.  
     appendonly no
548.  
      
549.  
     # The name of the append only file (default: "appendonly.aof")
550.  
     # AOF文件的名字
551.  
      
552.  
     appendfilename "appendonly.aof"
553.  
      
554.  
     # The fsync() call tells the Operating System to actually write data on disk
555.  
     # instead to wait for more data in the output buffer. Some OS will really flush
556.  
     # data on disk, some other OS will just try to do it ASAP.
557.  
     #
558.  
     # Redis supports three different modes:
559.  
     #
560.  
     # no: don't fsync, just let the OS flush the data when it wants. Faster.
561.  
     # always: fsync after every write to the append only log . Slow, Safest.
562.  
     # everysec: fsync only one time every second. Compromise.
563.  
     #
564.  
     # The default is "everysec", as that's usually the right compromise between
565.  
     # speed and data safety. It's up to you to understand if you can relax this to
566.  
     # "no" that will let the operating system flush the output buffer when
567.  
     # it wants, for better performances (but if you can live with the idea of
568.  
     # some data loss consider the default persistence mode that's snapshotting),
569.  
     # or on the contrary, use "always" that's very slow but a bit safer than
570.  
     # everysec.
571.  
     #
572.  
     # More details please check the following article:
573.  
     # http://antirez.com/post/redis-persistence-demystified.html
574.  
     #
575.  
     # If unsure, use "everysec".
576.  
     # redis的数据同步方式，三种
577.  
     # no，redis本身不做同步，由OS来做。redis的速度会很快
578.  
     # always，在每次写操作之后，redis都进行同步，即写入AOF文件。redis会变慢，但是数据更安全
579.  
     # everysec，折衷考虑，每秒同步一次数据。【默认】
580.  
      
581.  
     # appendfsync always
582.  
     appendfsync everysec
583.  
     # appendfsync no
584.  
      
585.  
     # When the AOF fsync policy is set to always or everysec, and a background
586.  
     # saving process (a background save or AOF log background rewriting) is
587.  
     # performing a lot of I/O against the disk, in some Linux configurations
588.  
     # Redis may block too long on the fsync() call. Note that there is no fix for
589.  
     # this currently, as even performing fsync in a different thread will block
590.  
     # our synchronous write(2) call.
591.  
     #
592.  
     # In order to mitigate this problem it's possible to use the following option
593.  
     # that will prevent fsync() from being called in the main process while a
594.  
     # BGSAVE or BGREWRITEAOF is in progress.
595.  
     #
596.  
     # This means that while another child is saving, the durability of Redis is
597.  
     # the same as "appendfsync none". In practical terms, this means that it is
598.  
     # possible to lose up to 30 seconds of log in the worst scenario (with the
599.  
     # default Linux settings).
600.  
     #
601.  
     # If you have latency problems turn this to "yes". Otherwise leave it as
602.  
     # "no" that is the safest pick from the point of view of durability.
603.  
     # redis的同步方式中，always和everysec，快照和写AOF可能会执行大量的硬盘I/O操作，
604.  
     # 而在一些Linux的配置中，redis会阻塞很久，而redis本身并没有很好的解决这一问题。
605.  
     # 为了缓和这一问题，redis提供no-appendfsync-on-rewrite选项，
606.  
     # 即当有另外一个进程在执行保存操作的时候，redis采用no的同步方式。
607.  
     # 最坏情况下会有延迟30秒的同步延迟。
608.  
     # 如果你觉得这样做会有潜在危险，则请将该选项改为yes。否则就保持默认值no（基于稳定性考虑）。
609.  
      
610.  
     no-appendfsync-on-rewrite no
611.  
      
612.  
     # Automatic rewrite of the append only file.
613.  
     # Redis is able to automatically rewrite the log file implicitly calling
614.  
     # BGREWRITEAOF when the AOF log size grows by the specified percentage.
615.  
     #
616.  
     # This is how it works: Redis remembers the size of the AOF file after the
617.  
     # latest rewrite (if no rewrite has happened since the restart, the size of
618.  
     # the AOF at startup is used).
619.  
     #
620.  
     # This base size is compared to the current size. If the current size is
621.  
     # bigger than the specified percentage, the rewrite is triggered. Also
622.  
     # you need to specify a minimal size for the AOF file to be rewritten, this
623.  
     # is useful to avoid rewriting the AOF file even if the percentage increase
624.  
     # is reached but it is still pretty small.
625.  
     #
626.  
     # Specify a percentage of zero in order to disable the automatic AOF
627.  
     # rewrite feature.
628.  
     # 自动重写AOF文件
629.  
     # 当AOF日志文件大小增长到指定百分比时，redis会自动隐式调用BGREWRITEAOF来重写AOF文件
630.  
     # redis会记录上次重写AOF文件之后的大小，
631.  
     # 如果当前文件大小增加了auto-aof-rewrite-percentage，则会触发重写AOF日志功能
632.  
     # 当然如果文件过小，比如小于auto-aof-rewrite-min-size这个大小，是不会触发重写AOF日志功能的
633.  
     # auto-aof-rewrite-percentage为0时，禁用重写功能
634.  
      
635.  
     auto-aof-rewrite-percentage 100
636.  
     auto-aof-rewrite-min-size 64mb
637.  
      
638.  
     ################################ LUA SCRIPTING ###############################
639.  
      
640.  
     # Max execution time of a Lua script in milliseconds.
641.  
     #
642.  
     # If the maximum execution time is reached Redis will log that a script is
643.  
     # still in execution after the maximum allowed time and will start to
644.  
     # reply to queries with an error.
645.  
     #
646.  
     # When a long running script exceed the maximum execution time only the
647.  
     # SCRIPT KILL and SHUTDOWN NOSAVE commands are available. The first can be
648.  
     # used to stop a script that did not yet called write commands. The second
649.  
     # is the only way to shut down the server in the case a write commands was
650.  
     # already issue by the script but the user don't want to wait for the natural
651.  
     # termination of the script.
652.  
     #
653.  
     # Set it to 0 or a negative value for unlimited execution without warnings.
654.  
     # LUA脚本的最大执行时间（单位是毫秒），默认5000毫秒，即5秒
655.  
     # 如果LUA脚本执行超过这个限制，可以调用SCRIPT KILL和SHUTDOWN NOSAVE命令。
656.  
     # SCRIPT KILL可以终止脚本执行
657.  
     # SHUTDOWN NOSAVE关闭服务，防止LUA脚本的写操作发生
658.  
     # 该值为0或者负数，表示没有限制时间
659.  
     lua-time-limit 5000
660.  
      
661.  
     ################################## SLOW LOG ###################################
662.  
      
663.  
     # The Redis Slow Log is a system to log queries that exceeded a specified
664.  
     # execution time. The execution time does not include the I/O operations
665.  
     # like talking with the client, sending the reply and so forth,
666.  
     # but just the time needed to actually execute the command (this is the only
667.  
     # stage of command execution where the thread is blocked and can not serve
668.  
     # other requests in the meantime).
669.  
     #
670.  
     # You can configure the slow log with two parameters: one tells Redis
671.  
     # what is the execution time, in microseconds, to exceed in order for the
672.  
     # command to get logged, and the other parameter is the length of the
673.  
     # slow log. When a new command is logged the oldest one is removed from the
674.  
     # queue of logged commands.
675.  
     # 记录执行比较慢的命令
676.  
     # 执行比较慢仅仅是指命令的执行时间，不包括客户端的链接与响应等时间
677.  
     # slowlog-log-slower-than 设定这个慢的时间，单位是微妙，即1000000表示1秒，0表示所有命令都记录，负数表示不记录
678.  
     # slowlog-max-len表示记录的慢命令的个数，超过限制，则最早记录的命令会被移除
679.  
     # 命令的长度没有限制，但是会消耗内存，用SLOWLOG RESET来收回这些消耗的内存
680.  
      
681.  
     # The following time is expressed in microseconds, so 1000000 is equivalent
682.  
     # to one second. Note that a negative number disables the slow log, while
683.  
     # a value of zero forces the logging of every command.
684.  
     slowlog-log-slower-than 10000
685.  
      
686.  
     # There is no limit to this length. Just be aware that it will consume memory.
687.  
     # You can reclaim memory used by the slow log with SLOWLOG RESET.
688.  
     slowlog-max-len 128
689.  
      
690.  
     ################################ LATENCY MONITOR ##############################
691.  
      
692.  
     # The Redis latency monitoring subsystem samples different operations
693.  
     # at runtime in order to collect data related to possible sources of
694.  
     # latency of a Redis instance.
695.  
     #
696.  
     # Via the LATENCY command this information is available to the user that can
697.  
     # print graphs and obtain reports.
698.  
     #
699.  
     # The system only logs operations that were performed in a time equal or
700.  
     # greater than the amount of milliseconds specified via the
701.  
     # latency-monitor-threshold configuration directive. When its value is set
702.  
     # to zero, the latency monitor is turned off.
703.  
     #
704.  
     # By default latency monitoring is disabled since it is mostly not needed
705.  
     # if you don't have latency issues, and collecting data has a performance
706.  
     # impact, that while very small, can be measured under big load. Latency
707.  
     # monitoring can easily be enalbed at runtime using the command
708.  
     # "CONFIG SET latency-monitor-threshold <milliseconds>" if needed.
709.  
     # 延迟监控器
710.  
     # redis延迟监控子系统在运行时，会抽样检测可能导致延迟的不同操作
711.  
     # 通过LATENCY命令可以打印相关信息和报告, 命令如下（摘自源文件注释）：
712.  
     # LATENCY SAMPLES: return time-latency samples for the specified event.
713.  
     # LATENCY LATEST: return the latest latency for all the events classes.
714.  
     # LATENCY DOCTOR: returns an human readable analysis of instance latency.
715.  
     # LATENCY GRAPH: provide an ASCII graph of the latency of the specified event.
716.  
     #
717.  
     # 系统只记录超过设定值的操作，单位是毫秒，0表示禁用该功能
718.  
     # 可以通过命令“CONFIG SET latency-monitor-threshold <milliseconds>” 直接设置而不需要重启redis
719.  
      
720.  
     latency-monitor-threshold 0
721.  
      
722.  
     ############################# Event notification ##############################
723.  
      
724.  
     # Redis can notify Pub/Sub clients about events happening in the key space.
725.  
     # This feature is documented at http://redis.io/topics/keyspace-events
726.  
     #
727.  
     # For instance if keyspace events notification is enabled, and a client
728.  
     # performs a DEL operation on key "foo" stored in the Database 0, two
729.  
     # messages will be published via Pub/Sub:
730.  
     #
731.  
     # PUBLISH __keyspace@0__:foo del
732.  
     # PUBLISH __keyevent@0__:del foo
733.  
     #
734.  
     # It is possible to select the events that Redis will notify among a set
735.  
     # of classes. Every class is identified by a single character:
736.  
     #
737.  
     # K Keyspace events, published with __keyspace@<db>__ prefix.
738.  
     # E Keyevent events, published with __keyevent@<db>__ prefix.
739.  
     # g Generic commands (non-type specific) like DEL, EXPIRE, RENAME, ...
740.  
     # $ String commands
741.  
     # l List commands
742.  
     # s Set commands
743.  
     # h Hash commands
744.  
     # z Sorted set commands
745.  
     # x Expired events (events generated every time a key expires)
746.  
     # e Evicted events (events generated when a key is evicted for maxmemory)
747.  
     # A Alias for g$lshzxe, so that the "AKE" string means all the events.
748.  
     #
749.  
     # The "notify-keyspace-events" takes as argument a string that is composed
750.  
     # by zero or multiple characters. The empty string means that notifications
751.  
     # are disabled at all.
752.  
     #
753.  
     # Example: to enable list and generic events, from the point of view of the
754.  
     # event name, use:
755.  
     #
756.  
     # notify-keyspace-events Elg
757.  
     #
758.  
     # Example 2: to get the stream of the expired keys subscribing to channel
759.  
     # name __keyevent@0__:expired use:
760.  
     #
761.  
     # notify-keyspace-events Ex
762.  
     #
763.  
     # By default all notifications are disabled because most users don't need
764.  
     # this feature and the feature has some overhead. Note that if you don't
765.  
     # specify at least one of K or E, no events will be delivered.
766.  
     # 事件通知，当事件发生时，redis可以通知Pub/Sub客户端
767.  
     # 空串表示禁用事件通知
768.  
     # 注意：K和E至少要指定一个，否则不会有事件通知
769.  
     notify-keyspace-events ""
770.  
      
771.  
     ############################### ADVANCED CONFIG ###############################
772.  
      
773.  
     # Hashes are encoded using a memory efficient data structure when they have a
774.  
     # small number of entries, and the biggest entry does not exceed a given
775.  
     # threshold. These thresholds can be configured using the following directives.
776.  
     # 当hash数目比较少，并且最大元素没有超过给定值时，Hash使用比较有效的内存数据结构来存储。
777.  
     # 即ziplist的结构（压缩的双向链表），参考：http://blog.csdn.net/benbendy1984/article/details/7796956
778.  
     hash-max-ziplist-entries 512
779.  
     hash-max-ziplist-value 64
780.  
      
781.  
     # Similarly to hashes, small lists are also encoded in a special way in order
782.  
     # to save a lot of space. The special representation is only used when
783.  
     # you are under the following limits:
784.  
     # List配置同Hash
785.  
     list-max-ziplist-entries 512
786.  
     list-max-ziplist-value 64
787.  
      
788.  
     # Sets have a special encoding in just one case: when a set is composed
789.  
     # of just strings that happens to be integers in radix 10 in the range
790.  
     # of 64 bit signed integers.
791.  
     # The following configuration setting sets the limit in the size of the
792.  
     # set in order to use this special memory saving encoding.
793.  
     # Sets的元素如果全部是整数（10进制），且为64位有符号整数，则采用特殊的编码方式。
794.  
     # 其元素个数限制配置如下：
795.  
     set-max-intset-entries 512
796.  
      
797.  
     # Similarly to hashes and lists, sorted sets are also specially encoded in
798.  
     # order to save a lot of space. This encoding is only used when the length and
799.  
     # elements of a sorted set are below the following limits:
800.  
     # sorted set 同Hash和List
801.  
     zset-max-ziplist-entries 128
802.  
     zset-max-ziplist-value 64
803.  
      
804.  
     # HyperLogLog sparse representation bytes limit. The limit includes the
805.  
     # 16 bytes header. When an HyperLogLog using the sparse representation crosses
806.  
     # this limit, it is converted into the dense representation.
807.  
     #
808.  
     # A value greater than 16000 is totally useless, since at that point the
809.  
     # dense representation is more memory efficient.
810.  
     #
811.  
     # The suggested value is ~ 3000 in order to have the benefits of
812.  
     # the space efficient encoding without slowing down too much PFADD,
813.  
     # which is O(N) with the sparse encoding. The value can be raised to
814.  
     # ~ 10000 when CPU is not a concern, but space is, and the data set is
815.  
     # composed of many HyperLogLogs with cardinality in the 0 - 15000 range.
816.  
     # 关于HyperLogLog的介绍：http://www.redis.io/topics/data-types-intro#hyperloglogs
817.  
     # HyperLogLog稀疏表示限制设置，如果其值大于16000，则仍然采用稠密表示，因为这时稠密表示更能有效使用内存
818.  
     # 建议值为3000
819.  
     hll-sparse-max-bytes 3000
820.  
      
821.  
     # Active rehashing uses 1 millisecond every 100 milliseconds of CPU time in
822.  
     # order to help rehashing the main Redis hash table (the one mapping top-level
823.  
     # keys to values). The hash table implementation Redis uses (see dict.c)
824.  
     # performs a lazy rehashing: the more operation you run into a hash table
825.  
     # that is rehashing, the more rehashing "steps" are performed, so if the
826.  
     # server is idle the rehashing is never complete and some more memory is used
827.  
     # by the hash table.
828.  
     #
829.  
     # The default is to use this millisecond 10 times every second in order to
830.  
     # active rehashing the main dictionaries, freeing memory when possible.
831.  
     #
832.  
     # If unsure:
833.  
     # use "activerehashing no" if you have hard latency requirements and it is
834.  
     # not a good thing in your environment that Redis can reply form time to time
835.  
     # to queries with 2 milliseconds delay.
836.  
     #
837.  
     # use "activerehashing yes" if you don't have such hard requirements but
838.  
     # want to free memory asap when possible.
839.  
     # 每100毫秒，redis将用1毫秒的时间对Hash表进行重新Hash。
840.  
     # 采用懒惰Hash方式：操作Hash越多，则重新Hash的可能越多，若根本就不操作Hash，则不会重新Hash
841.  
     # 默认每秒10次重新hash主字典，释放可能释放的内存
842.  
     # 重新hash会造成延迟，如果对延迟要求较高，则设为no，禁止重新hash。但可能会浪费很多内存
843.  
     activerehashing yes
844.  
      
845.  
     # The client output buffer limits can be used to force disconnection of clients
846.  
     # that are not reading data from the server fast enough for some reason (a
847.  
     # common reason is that a Pub/Sub client can't consume messages as fast as the
848.  
     # publisher can produce them).
849.  
     #
850.  
     # The limit can be set differently for the three different classes of clients:
851.  
     #
852.  
     # normal -> normal clients including MONITOR clients
853.  
     # slave -> slave clients
854.  
     # pubsub -> clients subscribed to at least one pubsub channel or pattern
855.  
     #
856.  
     # The syntax of every client-output-buffer-limit directive is the following:
857.  
     #
858.  
     # 客户端输出缓冲区限制，当客户端从服务端的读取速度不够快时，则强制断开
859.  
     # 三种不同的客户端类型：normal、salve、pubsub，语法如下：
860.  
     # client-output-buffer-limit <class> <hard limit> <soft limit> <soft seconds>
861.  
     #
862.  
     # A client is immediately disconnected once the hard limit is reached, or if
863.  
     # the soft limit is reached and remains reached for the specified number of
864.  
     # seconds (continuously).
865.  
     # So for instance if the hard limit is 32 megabytes and the soft limit is
866.  
     # 16 megabytes / 10 seconds, the client will get disconnected immediately
867.  
     # if the size of the output buffers reach 32 megabytes, but will also get
868.  
     # disconnected if the client reaches 16 megabytes and continuously overcomes
869.  
     # the limit for 10 seconds.
870.  
     #
871.  
     # By default normal clients are not limited because they don't receive data
872.  
     # without asking (in a push way), but just after a request, so only
873.  
     # asynchronous clients may create a scenario where data is requested faster
874.  
     # than it can read.
875.  
     #
876.  
     # Instead there is a default limit for pubsub and slave clients, since
877.  
     # subscribers and slaves receive data in a push fashion.
878.  
     #
879.  
     # Both the hard or the soft limit can be disabled by setting them to zero.
880.  
     # 当达到硬限制，或者达到软限制且持续了算限制秒数，则立即与客户端断开
881.  
     # 限制设为0表示禁止该功能
882.  
     # 普通用户默认不限制
883.  
     client-output-buffer-limit normal 0 0 0
884.  
     client-output-buffer-limit slave 256mb 64mb 60
885.  
     client-output-buffer-limit pubsub 32mb 8mb 60
886.  
      
887.  
     # Redis calls an internal function to perform many background tasks, like
888.  
     # closing connections of clients in timeout, purging expired keys that are
889.  
     # never requested, and so forth.
890.  
     #
891.  
     # Not all tasks are performed with the same frequency, but Redis checks for
892.  
     # tasks to perform accordingly to the specified "hz" value.
893.  
     #
894.  
     # By default "hz" is set to 10. Raising the value will use more CPU when
895.  
     # Redis is idle, but at the same time will make Redis more responsive when
896.  
     # there are many keys expiring at the same time, and timeouts may be
897.  
     # handled with more precision.
898.  
     #
899.  
     # The range is between 1 and 500, however a value over 100 is usually not
900.  
     # a good idea. Most users should use the default of 10 and raise this up to
901.  
     # 100 only in environments where very low latency is required.
902.  
     # redis调用内部函数执行的后台任务的频率
903.  
     # 后台任务比如：清除过期数据、客户端超时链接等
904.  
     # 默认为10，取值范围1~500,
905.  
     # 对延迟要求很低的可以设置超过100以上
906.  
     hz 10
907.  
      
908.  
     # When a child rewrites the AOF file, if the following option is enabled
909.  
     # the file will be fsync-ed every 32 MB of data generated. This is useful
910.  
     # in order to commit the file to the disk more incrementally and avoid
911.  
     # big latency spikes.
912.  
     # 当修改AOF文件时，该设置为yes，则每生成32MB的数据，就进行同步
913.  
     aof-rewrite-incremental-fsync yes