[teacher Zhao Yuqiang] RDB persistence of redis

Redis There are many different levels of persistence ：

• RDB Persistence can generate a point in time snapshot of a dataset within a specified time interval （point-in-time snapshot）.
• AOF （Append-only file） Persistent records all write commands executed by the server , And when the server starts , Restore the dataset by reexecuting these commands . AOF All commands in the file are as follows Redis The format of the agreement to save , The new command is appended to the end of the file . Redis You can also do it backstage AOF File rewriting （rewrite）, bring AOF The size of the file does not exceed the actual size required to save the dataset state .
• Redis It can also be used at the same time AOF Persistence and RDB Persistence . under these circumstances , When Redis Restart time , It will give priority to AOF File to restore the dataset , because AOF The data set saved by a file is usually larger than RDB The data set saved by the file is more complete .
• You can even turn off persistence , Let the data only exist when the server is running .

One 、RDB The persistence of

working principle ： Take a snapshot of the memory at regular intervals , Write data in memory to a file （rdb file ）. This is a Redis Default persistence method . When redis Generate dump.rdb When you file , The working process is as follows ：

• When reach RDB When generating conditions ,redis The main process fork A subprocess
• fork The subprocess will store the data set in memory dump To the temporary RDB in
• When subprocesses are temporary RDB File write complete ,redis With the new RDB The file replaces the old RDB file

The configuration parameters are as follows ：

RDB Sample test ： have access to redis-benchmark Stress test , Observe RDB Changes in file size .

bin/redis-benchmark -n 100000   Indicates execution 100000 Operations 

Two 、RDB The shortcomings of ：

Between two snapshots , If a power failure occurs , Data will be lost . give an example ： It's generating rdb after , Insert new value . A sudden power failure , Data may be lost .

3、 ... and 、 monitor RDB：

Redis The most direct way to monitor is, of course, to use what the system provides info The order came to do , Just execute the following command , Can get Redis System status report .

bin/redis-cli info | grep rdb_

• rdb_changes_since_last_save Show that last time RDB Save what changes later key frequency
• rdb_bgsave_in_progress Indicates whether or not the current bgsave operation ,1 It means that it is in progress ;0 Indicates that there is no
• rdb_last_save_time Last save RDB The time stamp of the file
• rdb_last_bgsave_time_sec Last saved time
• rdb_last_bgsave_status Last saved state
• rdb_current_bgsave_time_sec It's preserved now RDB Time spent on documents