Redis persistence

Persistent configuration

# If 900s Inside , There is at least one shareholder key It's been modified , Let's do persistence 
save 900 1
# If in 300s At least 10 individual key It's been modified , It's a persistent operation 
save 300 10
# If in 60s At least 10000 individual key It's been modified , Let's do persistence 
save 60 10000
# Persistence error continue working 
stop-writes-on-bgsave-error yes
# Is it compressed? rdn file 
rdbcompression yes
# preservation rdb When you file , Error checking 
rdbchecksum yes

dir ./

requirepass # Set the password 

maxclients 10000 # Maximum number of clients 
maxmemory #
maxmemory-policy  noevicition 

appendonly no # Not on by default aof 
appendfsync everysec # Once a second sync 

Persistence operation rdb– a key

Write the data set snapshot in memory to disk within the specified time interval , When it recovers, it reads the snapshot file directly into memory

Redis A separate subprocess will be created for persistence , The data is written to a temporary file , The tape persistence process ends , Replace the last persistent file with this temporary file , The whole process , The main process does not do any IO operation , If large-scale recovery is required , And not very sensitive to the integrity of data recovery , that rdb than aof More efficient

#redis Operate on memory , If you cut off the power, you will lose 
#rdb What is kept is dump.rdb  file  

# Trigger mechanism 
#1. Satisfy save Conditions 
#2.fulshall command 
#3. sign out redis

# recovery rdb file 
# Only need to rd Put the file in our redis Boot directory , It checks automatically when it starts up 

# advantage ： Suitable for large-scale data recovery , The requirement of data integrity is not high 
# shortcoming ： It takes a certain interval of time , In case of accidental shutdown , Data will be lost ,

Persistence operation aof

Log every write operation , take redis Record all instructions executed , Reading operation does not record , Only files can be added but not rewritten ,redis At the start , The file will be read to rebuild the data ,redis In case of restart, execute the write instruction from the front to the back according to the contents of the log file to complete the data recovery

# What is kept is  appendonly.aof
# Not on by default 
appendonly yes
# provided that aof There is something wrong with the document. ,redis Can't start , Need repair aof file ,redis-check-aof  File for repair 

# If aof The document passed the general assembly fork A new process 
# advantage ： Every change is synchronized , File integrity will be better , Synchronizing once per second may result in data loss , It is most efficient to never synchronize 
# shortcoming ： Relative to data file ,aof Far greater than rdbm Recovery is also faster than rdb slow ,aof Operating efficiency is also higher than rdb slow , So we redis The default allocation is rdb Persistence