MultiDB En

Pika has supported multiple DBs since version 3.1.0. For compatibility, some commands and configuration parameters changed. The changes are as follows:

1. info keyspace command:

Retained:

info keyspace [1|0]: Trigger statistics and display, or only display key information for all DBs.

Added:

info keyspace [1|0] db0: Trigger statistics and display, or only display key information for db0.

info keyspace [1|0] db0,db2: Trigger statistics and display, or only display key information for db0 and db2.

Note: DB names must be specified using db[0-7]; multiple DB names are separated by commas.

2. compact command:

Retained:

compact: Compact all DBs.

compact [string/hash/set/zset/list/all]: Compact a specific data structure, or all data structures, for all DBs.

Added:

compact db0 all: Compact all data structures for db0 only.

compact db0,db2 all: Compact all data structures for db0 and db2.

compact db1 string: Compact the string data structure for db1 only.

compact db1,db3 hash: Compact the hash data structure for db1 and db3.

Note: DB names must be specified using db[0-7]; multiple DB names are separated by commas.

3. slaveof command:

Retained:

slaveof 192.168.1.1 6236 [force]: Creates a sync relationship for the pika instance, affecting all DBs. The force parameter performs an instance-level full sync.

Removed:

slaveof 192.168.1.1 6236 1234 111222333: Specifying a write2file file number and write2file file offset when creating a global sync relationship is no longer supported.

4. bgsave command:

Retained:

bgsave: Perform a snapshot backup of all DBs.

Added:

bgsave db0: Back up db0 only.

bgsave db0,db3: Back up db0 and db3 only.

Note: DB names must be specified using db[0-7]; multiple DB names are separated by commas.

5. purgelogsto command:

Retained:

purgelogsto write2file1000: Deletes all write2file files in db0 before write2file1000.

Added:

purgelogsto write2file1000 db1: Deletes all write2file files in db1 before write2file1000. Only one DB can be operated on at a time.

Note: DB names must be specified using db[0-7].

6. flushdb command:

Retained:

flushdb [string/hash/set/zset/list]: Deletes a specific data structure in a DB.

Added:

flushdb: Deletes all data structures in a DB.

Note: Consistent with Redis, please select the correct DB before executing flushdb in Pika to avoid accidentally deleting data.

7. dbslaveof command:

dbslaveof db[0 ~ 7]: Sync a specific DB.

dbslaveof db[0 ~ 7] force: Fully sync a specific DB.

dbslaveof db[0 ~ 7] no one: Stop syncing a specific DB.

dbslaveof db[0 ~ 7] filenum offset: Sync a specific DB starting from a specified offset.

Note: This command requires a master-slave relationship to have been established between two Pika instances before it can control the sync state of a single DB.